UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies on synaptic potentiation in the hippocampus Goh, Joanne Wan Yoong 1986

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

Item Metadata

Download

Media
831-UBC_1986_A1 G53.pdf [ 8.31MB ]
Metadata
JSON: 831-1.0097124.json
JSON-LD: 831-1.0097124-ld.json
RDF/XML (Pretty): 831-1.0097124-rdf.xml
RDF/JSON: 831-1.0097124-rdf.json
Turtle: 831-1.0097124-turtle.txt
N-Triples: 831-1.0097124-rdf-ntriples.txt
Original Record: 831-1.0097124-source.json
Full Text
831-1.0097124-fulltext.txt
Citation
831-1.0097124.ris

Full Text

STUDIES ON SYNAPTIC POTENTIATION IN THE HIPPOCAMPUS By JOANNE WAN YOONG GOH B. Sc. (Pharm.), The University of Bri t ish Columbia, 1981 M. Sc . , The University of Brit ish Columbia, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Pharmacology & Therapeutics, Faculty of Medicine) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1986 © J o a n n e Wan Yoong Goh, 1986 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of Pharmacology & Therapeutics The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date 12 August, 1986 )E-6 (3/81) - i i -ABSTRACT The present investigation was conducted on transversely sectioned rat hippocampal slices to examine the mechanisms involved in synaptic potentia-t ion. Results indicate that long-term potentiation (LTP) induced by input tetanization requires extracellular C a + + , because during the induction of LTP postsynaptic depolarization must accompany presynaptic ac t ivi ty (LTP could be induced by raised K + [10 to 80 mM] in C a + + - f r e e medium). Since + ++ LTP (induced by raised K ) occurs in the absence of Ca and, therefore, presumably in the near absence of transmitter release, N-methyl-D-aspartate (NMDA) receptor activation is not obligatory. Moreover, NMDA receptors appear not to be involved in the CAj area. A necessity for both pre- and postsynaptic depolarization also accounts for the need for co-stimulation of afferents for LTP induction. Associative potentiation was found not to require tetanic stimulation of the test input; single pulse activation of the test input (at 0.2 Hz) paired with tetanic trains to a conditioning input (presumably to the same postsynaptic neurones) could produce LTP. A short-term potentiation (STP), which resembled post-tetanic potentiation (PTP) in time course, could be induced in an associative fashion by condi-tioning tetanic trains (paired with single test s t imuli ) , that were insuf-f i c ient to produce LTP. In the absence of conditioning s t imuli , interrup-tion of a regular 0.2 Hz test input stimulation for 10 minutes disclosed a subsequent potentiation. This potentiation could be distinguished from associative potentiation in that i t was not associated with a decrease in presynaptic terminal exc i tabi l i ty . A decrease in presynaptic terminal exc i tabi l i ty was characteristic of associative STP and LTP, and followed similar time courses. Since raised K + reversed rather than accentuated the decreased exc i tabi l i ty , i t was concluded that i t is not due to Na -inactivation and may be caused by a hyperpolarization which might also lead to an increase in evoked transmitter release. The hypothesis of Baudry and Lynch (1980a) that LTP is due to an increase in glutamate receptors seems unlikely; there was no increase in Na+-independent glutamate binding sites (determined by the same method as used by Lynch et- a l : [1982]) in associa-tion with LTP induced by a single brief 400 Hz (200 pulses) input tetanus. A decrease in the uptake of glutamate occurs with tetanic stimulation under conditions where there is no LTP (absence of C a + + and raised Mg + + and Mn + +) and, therefore, does not appear to be a mechanism producing LTP. Bhagavatula R. Sastry (Supervisor) - iv -TABLE OF CONTENTS CHAPTER Page ABSTRACT i i TABLE OF CONTENTS iv LIST OF TABLES ix LIST OF FIGURES x ABBREVIATIONS x i i i ACKNOWLEDGEMENTS xiv 1 INTRODUCTION 1 2 LITERATURE SURVEY 5 2.1 Anatomy of the hippocampal formation 5 2.2 Major afferent supply to the hippocampal formation 7 2.2.1 Perforant path 7 2.2.2 Alvear path 7 2.2.3 Commissural input 7 2.2.4 Septal input 8 2.2.5 Inputs from the brainstem 8 2.2.6 Inputs from the diencephalon 9 2.3 Internal c i rcui t ry of the hippocampal formation 9 2.4 Inhibitory influences in the hippocampal formation 11 2.5 Interpretation of waveforms in the hippocampus 14 2.6 Electrophysiology of hippocampal pyramidal neurones 17 2.6.1 Firing patterns and membrane properties 17 - V -CHAPTER Page 2.6.2 Synaptic activation 18 2.6.3 Afterhyperpolarization following synaptic activation 19 2.6.4 Ephaptic interactions 20 2.7 Transmitter candidates in the hippocampal formation 21 2.7.1 Excitatory amino acids 23 2.7.2 Acetylcholine 26 2.7.3 Y-Aminobutyric acid 27 2.7.4 Aromatic amines 27 2.8 Long-term synaptic potentiation in hippocampus 29 2.9 Post-tetanic potentiation 42 3 METHODS 44 3.1 Preparation of slices 44 3.2 Constitution of perfusing media 45 3.3 Recording systems 48 3.4 Stimulating systems 48 3.5 Exci tabi l i ty testing of Schaffer collateral terminal regions 50 3.6 Iontophoretic applications of drugs 52 3.7 Iontophoretic DLH experiments 52 3.8 APV, NMDA, NMLA and NMDLA experiments 55 3.9 Presynaptic interaction experiments 56 3.10 Associative short-term potentiation (STP) and LTP experiments 59 3.11 Experiments involving interruption of input stimulation 64 3.12 Elevated K + experiments 64 - vi -CHAPTER Page 3 3.13 H-Glutamate binding studies 66 3.14 3H-Glutamate uptake studies 69 3.15 Protein determination 70 4 RESULTS 71 4.1 Effects of DLH application at the CA^ cel l body on Schaffer collateral terminal exc i tabi l i ty 71 4.2 Effects of bath application of NMDLA on the CA^ population spike and Schaffer collateral terminal exc i tabi l i ty 72 4.3 Effects of bath application of NMDLA on the CA3 population spike 77 4.4 Effects of bath application of NMDLA in the presence of Ca + + - f ree medium on the CA^ population spike and Schaffer collateral terminal exc i tabi l i ty 79 4.5 Effects of bath application of NMDLA in the presence of APV on the CA-^  population spike and Schaffer collateral terminal exci tabi l i ty 79 4.6 Effects of bath application of APV on tetanus-induced LTP of the CA-^  population spike 84 4.7 Effects of bath application of APV on tetanus-induced LTP of the CA3 population spike 84 4.8 Iontophoretic applications of NMDLA at the CA^ apical dendritic zone 87 4.9 Interactions among presynaptic terminals in the CA-^  region 88 - v i i -CHAPTER Page 4.9.1 Conditioning through a separate electrode 88 4.9.2 Conditioning through the test electrode 91 4.10 Elevated extracellular K + and glutamate on the Schaffer collateral antidromic threshold 94 4.11 Associative short-term potentiation (STP) and LTP 97 4.11.1 Stratum radiatum conditioning 97 4.11.2 Stratum oriens conditioning 102 4.11.3 Alveus conditioning 102 4.12 Effects of a transient interruption of input stimulation on the EPSP and Schaffer collateral terminal exc i tabi l i ty 102 4.13 Effects of elevated extracellular K + on the EPSP and Schaffer collateral terminal exc i tabi l i ty 105 4.14 Ca++-dependence of K +-induced potentiation of the EPSP 111 + 3 4.15 Na -independent H-glutamate binding 114 3 4.16 H-Glutamate accumulation into whole slices 117 5 DISCUSSION 119 5.1 Ca++-dependence of LTP 119 5.2 NMDA receptor involvement in LTP 122 5.3 Interactions among presynaptic terminals in the CA^ region 129 5.4 Associative STP and LTP 133 5.5 Presynaptic involvement in LTP 136 5.6 Potentiation of the EPSP following interruption of input stimulation 143 - v i i i -CHAPTER Page 3 5.7 H-Glutamate binding and uptake studies 144 5.8 Physiological significance of studies on f i e l d potentials 150 6 CONCLUSIONS 151 7 REFERENCES 153 - IX -LIST OF TABLES TABLE Page 1. Effects of DLH applied on CA3 neuronal somata on the threshold for antidromic activation of the neurone at Schaffer collaterals in the C A ^ area. 74 2. Effects of various conditioning stimulations on the threshold for activation of single Schaffer collateral terminals. 90 3. Effect of increasing the number of pulses in each conditioning train on Schaffer collateral terminal exc i tabi l i ty . 92 4. Effect of C a + + - f r e e medium exposure on the conditioned threshold. 93 5. Effects of raising extracellular K + concentration on the Schaffer collateral terminal e x c i t a b i l i t y . 96 6. Effects of bath application of glutamate on Schaffer collateral terminal e x c i t a b i l i t y . 98 7. Post-conditioning potentiation induced by pairing tetanic trains of the alveus with a single stimulation of the test input. 104 8. Na+-independent ^H-glutamate binding following tetanic stimulation of stratum radiatum. 115 9. Na+-independent ^H-glutamate binding following a transient exposure to Cl~-free medium. 116 10. ^H-Glutamate accumulation into s l ices following a high frequency (400 Hz, 200 pulses) tetanic stimulation of stratum radiatum. 118 - X -FIGURE Page 1. Anatomical i l lus t ra t ion of a transverse section of the hippo-campal formation. 6 2. Waveforms recorded in the CA^ region evoked by stimulation of input fibres in stratum radiatum. 16 3. Tetanus-induced long-term potentiation (LTP) of synaptically-driven CA^ responses. 30 4. Diagram of s l ice chamber and perfusion system for maintaining hippocampal s l i ces . 46 5. Positioning of stimulating and recording electrodes for synaptically-evoked population responses. 49 6. Positioning of stimulating and recording electrodes for exc i tabi l i ty testing of the Schaffer collateral terminal regions of single CA3 c e l l s . 51 7. Experimental arrangement for monitoring exc i tabi l i ty changes in the Schaffer collateral terminal regions following a high frequency activation of CA3 ce l l bodies by an iontophoretic application of DL-homocysteate (DLH). 53 8. Experimental arrangement for observing the effects of act ivi ty in other presynaptic fibres on the exc i tabi l i ty of the test Schaffer collateral terminal. 57 9. Experimental arrangement for monitoring exc i tabi l i ty changes in the Schaffer collateral terminal induced by act ivi ty in the same and other nearby f ibres . 58 10. Experimental arrangement for associative induction of short-term (STP) and long-term potentiation (LTP) of the CA^ population EPSP. 61 11. Experimental arrangement for the associative induction of Schaffer collateral terminal exc i tabi l i ty alterations using conditioning stimulations that lead to associative STP and LTP of the population EPSP. 63 12. Effect of DL-homocysteate (DLH, 100 nA, 3 min) on the threshold for antidromic activation of a CA3 neurone, in the presence and absence of extracellular C a + + . 73 - xi -FIGURE Page 13. Effects of N-methyl-DL-aspartate (NMDLA, 100 WM, 2 min) applied to the bath on the CAi population spike and the threshold for antidromic activation of single CA3 ce l ls from the terminal regions of Schaffer col la terals . 75 14. Effects of bath application of NMDLA (100 uM, 2 min) on the mossy fibre-CA3 population spike. 78 15. Blockade of the effects of NMDLA (100 uM, 2 min) on the CAi population spike and the threshold for antidromic activation of a single CA3 ce l l from the terminal regions of Schaffer collaterals by i ts application in the presence of Ca + + - f ree medium (Mn + + 1 mM, M g + + 3 mM). 80 16. Blockade of the effects of NMDLA (100 uM, 2 min) on the CA]_ population spike and the Schaffer collateral antidromic threshold by concurrently administered APV (100 uM, 4 min). 82 17. Effects of APV (100 uM, 4 min) on the induction of LTP in the stratum radiatum-CA^ and mossy fibre-CA3 population spikes. 85 18. NMDLA-induced increase in the threshold for activation of a single Schaffer collateral from i ts terminal regions. 89 19. Failure of picrotoxin (100 uM) to counteract the increase in Schaffer collateral terminal exci tabi l i ty produced by paired pulse stimulation. 95 20. Associative induction of STP, LTP and the reduction in the Schaffer collateral terminal e x c i t a b i l i t y . 99 21. The limits of the temporal relationship between conditioning and test stimuli for the induction of associative potentiation. 103 22. Effects of a 10 minute interruption of input stimulation on the magnitude of the test population EPSP and Schaffer collateral terminal exc i tabi l i ty . 106 23. Effects of elevated extracellular K + (20 mM, 5 min) on the CA^ population EPSP and Schaffer collateral terminal e x c i t a b i l i t y . 108 24. The post-treatment potentiation and depression of the population EPSP induced by application of elevated K + in the absence (1 mM M n + + , 7 mM Mg + + ) and presence (4 mM C a + + , 4 mM Mg + +) of extracellular C a + + , respectively. 112 - x i i -FIGURE Page 25. Schematic i l lus t ra t ion of the hypothetical mechanism responsible for determining the temporal requirements for induction of associative STP. 137 - x i i i -ABBREVIATIONS ACh acetylcholine APV 2-amino-5-phosphonovalerate ChAT choline acetyl transferase DLH DL-homocysteate EDTA ethylenediamine tetraacetic acid EPSP excitatory postsynaptic potential GABA Y-aminobutyric acid GAD glutamic acid decarboxylase 5-HT 5-hydroxytryptami ne IPSP inhibitory postsynaptic potential LTP long-term potentiation NA nor adrenaline NMDA N-methyl-D-aspartate NMDLA N-methyl-DL-aspartate NMLA N-methyl-L-aspartate PTP post-tetanic potentiation STP short-term potentiation - xiv -ACKNOWLEDGEMENTS I thank my supervisor, Dr. B. R. Sastry for his guidance throughout this study. I am grateful to Mr. A. Auyeung, Mr. P. May and Ms. M. Ho-Asjoe for their help during the course of this investigation and to Dr. David Quastel for suggesting the use of elevated K + for depolarizing neurones in the si ice . Financial support from the Medical Research Council of Canada, the University of Bri t ish Columbia Graduate Summer Scholarship and the H. R. MacMillan Family Fellowship is gratefully acknowledged. - 1 -1 INTRODUCTION It is well known that long-term potentiation (LTP) of the stratum radiatum stimulation-evoked CA-^  population spike can be produced following a high frequency tetanic stimulation to the input fibres (Schwartzkroin and Wester, 1975). The mechanisms responsible for LTP, however are not entirely clear and there is debate as to whether the phenomenon is due to a pre- or postsynaptic alteration. The present investigation was generally aimed at determining mechanism(s) of LTP and, in view of evidence that LTP is "asso-c ia t ive" , i . e . , cannot be induced by tetanic stimulation of a "weak" input alone (McNaughton, 1982; McNaughton e t - a l : , 1978; Robinson and Racine, 1982) was spec i f i ca l ly aimed at answering the following questions: 1) is C a + + directly required for LTP induction?; 2) is tetanic stimulation of a test input real ly necessary for the induction of LTP?; 3) is LTP sustained by a presynaptic change?; 4) what is the mechanism of the reduced exc i tabi l i ty in presynaptic terminal regions that accompanies LTP?; 5) does an increase in glutamate receptors correlate with LTP?; 6) can a decrease in glutamate up-take account for LTP?; 7) are NMDA receptors involved in production of LTP? It is generally believed that the induction of LTP requires the pre-sence of extracellular C a + + (Dunwiddie and Lynch, 1979; Wigstrom et- a l , 1979); tetanic stimulation of an input during perfusion of hippocampal slices with C a + + - f r e e medium fai led to produce LTP. However, this treat-ment also succeeded in blocking synaptic transmission. Therefore, i t is unclear whether LTP induction was counteracted due to a removal of C a + + or due to an abolition of synaptic transmission. Studies were, therefore, con-- 2 -++ ducted to examine whether LTP could be induced in Ca -free medium i f both the presynaptic terminals as well as postsynaptic neurones were depolarized using elevated extracellular K + concentrations. That the induction of LTP requires the co-activation of several input fibres is shown by the finding that "associative" LTP can be induced in response to a "weak" test input (which does not exhibit LTP when tetanized), by concurrent tetanization of a presumably separate "strong" input to the recorded neurones (McNaughton, 1982; McNaughton et - a l : , 1978; Robinson and Racine, 1982). This observation led the above authors to think that the common link between the two separate inputs is the postsynaptic neurone and, therefore, that LTP is postsynaptic. However, there is no reason to assume that presynaptic fibres do not communicate with each other. Studies were, therefore, conducted to examine this poss ibi l i ty and i t was indeed found that presynaptic fibres in the CA-^  region interact with each other, possi-bly via the postsynaptic neurone. The results prompted further studies to determine i f the presynaptic exc i tabi l i ty change in Schaffer collateral terminals associated with LTP was induced by associative interactions. Furthermore, experiments were also conducted to examine whether a potentia-tion of the test EPSP could be e l i c i t e d by associative conditioning without delivering a tetanic stimulation to the test input. The temporal cons-traints governing the position of the test and conditioning stimuli for induction of associative potentiation were also investigated. Since i t has been established that a high frequency activation of an input results in PTP and LTP, i t was of interest to examine the effects of non-activation of the input. These experiments were conducted by interrupt-- 3 -ing the control rate of input stimulation of 0.2 Hz with a "rest" period of 10 minutes. The EPSP as well as the Schaffer collateral terminal excitabi-l i t y were monitored before and after the quiescent period. A previous report (Sastry, 1982) suggested that a decrease in excita-b i l i t y (reflecting a hyperpolarization) of presynaptic terminals could be a mechanism responsible for LTP. An attempt was, therefore, made to correlate the time course and magnitude of the decrease in presynaptic terminal excit-a b i l i t y with the potentiation of the EPSP. The nature of the decrease in Schaffer collateral terminal exc i tabi l i ty is not known and several possibi-l i t i e s exist . Experiments were conducted to ascertain i f the exci tabi l i ty + change could be due to Na -inactivation or hyperpolarization. N-Methyl-D-aspartate (NMDA) receptors are thought to play a role in the induction of LTP (Col 1ingridge et - a l . , 1983b; Harris e t - a l ; , 1984). The effects of exogenously applied N-methyl-DL-aspartate (NMDLA) were examined on the CA^ population spike and Schaffer collateral terminal excitabi-l i t y . Studies were also carried out to ascertain i f the terminal excitabi-l i t y change could be induced by localized application (iontophoretically) of NMDLA in the CA-^  apical dendritic zone where these fibres terminate. The reason for localizing the application was to exclude the poss ibi l i ty that an increase in the f i r i n g rate of CA3 neurones during bath application of the amino acid was the reason for producing the presynaptic change. It has been reported that 2-amino-5-phosphonovalerate (APV), a "selective NMDA antag-onist" counteracts the induction of LTP (Coll ingridge et- - a l ; , 1983b; Harris et - al .- , 1984). To further investigate a presynaptic role in NMDLA-induced potentiation of the population spike, the a b i l i t y of APV to antagonize the - 4 -effects of NMDLA on the CA^ population spike and Schaffer collateral terminal exci tabi l i ty was examined. Al l previous studies concerning the NMDA hypothesis were conducted in the CA-^  region of the hippocampus (Col 1 ingridge et - -al-;, 1983b; Harris et- al . - , 1984). It was, therefore, decided to determine i f tetanus-induced LTP of the mossy fibre-CA^ system involved NMDA receptors by applying APV during the tetanic stimulation. The effects of exogenously applied NMDLA were also examined on the CA^ popula-tion spike. Baudry and Lynch (1980a) hypothesized that LTP could be accounted for by a postsynaptic mechanism whereby LTP of the synaptically-evoked popula-++ tion response is a consequence of a Ca -triggered increase in the number of subsynaptic neurotransmitter receptors. The suspected transmitter re-leased by the Schaffer collaterals (which comprise part of the stratum radiatum) is glutamate (Storm-Mathisen, 1977a). It was shown by Wieraszko (1983) that stimulus-evoked uptake of D-aspartate, a marker for glutamater-gic terminals, is decreased during LTP. This raises the interesting possi-b i l i t y that a decrease in transmitter uptake could result in a potentiation of the synaptically-evoked response due to a greater avai labi l i ty of the amino acid to subsynaptic receptors. Studies were conducted to 1) examine + i f an increase in Na -independent glutamate binding sites (presumed to be receptors by Baudry and Lynch [1980a]) is necessary for the visualization of LTP; 2) investigate the poss ibi l i ty that a decrease in glutamate uptake can be a mechanism for the maintenance of LTP. - 5 -2 LITERATURE SURVEY 2.1 Anatomy of the hippocampal formation The hippocampal formation is a structure that comprises part of the limbic system. It is made up of the hippocampus proper (also called Amnion's horn), the dentate gyrus and most of the subiculum. The highly organized lamellar organization of the hippocampal formation has made this cortical structure a desirable area to perform neurophysiological studies upon. The cel l body layer of the hippocampus, i f divided into the C A ^ a ^ CA^ ^ 3 ( a b c) a n c ' ^ 4 r e 9 i ° n s a n c ' t n e dentate gyrus, is made up of an inferior and superior blade containing its cel l body layers (Lorente de No, 1934) (Figure 1). These two structures, which are t ight ly interlocked have the CA^ area of the hippocampus penetrating the hilus (which is the t r i -angular shaped region between the inferior and superior blades) of the den-tate gyrus. The predominant cell type in the hippocampus is the pyramidal cel l (Blackstad, 1956; Golgi , 1886; Lorente de No, 1934), whereas granule cel ls are found in the dentate gyrus (Golgi, 1886; Lorente de No, 1934). A limited number of inhibitory interneurones (basket cells) can be found interspersed in both the pyramidal as well as granule ce l l layers (Ramon y Cajal , 1968). In a dorsal to ventral order, the layers of the hippocampus-dentate gyrus system are stratum oriens, stratum pyramidale, stratum radia-tum, stratum lacunosum, stratum moleculare, stratum granulosum and stratum polymorphe (Ramon y Cajal , 1911) (Figure 1). - 6 -Figure 1 Anatomical i l l u s t r a t i o n of a transverse section of the hippocampal formation. The hippocampal formation consists of the hippocampus (also cal led Amnion's horn or Cornu Ammonis) which is a C-shaped structure that folds t ight ly into another C-shaped structure, the dentate gyrus (also cal led Fascia-dentata) . The hippocampus is divided into four subfields: C A i , CA 2 , CA3 and C A 4 . Enlargement on r ight shows layered structure of the hippocampal formation. PROS - prosubiculum HF - hippocampal f issure Sch - Schaffer co l la tera l Comm - commissural input MF - mossy f ib re PP - perforant path input B - basket ce l l A represent pyramidal ce l ls of the hippocampus • represent granule ce l ls of the dentate gyrus - 7 -2.2 Major afferent-supply to the hippocampal-formation 2.2.1 Perforant- path; Anatomical evidence indicates that the major input to the dentate granule cel ls arises from the entorhinal cortex and is known as the perforant path (Ramon y Cajal , 1911; Lorente de No, 1934). Electrophysiological evidence provided by L$mo (1971) shows large f i e l d potentials recorded in the dentate granular cell layer by stimulation in the entorhinal area. The extracellularly recorded excitatory postsynaptic potential (EPSP) is maximal in size and is characterized by a negative wave in stratum moleculare (Andersen et - al . - , 1966b; L#mo, 1971), an observation which is consistent with the anatomical localization of perforant path synapses (Blackstad, 1958; Nafstad, 1967; Hjorth-Simonsen and Jeune, 1972). Although the perforant path innervates primarily the dentate gyrus, i t has also been shown with histological techniques that some fibres from this input terminate in the prosubiculum, CAp CA ,^ and CAg areas (Lorente de No, 1934). 2.2.2 Alvear path; Another input that arises from the entorhinal area to innervate the hippocampus is the alvear path which courses along the dor-sal surface of the hippocampus to terminate in the prosubiculum and CA-, xa regions (Lorente de No, 1934). 2.2.3 Commissural -input. The CAg and CA 4 pyramidal cel ls give rise to axons which form the commissural fibres that course through the ven-tral commissure (with a few that run through the dorsal commissure) to terminate in the contralateral hippocampus (Andersen, 1960a, 1960b; Blackstad, 1956; Deadwyler et a l . , 1975; Gottlieb and Cowan, 1973). Anato-mical studies have demonstrated that the commissural input, which enters the - 8 -hippocampus via the fimbria, innervates a l l f ie lds of the hippocampus and the dentate gyrus (Gottlieb and Cowan, 1973; Hjorth-Simonsen and Laurberg, 1977). In support of these histological findings, evidence from electrophy-siological analysis has shown that stimulation of commissural fibres pro-duces synaptic excitation of dentate granule cel ls (Cragg and Hamlyn, 1957) and the apical as well as basilar dendrites of CA^ pyramidal neurones (Andersen, 1960a, 1960b). Degeneration studies conducted by Blackstad (1958) i l lus t ra te a greater number of commissural terminations in the apical dendrites as compared to the basal dendrites of CA^ neurones, an observa-tion that is supported by electrophysiological evidence (Cragg and Hamlyn, 1957). 2.2.4 Septal- input: The septo-hippocampal fibres course through the dorsal as well as the body of the fornix to enter the alveus. This input terminates primarily in the hilus of the dentate gyrus and stratum oriens and stratum radiatum of the Zk^ and CA^ regions. A sparse population of septo-hippocampal fibres enters the molecular layer of the dentate gyrus, perhaps with the perforant path (Mellgren and Srebro, 1973). The hippocam-pal formation receives afferents that originate from the medial as well as lateral septal areas, the former input terminating in the dorsal hippocampus while fibres from the latter project to the ventral hippocampus (Siegel and Tassoni, 1971). It is believed that septo-hippocampal fibres are important in normal hippocampal functions, especially in the generation of slow wave or theta electr ical act ivi ty (Green and Arduini, 1954). 2.2.5 Inputs-from- the- brainstem.- Efferents from the brainstem nuclei such as the locus coeruleus, raphe nuclei and ventral tegmental nuclei form - 9 -monosynaptic connections with different regions of the hippocampal forma-t ion . It appears that the locus coeruleus projects to the subiculum, the CA-^  area (Pasquier and Reinoso-Suarez, 1978) and the dentate gyrus (Blackstad et a l ; , 1967; Lindvall and Bjbrklund, 1974; Loy et- a l ; , 1980; Storm-Mathisen, 1978). Fibres from the raphe nuclei innervate the dentate gyrus, the subiculum and parts of the hippocampus (Azmitia and Segal, 1978; Azmitia and Marovitz, 1980; Fuxe and Jonsson, 1974; Kbhler and Steinbusch, 1982; Segal, 1980; Steinbusch, 1981) while a projection from the ventral tegmental area, which is thought to be dopaminergic, terminates primarily in the dentate area (Simon et a l ; , 1979). Theta act ivi ty in the hippocampus is thought to be influenced by inputs from the locus coeruleus and raphe nuclei to the septal area and hippocampus (Gray, 1977). 2.2.6 Inputs - from- the- diencephalon. A direct input from the thalamus innervates the subiculum and CA^ regions of the hippocampus. Fibres emerge from anterior, posterior and midline nuclei of the thalamus to termi-nate in the subiculum and CA^ area (Schwerdtfeger, 1984). In humans, another input from the lateral geniculate nucleus also projects direct ly to the hippocampus (Babb e t - a l ; , 1980). The hypothalamus has two major inputs to the hippocampal formation, one projecting to the subiculum and the other to the dentate gyrus (Schwerdtfeger, 1984). 2.3 Internal c i rcui t ry of the-hippocampal-formation Dentate granule cel ls give rise to short axons called the mossy fibres which travel in the hi lar region to terminate on the dendrites of pyramidal cel ls in the C A ? , CA-, and CA* regions of the hippocampus (Blackstad et_ a l . , 1970; Lorente de No, 1934; Ram6n y Cajal , 1968) (Figure 1). The mossy f ibre bundle that arises from the inferior blade of dentate granule cel ls terminate a short distance away at the CAg c and CA^ regions of Lorente de No (1934). The mossy fibres that arise from the superior blade, however, traverse through CA^ and end in the CA^ region (Blackstad et a l : , 1970; Lorente de No, 1934). According to Hamlyn (1961), the whole f i e l d of CA3 is innervated by the mossy fibres which course through and activate the apical dendrites of the pyramidal neurones. These synapses are en -passant so that a single presynaptic volley wil l normally produce excitation of a number of CA^ neurones (Blackstad and Kjaerheim, 1961). It has been shown both histological ly (Hjorth-Simonsen and Jeune, 1972) and physiologically (Ltfmo, 1971) that mossy fibres exhibit a highly organized parallel lamellar arrangement. Pyramidal cel ls of the CA^ region give off axons which leave the hippocampus via the fimbria to cross over to the contralateral hippocampus as commissural fibres (described e a r l i e r ) . These axons branch off before leaving the hippocampus to form the Schaffer collaterals (Schaffer, 1892; Ramon y Cajal , 1911) and cross over the pyramidal ce l l body layer to termi-nate, in the stratum lacunosum of the CA^ and CA2 areas (Lorente de No, 1934) (Figure 1). Electrophysiological evidence supplied by Andersen (1960a) has confirmed the anatomical data above. Powerful synaptic excita-tion is seen in CA-^  neurones upon stimulation of CA3 ce l ls or the Schaf-fer collateral input and this activation is represented by a large negative wave which is maximal when recorded in the apical dendrites of CA^ neu-rones . - 11 -The axons of CA^ cel ls leave the ce l l body from the basilar side and head dorsally for a short distance before entering the alveus to course along the dorsal surface of the hippocampus and ultimately exit through the fimbria (Lorente de No, 1934). Some of these axons divide to form a recur-rent collateral which also travels in the alveus but in the opposite direc-tion towards the prosubiculum (Lorente de N6, 1934; Hjorth-Simonsen, 1973; Chronister and Zornetzer, 1973) (Figure 1). Sequential activation of neurones in the hippocampus via a trisynaptic loop consisting of 1) perforant path-dentate granule c e l l s ; 2) mossy f i b r e s -CA3 pyramids; 3) Schaffer col laterals-CA^ pyramids has been described (Andersen et a l . , 1971) (Figure 1). It was found that the perforant path, mossy f ibres , Schaffer collaterals and alvear fibres which constitute the outputs from the entorhinal-dentate-CA^-CA^ areas, respectively, are a l l arranged in parallel bands that are oblique to the longitudinal axis of the hippocampus (Andersen et a l . , 1971; L#mo, 1971). Therefore, one can expect a transversely sectioned s l i ce of the hippocampal formation to contain an intact trisynaptic loop. This, in fact , has been established electrophysio-logical ly by Skrede and Westgaard (1971) who demonstrated that in transverse slices of 3.00 to400 urn thick, a l l the synaptic inputs described were intact and functional. 2.4 Inhibi tory inf1uences in the -hippocampal formation The presence of inhibitory interneurones called basket cel ls in the hippocampus and the dentate gyrus has been well established (Andersen et  a l . , 1964a, 1964b; Ramon y Cajal , 1911; Spencer and Kandel, 1961c). These cel ls are thought to mediate recurrent inhibition and, therefore, 1) the - 12 -inhibitory postsynaptic potential (IPSP) e l i c i ted following afferent excita-tion has a longer latency than the excitatory postsynaptic potential (EPSP) and 2) antidromic stimulation of CAg axons produced either large IPSPs or antidromic invasion of CAg neurones followed by an IPSP (Andersen et- - a l . , 1964a, 1964b; Spencer and Kandel, 1961). Andersen et a l . (1964a) observed that the IPSP was maximum at the cel l body layer and was characterized by an extracellular p o s i t i v i t y . This suggests that the inhibition is mediated by a hyperpolarization that is generated at or close to the cel l body. Since the latency to onset of the IPSP is typical ly longer than that of the EPSP (when synaptic inputs are stimulated) and also than that of antidromic inva-sion (when axons of pyramidal cel ls are stimulated antidromically), i t has been postulated that the inhibition is produced by an interneurone which terminates on the cell bodies of pyramidal cel ls (Andersen et- - a l ; , 1964a, 1964b). Basket cel ls exhibit repetitive f i r i n g . The latency to onset of the IPSP decreased and the frequency as well as number of discharges of the interneurones (observed as a high frequency ripple on the r is ing phase of the response) increased with increasing afferent stimulation intensity (Andersen et a l . , 1964a). It was suggested that the inhibitory transmitter of the basket cel ls is y-aminobutyric acid (GABA) (Andersen, 1975; Andersen et • -al.-, 1964a; Curtis et- a l ; , 1970). Using specific antisera to glutamic acid decarboxy-lase (GAD), the enzyme for catalyzing the formation of GABA, i t was shown that dense GAD-positive staining was present at the ce l l bodies of pyramidal and granule c e l l s , corresponding to the location of basket cel l terminations (Ribak et a l . , 1978). The IPSP which is reduced by GABA antagonists (Alger - 13 -and Nicol 1 , 1982; Nicoll and Alger, 1981) is mediated by an increase in C l ~ conductance of the cel l (Kandel e t - a l ; , 1961). In addition to the classical recurrent inhibition described above, there appears to be feed-forward inhibition in the hippocampus. Andersen (1975) observed that some basket cel ls discharge with very weak input v o l -leys. In fact , interneurones can be activated with s ignif icantly lower stimulus intensities than postsynaptic pyramidal ce l ls and some interneu-rones may discharge in response to stimulations that are subthreshold to evoke a synaptic response in the pyramidal cel ls (Buzsaki and Eidelberg, 1981; Buzsaki e t - a l ; , 1983; Fox and Ranck, 1981; Knowles and Schwartzkroin, 1981). At higher stimulus strengths, i t has been shown that interneuronal discharge may precede the onset of the synaptically-induced population spike (Buzsaki and Eidelberg, 1982). It is not known whether feed-forward i n h i b i -t ion, l ike recurrent inhibi t ion , is mediated by basket c e l l s . Knowles and Schwartzkroin (1981) provide evidence for interneurones which are morpholo-gica l ly dissimilar to the classical basket cel ls to mediate both recurrent and feed-forward inhibition in the hippocampus. In addition to the recurrent and feed-forward IPSPs discussed, a long-lasting late hyperpolarizing potential is observed in pyramidal ce l ls f o l -lowing orthodromic but not antidromic stimulation, suggesting a feed-forward mode of activation (Newberry and N i c o l l , 1984a). GABA application onto pyramidal cel l dendrites e l i c i t s an i n i t i a l depolarization followed by a hyperpolarization of the cel l (Alger and N i c o l l , 1979). The i n i t i a l depol-arizing phase has been attributed to GABA acting on the dendrites of the pyramidal neurones because i t seems to involve a C l ~ conductance and is - 14 -blocked by GABA antagonists (Alger and N i c o l l , 1979). The late hyperpol-arizing potential is believed to be due to an increase in K + conductance at the dendritic zone that is not dependent on intracel lular C a + + and is not blocked by the classical GABA antagonists bicuculline and picrotoxin (Newberry and N i c o l l , 1984a). A novel bicucul1ine-insensitive GABA binding site has been described (Bowery et- a l ; , 1979, 1981). To distinguish the classical GABA receptor from this new receptor, the former has been termed the GABAA receptor and the latter the GABAg receptor. The GABAg receptor is thought to be present presynaptically (Bowery et- a l ; , 1980, 1981; Olpe et - al-.-, 1982) and a selective agonist at this site is baclofen (Bowery et- a l . , 1979, 1980, 1981). The late hyperpolarizing potential of Newberry and Nicoll (1984a) was suggested to be due to activation of dendri-t i c GABAg receptors because an application of baclofen generates a similar type of potential (Newberry and N i c o l l , 1984b). Baclofen has been shown to inhibit excitatory transmission in the hippocampus, an effect that has been attributed to a blockade of excitatory transmitter release (Olpe et- a l ; , 1982). Since axo-axonal synapses in the hippocampus have never been demon-strated, the relevance of the above results to physiological GABAergic pre-synaptic inhibition is not clear at present. 2.5 Interpretation-of waveforms - in the hippocampus Since a l l recordings done in the present study were extracellular, I shall l imit myself to a discussion of these responses. A unique and very useful feature of the CA-^  region of the hippocampus is that two dis t inct ly separate sets of input fibres can activate the same population of pyramidal c e l l s . Stimulation of stratum radiatum selectively activates the apical - 15 -dendrites and stimulation of stratum oriens results in excitation of the basilar dendrites (Figure 1). At a particular recording s i te , an extracel-lular negativity usually represents either a local depolarization or a dis -tant hyperpolarization and, conversely, a positive waveform indicates a local hyperpolarization or a distant depolarization, depending on the source-sink relationship of current flow. Upon stimulation of stratum radiatum (which includes the Schaffer col la terals ) , an EPSP is generated at the api -cal dendrites of CA^ neurones. A recording electrode placed at the source of the EPSP (100-200 ym from cell body layer) wi l l recognize a negatively directed waveform (Figure 2C_) due to a mixture of f ie lds caused by depol-arization of the apical dendrites and hyperpol arization of the soma by the IPSP. On the other hand, i f the recording site is the CA^ pyramidal cel l body layer, the EPSP wi l l reverse direction and manifest i t s e l f as a posi -tive response (Figure 2J3) because this region is the sink for depolarizing current generated at the dendrites and the source for a hyperpolarizing IPSP. As we move the recording electrode towards the basal dendrites, we find a positively directed EPSP with a smaller magnitude than that recorded in the cel l body layer (Figure 2A). When the stimulation strength is i n -creased to evoke a synchronous population spike (whose size is proportional to the number of postsynaptic neurones that are activated to discharge action potentials) , i ts direction is negative in the ce l l body layer (Figure 2E) and basal dendrites (Figure 20) but is positive in the apical dendrites (Figure 2_F). Note that the latency to onset of the spike is shortest with the cel l body recording (Figure 2E_), suggesting that the site for action potential generation in CA^ neurones is at or close to the soma. Similar - 16 -stratum ' radiatum 2mV 10 ms Figure 2 Waveforms recorded in the CA^ region evoked by stimulation of input f ibres in stratum radiatum. Records in A, B_ and C are extracellular population EPSPs recorded in the basilar dendrites, ceTl body layer and apical den-dr i tes , respectively. Upon increasing the stimulus intensity, population spikes can be e l ic i ted (D_, E and F) . Note that the population spike is posit ive at the apical dendrTtes ancJ negative at the ce l l body and basilar dendrites. Each response is an average of 4 consecutive sweeps. Records A and D, B and E and C and F_ were obtained from 3 different experiments. Negativity is downwards. - 17 -observations were documented by Andersen (1960a). 2.6 Electrophysiology of hippocampal pyramidal neurones 2.6.1 Fir ing patterns -and membrane- properties. Characteristics of neurones in the hippocampus have been well documented in- v-ivo (Kandel and Spencer, 1961; Kandel et a l . , 1961; Spencer and Kandel, 1961a, 1961b). The pyramidal neurone has a soma, two distinct dendritic systems (apical and basilar) and an axon that courses in the alveus. Spike generation in these cel ls following various modes of activation is similar to that seen in moto-neurones where the r is ing phase of the action potential shows a small i n -f l e c t i o n . Excitatory and inhibitory postsynaptic potentials were graded and i t was concluded that depolarization is related to excitation and hyperpol-arization is related to inhibi t ion . In addition, stimulation of axons ant i -dromically produces a marked inhibition of the neurone through recurrent collaterals (Kandel et a l . , 1961). Spontaneous act ivi ty in pyramidal cel ls consisted of either single spikes or bursts of action potentials. Single spikes or bursts were associated with a post-spike potential which lasted roughly 20 ms and was depolarizing. It appears that these after-potentials were additive and the magnitude corresponded to the number of spikes in a burst. However, the time constant of decay was independent of the duration of bursts. A burst is s e l f - l i m i t i n g and the membrane repolarizes even in the presence of a long depolarizing pulse. A short depolarizing current pulse, however, can trigger a burst which f a i l s to terminate immediately upon cessation of the stimulation, perhaps because the summation of the depolarizing after potentials can sustain repetitive f i r i n g (Kandel and Spencer, 1961). The depolarizing after potential could be mediated by a slow voltage-dependent Ca current (Johnston et- a l ; , 1980). The f i r i n g level for an action potential was constant for different methods of excita-t ion. Membrane properties determined for these neurones included time cons-tant (9.9 ms), rheobase for intracellular stimulation (1-5 X 10"^ A) and total neurone resistance (around 13 Mn) (Spencer and Kandel, 1961a). In some neurones, there appeared to be what was termed a fast prepotential which was a small potential that preceded the somatic spike. It was esta-blished that this fast prepotential was spike act ivi ty generated within the neurone from which i t was recorded. Since the potential was so small, i t was concluded that i t was generated at some site distant from the soma. Antidromic activation of the cel l fa i led to evoke any fast prepotentials so i t is unlikely that they arose from axons. It was suggested that the pre-potentials represented dendritic spikes and may have some role in impulse generation (Spencer and Kandel, 1961b). With the advent of the hippocampal s l ice preparation, Schwartzkroin (1975, 1977) has shown that v i r tual ly a l l the properties of hippocampal neu-rones in -vitro are similar to those found in -vivo, suggesting that cel ls in the s l ice do not become abnormal following the s l i c i n g procedure. There-fore, i t is reasonable to believe that data obtained from a s l ice prepara-tion is a true reflection of the hippocampal system in intact animals. 2.6.2 Synaptic activation. The excitatory synaptic contacts on pyra-midal ce l ls are located almost exclusively on dendritic spines (Andersen et  a l ; , 1966b). On the other hand, inhibitory interneurones terminate predomi-nantly at or near the somatic area. An estimate of the size of the EPSP produced by activation of a single presynaptic f ibre is 0.1 mV and the d i s -- 19 -charge threshold for the pyramidal cel l is about 6 mV (Andersen and Langmoen, 1981). The usual resting membrane potential of these pyramidal cel ls was reported to be around -60 mV (Andersen et - a U , 1980a). When two separate inputs to the same pyramidal cel l dendrite were stimulated simul-taneously, the EPSPs summated in a linear fashion. Similarly , when IPSPs were evoked, they also added algebraically to the summated EPSPs (Andersen and Langmoen, 1981). These observations suggest that roughly 60 synapses have to be activated simultaneously to synaptically drive a pyramidal neu-rone to discharge. Since these cel ls are thought to have 10,000 synaptic contacts (Hamlyn, 1963), i t appears that activation of less than one percent of these synapses is necessary to evoke a spike. 2.6.3 Afterhyperpolarization following synaptic activation and repeti-^  t-i-ve f i r i n g . Burst f i r i n g in hippocampal neurones induced in a variety of ways is accompanied by a post-activation hyperpolarization of the c e l l which can last as long as several minutes (Alger and N i c o l l , 1980; Gustafsson and Wigstrb'm, 1981; Hablitz, 1981; Hotson and Prince, 1980; Nicoll and Alger, 1981; Schwartzkroin and Stafstrom, 1980; Segal, 1981). The nature of the hyperpol arization has been claimed to be due to various factors. Segal (1981) feels that a Na+-pump can be responsible for this process following repetitive activation of neurones by the excitatory agent glutamate. Others believe that a prolonged Ca + + -activated K + conductance can play a role (Alger and N i c o l l , 1980; Hotson and Prince, 1980; Nicoll and Alger, 1981; Schwartzkroin and Stafstrom, 1980). Recently, i t was demonstrated that a prolonged synaptically driven IPSP could in part account for the hyperpol-arization following synaptic activation of pyramidal cel ls (Newberry and - 20 -N i c o l l , 1984a, 1984b). This late feed-forward IPSP is thought to be media-ted by GABA acting on GABAg receptors leading to an increase in a K + conductance whose activation is independent of Ca . This potential is not sensitive to classical GABA antagonists l ike bicuculline and picrotoxin and may represent an EGTA-resistant burst afterhyperpolarization (Alger and N i c o l l , 1980; Hablitz, 1981; Schwartzkroin and Stafstrom, 1980). Since i t has been reported that the afterhyperpolarization is made up of more than one component (Gustafsson and Wigstro'm, 1981; Schwartzkroin and Stafstrom, 1980), i t is possible that two or a l l three of the processes that have been discussed are involved. Bear in mind, however, that the late IPSP is only seen following synaptic but not antidromic or direct activation of pyramidal eel Is. 2.6.4 Ephaptic interactions; A synchronous extracellular f i e l d poten-t i a l in the CA^ area can influence the membrane potentials of surrounding pyramidal cel ls that are not included in the direct ly stimulated popula-t ion. These so called "ephaptic" interactions that are generated by tran-sient extracellular electr ical f ie lds have been implicated in synchronizing and f a c i l i t a t i n g neural discharges in the hippocampus (Richardson et - -aT., 1984; Taylor and Dudek, 1982; Turner et a l ; , 1984). It has been postulated that ephaptic coupling provides a positive feedback for excitation of neu-rones so that an increase in the f i e l d potential would mean an enhancement in ephaptic interactions and, hence an amplification of the population res-ponse further (Turner et a l . , 1984). In addition to ephaptic interactions, there are other ways in which hippocampal cel ls can "communicate". It has been demonstrated physiologi-- 21 -cal ly and anatomically that electrotonic coupling exists between pyramidal cel ls and granule c e l l s , probably through gap junctions (MacVicar and Dudek, 1981, 1982). According to these authors, such a coupling could result in synchronization of rhythmic act ivi ty as well as epileptic discharges in a population of c e l l s . Changes in the extracellular concentration of ions could result in exci tabi l i ty changes in neurones. Depolarization of the cel lular membrane of quiescent neurones could be achieved during neuronal act ivi ty of other cel ls in the same area by an elevation of extracellular + ++ K or a f a l l in extracellular Ca concentrations. It has been shown both in vivo and in - vitro using ion-sensitive electrodes that repetitive stimulation of inputs in the hippocampus leads to a marked elevation in extracellular K + and a depletion of C a + + (Benninger et a l : , 1980; Fr i tz and Gardner-Medwin, 1976; Krnjevic et a l ; , 1980, 1982). Although studies to date regarding ephaptic interactions and electrotonic coupling were done on neuronal somata, there is no reason to exclude the poss ibi l i ty that such processes also play a role presynaptically to modulate transmitter release. Weight and Erulkar (1976) reported that repetitive postsynaptic action potentials can alter transmitter release presynaptically, possibly through an increase in extracellular K + concentrations. 2.7 Transmitter candidates in the hippocampal formation Before a compound is accepted to be a chemical neurotransmitter, i t is generally agreed upon that certain c r i t e r i a must be met (Werman, 1966). 1) The agent, when applied onto postsynaptic neurones, produces an action that is identical to that of synaptically released transmitter. 2) During stimu-lation of the presynaptic f ibres , the substance is released from the termi-nals. 3) The substance is released from axon terminals in a Ca -depen-dent manner. 4) Drugs which interact with the endogenous transmitter should also interact with the exogenously applied suspected transmitter in an iden-t i ca l manner. 5) The substance must be present in the releasing neurones. 6) The releasing neurone must have enzymes for the synthesis and release of the suspected transmitter. 7) Various precursors and intermediate compounds for the synthesis of the substance should be present. 8) Inactivating mechanisms such as enzymatic breakdown processes or uptake mechanisms could be present. These c r i t e r i a may be useful to assess the s u i t a b i l i t y of any substance as a transmitter candidate at a given synapse. However, i t has to be borne in mind that these are guidelines and not necessarily r i g i d rules because there may be exceptions. For instance, c r i t e r i a 2) and 4) may not be satisfied i f available detection methods are not suff ic ient ly sensitive to monitor low concentrations of transmitter or i f extensive inactivation occurs following release. The detection of acetylcholine in brain and s p i -nal cord superfusates has only been successful following blockade of cholinesterase (Kanai and Szerb, 1965; Kuno and Rudomin, 1965; Mitchell and P h i l l i s , 1962; Mitchel l , 1963; Szerb, 1963). Criterion 1) is also not with-out exception as i t is possible that the nature of the response to a p a r t i -cular transmitter is time and/or concentration dependent. For example, an excitatory agent when applied exogenously could produce a depolarization blockade or desensitization of receptors i f applied at a high concentration or for a prolonged period. Under physiological conditions, inactivating mechanisms may terminate the action of released transmitter rapidly so that the phase of depolarization blockade or desensitization may not be appa-- 23 -rent. In addition, i t is possible that a given compound has dissimilar effects at different regions of the neurone. For example, GABA has been shown to hyperpolarize neuronal somata but depolarize dendrites of hippo-campal cel ls (Alger and N i c o l l , 1979; Langmoen et a l ; , 1978). In spite of some inadequacies, however, the c r i t e r i a for identif icat ion of transmitters have generally proven to be quite useful . 2.7.1 Excitatory - amino acids. Glutamate and aspartate are the most l i k e l y candidates to serve as neurotransmitters in major excitatory pathways of the hippocampus (Storm-Mathisen, 1977a). Nadler et a l . (1976) found that glutamate and aspartate were released in a Ca+ +-dependent manner following K +-induced depolarization of hippocampal fragments. Electr ical stimula-tion of the perforant path resulted in an increase in endogenous glutamate release (Crawford and Connor, 1973; White et a l ; , 1977) and lesions of this pathway caused reduced glutamate release (Nadler et a l ; , 1976, 1978). Con-trasting results, however, were presented by Di Lauro et- a l ; (1981) who claim that aspartate is the major transmitter of the perforant path f i b r e s . Their argument was that studies done by Nadler et a l . (1976, 1978) allowed sufficient time for the regeneration of aspartergic terminals prior to conducting their assays. Following interruption of the commissural input, the release of aspartate but not glutamate was s ignif icant ly reduced (Nadler et a l ; , 1978). Kainic acid-induced destruction of CAg cel ls (which give rise to Schaffer collaterals and commissural fibres) results in a decrease in both aspartate and glutamate in the dorsal hippocampus (Fonnum and 3 Walaas, 1978). Autoradiographic studies using H-glutamate indicate that the label is concentrated in the inner one third of the dentate area, - 24 -stratum oriens and stratum radiatum (where commissural and Schaffer fibres terminate) and in the hi lar region of the dentate, where mossy fibres are found (Iversen and Storm-Mathisen, 1976; Storm-Mathisen and Iversen, 1979). Destruction of commissural and Schaffer collateral fibres led to a marked decrease in glutamate uptake (the main inactivating mechanism for glutamate) from the CA^ area (Storm-Mathisen, 1977b). D-Aspartate has been thought to be a marker for both glutamate and aspartate terminals (Balcar and Johnston, 1972; Davies and Johnston, 1976; Roberts and Watkins, 1975) but appears to selectively favour glutamate terminals (Malthe-S(6rrensen et- a l ; , 1979). Loaded D-aspartate was released in a Ca -dependent manner follow-ing stimulation of commissural and Schaffer fibres in the hippocampal s l ice (Malthe-Stfrrensen et a l . , 1979). Uptake studies have to be viewed with caution because, f i r s t l y , the high a f f i n i t y uptake system does not d i s t i n -guish between glutamate and aspartate as both amino acids u t i l i z e the same carrier (Balcar and Johnston, 1972; Roberts and Watkins, 1975; Young et a l ; , 1974) and, secondly, endogenous and added glutamate are localized in d i f f e r -ent synaptosomal compartments (Kvamme, 1981) so that experiments using exo-genously loaded glutamate or aspartate may give misleading results . The synthesis of releasable glutamate can be brought about by loading slices with its precursors glucose and glutamine (Hamberger et al- . , 1978, 1979a, 1979b). Electrophysiological studies on hippocampal neurones demonstrate a powerful excitation of pyramidal cel ls by glutamate and aspartate (Dudar, 1974; Schwartzkroin and Andersen, 1975; Zanotto and Heinemann, 1983) with the most sensitive areas located at the dendrites where excitatory inputs terminate. The evidence presented strongly supports the poss ibi l i ty that an - 25 -acidic amino acid is the neurotransmitter in the perforant path, commissural input and Schaffer collateral input. However, i t is not possible at this point to be certain whether glutamate and/or aspartate are involved in these pathways. In contrast to the inputs discussed, the other major afferent system, the mossy fibres (arising from dentate granule cel ls ) does not seem to u t i l i z e either glutamate or aspartate as a transmitter (Nadler e t - a l ; , 1978; White et• al- . , 1979). The report by Crawford and Connor (1973) claims that mossy fibres release glutamate. Their evidence is based on collection of evoked release of endogenous glutamate by stimulation in the entorhinal cortex. Such stimulation would activate the trisynaptic loop consisting of perforant path, mossy fibres and Schaffer col la terals . Therefore, the va l id i ty of such an assumption is highly questionable because a selective activation of the mossy fibres using this mode of stimulation is impossi-ble . The transmitter of the mossy fibres remains largely unknown but i t is interesting that granule cel ls and mossy fibres showed dense staining for dynorphin (McGinty et a l ; , 1983). Electrophysiological and pharmacological studies support the presence of multiple receptors for glutamate in the hippocampus (Col 1ingridge e t - a l ; , 1983a; Fagni et - a l 1 9 8 3 ; Koerner and Cotman, 1982). These have been characterized by Fagni et - al.- (1983) and are divided into four subtypes: 1) the NMDA receptor which is activated by N-methyl-D-aspartate and exhibits desensitization; 2) the kainate receptor which is activated by kainic acid and does not desensitize; 3) the G2 receptor which is activated by glutamate and aspartate and shows desensitization; 4) the Gl receptor which is a c t i -- 26 -vated by DL-homocysteate and does not desensitize. Bear in mind that these four subtypes are glutamate receptors and are, therefore, activated to varying degrees by this amino acid. Each subtype is identified by the most potent agonist at that particular receptor. On the basis of electrophysio-logical studies, the above authors postulate that the Gl receptor is the synaptic receptor. 2.7.2 Acetylcholine; Denervation studies have demonstrated that acetylcholinesterase (AChE) and choline acetyl transferase (ChAT) (the en-zymes responsible for the breakdown and synthesis of acetylcholine [ACh], respectively) disappear almost entirely from the hippocampal region after interruption of the fimbrial fibres or lesions in the septum (Lewis e t - a l ; , 1967; Mellgren and Srebro, 1973; Shute and Lewis, 1961). Fibres showing accumulation of AChE have been traced from the hippocampus back to the medial septum (Lewis and Shute, 1967). The densest innervation of choliner-gic fibres occurs in stratum oriens in the CA3 region, the hilus of the dentate and to a lesser extent, the dentate molecular layer (Crutcher et  a l . , 1981; Kimura et a l . , 1981; Lynch e t - a l ; , 1978; Storm-Mathisen, 1977a). Spontaneous as well as stimulus-evoked (by stimulation of medial septum) release of ACh has been collected from the hippocampus (Dudar, 1975; Smith, 1972). No evoked release of ACh is seen with stimulation of the lateral septal nucleus, contralateral hippocampus or caudate nucleus and the evoked release mentioned above is dependent on the integrity of the septo-hippocam-pal projection. Electrophysiological and pharmacological studies have also supported the cholinergic nature of the septo-hippocampal pathway (Brucke et  a l . , 1963; Green and Arduini , 1954; Stumpf, 1965). When applied iontophore-- 27 -t i c a l l y , ACh usually produces a slow excitation in hippocampal cells that is thought to be mediated by activation of muscarinic receptors (Biscoe and Straughan, 1966; Bland et a l , 1974; Herz and Nacimiento, 1965; Salmoiraghi and Stefanis, 1965; Stefanis, 1964; Steiner, 1968). 2.7.3 Y-Aminobutyric acid; The well characterized basket c e l l s , which are inhibitory interneurones in the hippocampus, are thought to u t i l i z e Y-aminobutyric acid (GABA) as their neurotransmitter (Andersen, 1975; 3 Storm-Mathisen, 1977a). H-GABA localization and glutamic acid decarboxy-lase (GAD, an enzyme for the synthesis of GABA) act ivi ty were measured in hippocampus (Iversen and Bloom, 1972; Ribak et - a l . , 1978; Storm-Mathisen, 1972, 1976). The lack of reduction in GAD act ivi ty and GABA uptake after lesions of afferent pathways suggests that GABA-containing cel ls are i n t r i n -sic to the hippocampal formation (Nadler e t - a l ; , 1974; Storm-Mathisen, 1972, 1975; Storm-Mathisen and Fonnum, 1972; Storm-Mathisen and Guldberg, 1974). These biochemical findings are consistent with electrophysiological studies which show that interneurones mediate the inhibition of pyramidal cel ls (Andersen et a l . , 1964a, 1964b). Physiological and pharmacological data indicate that pyramidal cel ls are inhibited by iontophoretically applied GABA and the response to this agent as well as IPSPs are blocked by bicucul-l i n e , a GABA antagonist (Biscoe and Straughan, 1966; Curtis et a l ; , 1970; Stefanis, 1964). The sensi t ivi ty to GABA appears to be greatest at the cel l bodies, corresponding with the termination of GABAergic nerve terminals (Schwartzkroin et a l ; , 1974). 2.7.4 Aromatic -amines. In the rat , noradrenaline (NA) containing nerve terminals and axons are diffusely distributed in the hippocampus but - 28 -are concentrated in the hilus of the dentate and to a lesser extent in the CA^ and CA^ regions (Blackstad et a l . , 1967). The afferent fibres enter the hippocampal formation via the fimbria, the fornix, the cingulum bundle and the amygdaloid area (Fuxe, 1965; Fuxe et- a l ; , 1969; Lindvall and Bjorklund, 1974; Ungerstedt, 1971). Following lesions in the locus coeru-leus and NA containing pathways ascending from the locus coeruleus, exten-sive reduction of NA and NA markers were observed in the hippocampal region (Anden et a l . , 1966; Lindvall and Bjorklund, 1974; Thierry et - al- : , 1973; Ungerstedt, 1971). It was also found that dopamine-e-hydroxylase, a synthe-sizing enzyme for NA, was drast ical ly reduced after destruction of the locus coeruleus (Ross and Reis, 1974). Therefore, i t appears that the major or, perhaps, only noradrenergic input to the hippocampus arises from the locus coeruleus and there are no intr insic NA neurones in the hippocampal forma-t ion . The effect of iontophoretically applied NA on hippocampal neurones is inhibitory (Biscoe and Straughan, 1966; Herz and Nacimiento, 1965; Salmoiraghi and Stefanis, 1965; Stefanis, 1964) and appears to mimic the action of stimulation of the locus coeruleus in producing inhibition (Segal and Bloom, 1974a, 1974b). This inhibit ion was shown to be different from recurrent basket cel l inhibition and was not blocked by bicuculline (Storm-Mathisen, 1977a). The serotonergic projection to the hippocampus arises from the medial and dorsal raphe nuclei (Azmitia and Segal, 1978; Moore and Halaris, 1975). Although this input is distributed diffusely throughout the hippocampus, i t appears to be concentrated in the stratum moleculare and the subiculum (Fuxe et a l . , 1970; Fuxe and Jonsson, 1974). Serotonin (5-HT) containing axons - 29 -invade the hippocampus via the fimbria and the cingulum bundle (Bjorklund et  a l . , 1973; Fuxe, 1965; Moore and Halaris, 1975). Following destruction of the raphe nuclei , 5-HT, 5-HT uptake and tryptophan-5-hydroxylase (a synthe-sizing enzyme for 5-HT) decrease drast ical ly in hippocampus and various forebrain regions (Kuhar et- a l ; , 1972). As with NA, iontophoretically applied 5-HT also causes inhibition of hippocampal pyramidal cel ls (Biscoe and Straughan, 1966; Herz and Nacimiento, 1965; Salmoiraghi and Stefanis, 1965; Segal, 1975; Stefanis, 1964). Electr ical stimulation in the raphe nuclei appears to produce an inhibition of prolonged time course in pyrami-dal cel ls (Segal, 1975). The inhibition produced by raphe stimulation was enhanced by the 5-HT uptake blocker p-chlorophenylalanine and inhibited by the tryptophan-5-hydroxylase inhibitor p-chlorophenylimipramine (Segal, 1975, 1976). These data support the existence of a raphe-hippocampal i n h i -bitory pathway using 5-HT as a neurotransmitter. 2.8 Long-term synaptic potentiation in hippocampus The phenomenon of long-term potentiation (LTP) of synaptic transmis-sion can be e l i c i ted in various regions of the intact hippocampus (Bliss and Gardner-Medwin, 1973; Bliss and Lfimo, 1973; Douglas and Goddard, 1975) as well as in vitro hippocampal slices (Alger and Teyler, 1976; Andersen et  al.- , 1977; Lynch et a l . , 1976; Schwartzkroin and Wester, 1975; Yamamoto and Chujo, 1978) (Figure 3). The increase in synaptic efficacy during.LTP can be observed as an enhancement in the size of the population spike or the population EPSP, a reduction in the latency to onset of the population spike, an increase in the rate of rise and amplitude of the intracellular EPSP or a decrease in threshold for synaptically evoking an all-or-none - 30 -10 ms Figure 3 Tetanus-induced long-term potentiation (LTP) of synaptically-driven CA^ responses. Records in the top row show extracellular population spikes recorded in the CAj c e l l body layer and records in the bottom row are population EPSPs monitored at the apical dendritic zone. Both sets of responses were evoked by stimulation of stratum radiatum. On the extreme lef t are control responses. Subsequent to a 400 Hz, 200 pulses tetanus delivered to stratum radiatum, the post-tetanic responses were followed for 60 minutes. Records shown are for 1, 15, 30 and 60 minutes post-tetanus. Negativity is downwards. - 31 -action potential in a CA^ c e l l . The induction of LTP is believed to be Ca -dependent and cannot be e l i c i t e d i f the tetanic stimulation is d e l i -vered in C a + + - f r e e medium (Dunwiddie and Lynch, 1979; Wigstrb'm et- - a l . , 1979). Furthermore, a C a + + agonist, S r + + , can support both synaptic transmission and LTP in the hippocampal s l ice (Wigstrb'm and Swann, 1980). A transient elevation of extracellular C a + + can produce a post-application LTP-like state (Turner et a l . , 1982), a condition that is associated with a prolonged increase in intracellular C a + + content (Baimbridge and M i l l e r , 1981). In the CA^ region, LTP has been demonstrated to be input specific ( i . e . , i t can be observed only in a previously tetanized but not a non-tetanized input impinging on the same population of CA^ neurones) (Andersen et al- . , 1977; Lynch et a l . , 1976). In contrast to the CA^ area, LTP in the CAg region appears not to exhibit input select ivi ty (Misgeld et  a l ; , 1979; Yamamoto and Chujo, 1978). The induction of LTP requires a cooperativity of input f ibres , that i s , a sufficient number of presynaptic fibres has to be activated. A "weak" subsynaptic response that is evoked by stimulation of the input with low stimulus intensity wi l l usually exhibit marginal or no LTP following a tetanus to the same input. However, i f the tetanus of the "weak" input is either paired with a tetanus to another "strong" separate input (to the same population of postsynaptic neurones) or the stimulus intensity is increased during the tetanus to recruit more f ibres , associative LTP of the "weak" input results (Barrionuevo and Brown, 1983; Lee 1983a; McNaughton, 1982; McNaughton et - a l ; , 1978; Robinson and Racine, 1982). It is believed by some investigators that the mechanisms underlying LTP of the population spike and the population EPSP are dissimi-- 32 -lar because a potentiation of one can be observed without any apparent change in the other (Bliss et a l . , 1983; Bliss and Umo, 1973). In addi-t ion, i t was observed that the time courses of decay for LTP of the popula-tion spike and population EPSP are not necessarily the same (Douglas and Goddard, 1975). Since a l l the studies above were conducted on extracellular f i e l d potentials, caution has to be exercised in interpreting the results . An extracellular recording at the ce l l body layer of the CA^ region wil l yield a positively directed population EPSP that is interrupted by a nega-t ively going population spike. The reason for this is that the EPSP is generated by a synaptically-induced depolarization of the dendrites whereas the site for spike in i t ia t ion is close to the soma (Andersen et- a l ; , 1980a). Obviously, these two f ie lds of opposing polarity overlap and i t is possible that an increase in size of one wil l result in an apparent diminu-tion of the other. Furthermore, any changes in the IPSP which is a positive potential when recorded at the cel l body layer may cause a change in the size of the EPSP i f their time courses overlap. Therefore, with these uncertainties, firm conclusions regarding the differences between LTP of the population spike and population EPSP cannot be reached without intracellular studies. Early experiments involving hippocampal lesions and ablations have led many investigators to agree that the hippocampus is in some way involved in learning and memory (Best and Best, 1976; Coleman and Lindsley, 1977; Green, 1964; Isaacson, 1974; O'Keefe, 1983; O'Keefe and Nadel, 1978; Olds et al- . , 1972; Penfield and Milner, 1958; Scoville and Milner, 1957; Segal and Olds, 1973). However, there is disagreement as to what kind of memories the - 33 -hippocampus may be responsible for . Consistent with this hypothesis, i t was shown that the f i r i n g rates of hippocampal neurones gradually increased during the course of classical conditioning of the rabbit nic t i ta t ing mem-brane response (Berger and Thompson, 1978; Berger et a l . , 1976). It was or iginal ly proposed by Hebb (1949) that memory involves an increase in synaptic transmission and that the information is stored as a result of long-lasting changes in some synaptic properties. Therefore, i t was conclu-ded that LTP in the hippocampus could be a physiological mechanism for learning and memory at the cel lular level (Berger and Thompson, 1978; Chung, 1977; Teyler, 1976). Evidence for this claim was provided by Berger (1984) who showed that prior induction of LTP in the hippocampus of intact animals by tetanically stimulating an input results in a faster rate of classical conditioning when compared to untetanized control animals. If LTP i s , indeed, linked to learning, then as an animal learns a task, one should see an increase in the size of the synaptically evoked population response. This appears not to be true because the amplitudes of the population EPSP as well as the population spike induced by perforant path stimulation were measured before and after classical conditioning and neither was found to be increased after learning of the task (Laroche, 1985). Some evidence ques-tions whether the hippocampus is even involved in learning and memory. Rats with large bilateral lesions of the hippocampus were s t i l l able to learn and remember tasks as well as undamaged controls i f the lesions were carried out in stages (Isseroff e t • a l . , 1976; Stein et a l ; , 1969). Much of the research that is used to implicate the hippocampus in memory processes ut i l izes lesion techniques which inadvertently cause extensive damage to other brain - 34 -areas and fibres that pass through this structure. In a recent study that used neurotoxins to selectively destroy hippocampal neurones without damag-ing fibres-of-passage or afferents to the hippocampus, i t was demonstrated that under such conditions, the performance of complex memory tasks was minimally affected (Jarrard, 1983, 1985). In the same studies, aspiration of the hippocampus and interruption of the main afferent and efferent path-ways resulted in markedly poorer performance on the same tasks (Jarrard, 1985). In light of the contradicting evidence, i t is not possible to say with certainty that the hippocampus is involved in learning and memory, much less speculate on the role of hippocampal LTP in these processes. The elucida-tion of the mechanisms responsible for LTP, whether i t has any relevance to learning and memory, wil l provide substantial information with regard to understanding the plastic properties of synapses in the central nervous system. LTP is not specific to the hippocampus as i t has been shown to be present in the neocortex (Kasamatsu et a l ; , 1981; Lee, 1983b), mammalian sympathetic ganglion (Brown and McAfee, 1982; Koyano et- a l : , 1985), abdomi-nal ganglion of Aplysia (Caste!lucci and Kandel, 1976; Castellucci et- a-lv, 1970) and crayfish neuromuscular junction (Baxter et a l . , 1985). It was established in some systems (crayfish neuromuscular junction, abdominal ganglion of Aplysia and rat sympathetic ganglion) using the method of quan-tal analysis that LTP could be accounted for by a presynaptic mechanism through increased transmitter release (Baxter et a l ; , 1985; Briggs et a l : , 1985; Castellucci and Kandel, 1976; Koyano et - al . - , 1985). It was also reported that LTP of the septo-hippocampal input to CA^ neurones involved - 35 -an increase in quanta! content but there was no change in the quantal unit . Furthermore, the same authors showed that there was no change in the sensi-t i v i t y of subsynaptic receptors on CAg neurones to applied acetylcholine, the transmitter released by the septo-hippocampal input (Voronin, 1980, 1983). Mellgren and Srebro (1973) reported that the septo-hippocampal projection to the hippocampus terminates in stratum oriens and stratum radiatum of CAg neurones. Therefore, these synaptic contacts on the CAg pyramidal cel ls could be located at some distance from the soma. In the studies of Voronin (1980, 1983) recordings were done in CAg cel l bodies, so i t is possible that increases in the quantal unit did occur at the den-drites but were not recognized at the soma due to decrement of the poten-t i a l s during electrotonic propagation to the recording s i te . Perhaps, a better system to conduct quantal analysis upon is the mossy fibre-CAg synapse where i t has been demonstrated that the excitatory synapses termi-nate very close to the soma (Blackstad et a l ; , 1970; Blackstad and Kjaerheim, 1961; Brown and Johnston, 1983; Johnston and Brown, 1983; Hamlyn, 1961; Hamlyn, 1962; Lorente de No, 1934). Therefore, the locus for LTP in the hippocampus is not clear. Since there is evidence for the involvement of both pre- and postsynaptic elements in LTP of int r ins ic hippocampal sys-tems, the general consensus is that both components play a role . The following observations suggest that LTP is presynaptic. 1) LTP in the CA-^  area is input specific (Andersen et - a l ; , 1977; Lynch et- a l ; , 1976). 2) High frequency tetanic stimulations that produce minimal synchro-nous discharge of the postsynaptic neurones are more favourable for inducing LTP whereas lower frequency trains which produce frequency f a c i l i t a t i o n tend - 36 -to result in homo- and heterosynaptic depression (Bliss and Ltfmo, 1973; Dunwiddie and Lynch, 1978; Sastry e t - a l . , 1984a). 3) Passive membrane pro-perties of CA-^  neurones such as resting membrane potential , input res i s -tance and time constant were unaltered during LTP (Andersen et - al .- , 1980b; Barrionuevo and Brown, 1983). 4) Synchronous postsynaptic discharge is not required during the tetanic stimulation to successfully induce LTP (Wigstrom et a l . , 1982). The most obvious interpretation of points 1) to 4) is that LTP is presynaptic, but another possible explanation is that a postsynaptic change which occurs at the dendritic region is responsible. 5) More d e f i n i -tive evidence for a presynaptic involvement is provided by several labora-tories . These investigators demonstrate that stimulus-evoked release of both exogenously loaded as well as endogenous neurotransmitter is increased during LTP (Bliss e t - a l ; , 1985; Dolphin et a l : , 1982; Skrede and Malthe-Stfrrensen, 1981). 6) A decrease in presynaptic terminal exci tabi l i ty is associated with LTP and this exc i tabi l i ty change is not seen at non-terminal axonal regions (Sastry, 1982). The following evidence has been presented in support of a postsynaptic locus for LTP. 1) The number of Na+-independent glutamate binding sites (presumed by the authors to be subsynaptic receptors although pharmacologi-cal and physiological evidence is lacking) is increased following tetanic stimulations to an input (Baudry and Lynch, 1980a). 2) Intracellular injec-tions of EGTA into CA-^  cel ls blocked the induction of LTP (LTP induction is thought to be Ca -dependent) (Lynch et a l . , 1983). 3) The associative nature for the induction of LTP (where a presumably separate input can "cooperate" with the test input to f a c i l i t a t e production of LTP in the la t -- 37 -ter) has convinced many investigators that the common link and, therefore, locus for LTP is the postsynaptic neurone (Douglas et a l . , 1982; McNaughton, 1982; McNaughton et- a l ; , 1978; Robinson and Racine, 1982). A subsequent investigation has shown, however, that the cooperativity among afferents does not correlate with enhanced postsynaptic discharge during conditioning, but rather could be due to interactions among presynaptic fibres (Lee, 1983a). 4) Changes in inhibitory processes (presumably postsynaptic i n h i b i -tion) are thought to play a role in LTP. Unlike the situation in the CA^ region, LTP in the CAg region is not input specific and the size of the EPSP appears to be inversely related to the magnitude of the IPSP in CAg neurones (Misgeld et aT., 1979; Yamamoto and Chujo, 1978). It has been shown recently, however, that a decrease in inhibition is not responsible for LTP in the CAg region ( G r i f f i t h et a 1;, 1986). In the dentate gyrus, enhanced inhibition of granule ce l ls during afferent tetanization succeeded in blocking the induction of LTP (Douglas et a l ; , 1982). Furthermore, blockade of inhibition using GABA antagonists fac i l i ta ted induction of LTP in the CA^ area (Wigstrb'm and Gustafsson, 1983), Although i t appears that the induction of LTP may be enhanced by the blockade of presumed post-synaptic inhibi t ion , the maintenance of the phenomenon, at least in the CA^ region is not dependent on inhibitory processes (Haas and Rose, 1982, 1984). 5) Anatomical changes in dendritic morphology have been suggested as possible postsynaptic mechanisms for LTP. A swelling of dentate granule ce l l spines was reported following tetanic stimulations to the perforant path (Fifkova and Van Harreveld, 1977; Van Harreveld and Fifkova, 1975). These authors postulated that this prolonged change can account for the - 38 -increase in synaptic efficacy during LTP. This explanation is not entirely plausible because i t was shown earlier that swelling of spines accompanies spreading depression (Van Harreveld and Khattab, 1967). Furthermore, the authors used a tetanic stimulus frequency (30 Hz) that could produce f r e -quency f a c i l i t a t i o n during the conditioning train and a following post-tetanic homo- and heterosynaptic depression of synaptic transmission (Dunwiddie and Lynch, 1978; Sastry et a l : , 1984a). It was demonstrated in a later study (Lee e t - a l ; , 1980) that a higher stimulus frequency (100 Hz), which is more favourable for e l i c i t i n g LTP, fa i led to produce any swelling of spines. A recent report further disproves the spine swelling theory by showing that there is no correlation between spine swelling and LTP (Chang and Greenough, 1984). The brief bursts of high frequency stimuli in the study of Lee et a l . (1980) produced a rearrangement of synapses to result in an increase in the density of dendritic shaft synapses and a decrease in spine v a r i a b i l i t y that is thought to improve synaptic transmission. Using computer simulated models, i t was deduced that various alterations in den-d r i t i c spines can be conducive to synaptic amplification (Horwitz, 1981; Mil ler et a l . , 1985). Several hypotheses have been presented regarding the locus and c e l l u -lar mechanisms that may be responsible for LTP. 1) Baudry and Lynch (1980a) suggested that an increase in the number of subsynaptic neurotransmitter (glutamate) receptors can account for LTP. A tetanic stimulation of an input presumably results in activation of the postsynaptic membrane to allow ++ ++ Ca influx . This increase in intracellular Ca then activates a pro-tease which then unmasks new glutamate receptors (Baudry and Lynch, 1980b; - 39 -Baudry et a l ; , 1981a, 1981b; Vargas and Costa, 1981). Evidence that has been reported demonstrates an increase in Na -independent glutamate bind-ing sites following tetanic stimulations in hippocampal slices (Baudry et  a l . , 1980; Lynch et- a l ; , 1982). This increase in binding s i tes , which may or may not be receptors, appears to be irreversible (Baudry et a l . , 1983) ++ and is inducible by elevated Ca concentrations (Baudry and Lynch, 1979). 2) Changes in dendritic spine morphology have also been postulated to play a role in the synaptic enhancement observed during LTP. Based on anatomical and computer simulated data, several investigators have concluded that certain changes in the shape and size of dendritic spines and the rear-rangement of synaptic contacts can contribute to fac i l i ta ted synaptic trans-mission (Desmond and Levy, 1981, 1984; Lee et a l . , 1980; Horwitz, 1981; Mil ler et aT: , 1985). As discussed previously, the hypothesis regarding spine swelling (Fifkova and Van Harreveld, 1977; Van Harreveld and Fifkova, 1975) appears not to be a feasible one. 3) The contribution of inhibitory processes to the induction and maintenance of LTP has been investigated. It was postulated that the generation of LTP is f a c i l i t a t e d i f the conditioning tetanus is delivered during a blockade of inhibition with GABA antagonists (Wigstrom and Gustafsson, 1983.)".- However, i t appears that the maintenance of LTP in the CA^ area is not dependent on a reduction in the ortho-dromically-evoked or recurrent IPSP. In contrast, LTP in the CAg region is not input specific and is accompanied by a diminution of the IPSP (Misgeld et a l . , 1979; Yamamoto and Chujo, 1978). Apparently, in the CAg region, the size of the EPSP during LTP is inversely related to the magni-3 tude of the IPSP. 4) The stimulus-evoked uptake of D- H-aspartic acid (a - 40 -marker for glutamatergic nerve terminals) is decreased following high f r e -quency tetanic stimulations to an input (Wieraszko, 1983). This observation led to the hypothesis that a reduction in the uptake of neurotransmitter could account for LTP because of greater a v a i l a b i l i t y of the substance to the subsynaptic receptors. Although this may be true, i t is known that a blockade of transmitter uptake does not necessarily result in enhanced synaptic transmission (Curtis et a l . , 1976). It is possible that uptake mechanisms are activated with a slower time course than synaptic transmis-sion so that the interaction with subsynaptic receptors occurs before the transmitter is eliminated. 5) Presynaptic terminal alterations resulting in an increased release of transmitter has been thought to occur during LTP. 3 It has been demonstrated that evoked release of loaded H-aspartate is enhanced following tetanic stimulation of inputs in stratum radiatum (Skrede and Malthe-S^rrensen, 1981). The increase in stimulus-evoked endogenous glutamate release parallels that of LTP in the dentate region after tetanic stimulation of the perforant path (Bliss et- a l ; , 1985; Dolphin et - - a l . , 1982). The exci tabi l i ty of presynaptic terminals is reduced during LTP (Sastry, 1982). This exc i tabi l i ty change has a similar time course to LTP and, therefore, may play a role in this potentiation. It was suggested that a hyperpolarization of the presynaptic terminals would result in a reduced exci tabi l i ty as measured by Wall's technique (Wall, 1958; Wall and Johnson, 1958) and an increase in the action potential height (Eccles and Krnjevic, 1959a, 1959b; Lloyd, 1949) which brings about an enhancement in the evoked release of transmitter (Hubbard and W i l l i s , 1962; Takeuchi and Takeuchi, 1962). 6) N-Methyl-D-aspartate (NMDA) receptor activation is thought to be - 41 -necessary for LTP induction (Col 1ingridge et a l . , 1983b; Harris et - a 1., 1984). 2-Amino-5-phosphonovalerate (APV), reportedly a specific NMDA recep-tor antagonist (Col 1ingridge et- a l ; , 1983a) counteracts the development of LTP when the tetanus is given during application of the drug (Col 1ingridge et a l . , 1983b; Harris e t - a l ; , 1984). Iontophoretic application of NMDLA at the apical synaptic zone of CA^ neurones produced a post-application potentiation of the population EPSP evoked by stimulation of stratum radia-tum (Col 1ingridge et a l ; , 1983b). A prolonged negative wave that is recor-ded in the synaptic region during tetanic stimulation of the input is thought to be mediated by an activation of NMDA receptors because i t can be reduced by application of APV (Wigstrb'm and Gustafsson, 1984). This nega-tive wave may be important for LTP induction because a blockade of i n h i b i -tion with picrotoxin during tetanic stimulation produces an enhancement of the wave as well as fac i l i ta ted LTP development following the treatment. These authors suggested that the dendritic depolarization achieved by NMDA receptor activation during tetanization is fac i l i ta ted during a reduction in postsynaptic inhibition and that i t is this depolarization which produces optimal conditions for in i t ia t ing LTP. 7) Neuroleptic drugs which are c a l -modulin antagonists impair the induction of LTP (Finn et a l . , 1980; Mody et  a l . , 1984) raising the poss ibi l i ty that this C a + + buffering protein is involved in LTP. 8) Selective depletion of noradrenaline and 5-HT in rats results in a reduction in tetanus-induced LTP, suggesting that these mono-amines play a role in this phenomenon (Bliss et a l 1 9 8 1 , 1983). Further-more, i t was shown that applied noradrenaline results in an LTP-like pheno-menon in the dentate gyrus (Neuman and Harley, 1983). 9) LTP could not be - 42 -e l i c i t e d in cel ls injected intracel lular ly with the K + channel blocker CsCl (Haas and Rose, 1984). Therefore, a reduction of a potassium conduc-tance could be reponsible for LTP. 10) A recent report (Malenka et - a l . , 1986) provides evidence for protein kinase C (a Ca+ +-dependent phospho-l i p i d kinase which is selectively activated by phorbol esters) involvement in the induction of LTP. 2.9 Post-tetanic potentiation Post-tetanic potentiation (PTP) of synaptically transmitted responses (Eccles and Krnjevic, 1959a, 1959b; Lloyd, 1949) which has a time course of several minutes (1-4) differs from LTP in that the latter phenomenon is characterized by a much longer duration. The suggestion that PTP in the CNS is generated presynaptically has been with us since Lloyd (1949) supplied evidence in favour of this hypothesis. He observed that PTP in the spinal cord was input specific and there was no potentiation of non-tetanized heterosynaptic afferents that were known to terminate on the same postsynap-t i c motoneurones. Furthermore, a tetanus of afferents impinging on a cer-tain population of motor horn cel ls resulted in no potentiation but rather a depression of the antidromic compound action potential evoked by stimulation of the axons of these cel ls in the ventral root. A hyperpolarization of afferent fibres could be responsible for PTP because changes in the inten-s i ty and duration of the potentiation are closely associated with a positive after-potential that is recorded extracellularly from the tetanized fibres (Lloyd, 1949). This idea was later corroborated by the findings of Eccles and Krnjevic (1959b) who recorded i n t r a c e l l u l a r ^ in primary afferents in the spinal cord and observed a period of hyperpolarization of these fibres - 43 -following their tetanization. The exc i tabi l i ty of primary afferent projec-tions was also found to be decreased post-tetanically and this change paral-le ls the degree of synaptic potentiation (Wall and Johnson, 1958), an obser-vation that is in accordance with a hyperpolarization of the f ibres . The presynaptic axonal spike height appears to be increased during the period of hyperpolarization that accompanies PTP and a r t i f i c i a l anodal polarization of the spinal dorsal roots also produces a similar effect (Eccles and Krnjevic 1959b). However, attempts to alter postsynaptic potentials by polarizing presynaptic dorsal root fibres f a i l e d , the explanation given by the above authors being that the site of the polarizing electrode was too far from the terminals to produce adequate electrotonic changes in the membrane potential at the nerve endings. In studies conducted to examine the relationship between presynaptic spike size and postsynaptic potentials, i t was d i s -covered in the squid giant synapse that small presynaptic changes had a very profound postsynaptic effect (Hagiwara and Tasaki, 1958). An enhancement of the presynaptic spike height is accompanied by a f a c i l i t a t i o n of evoked transmitter release (Hubbard and W i l l i s , 1962; Takeuchi and Takeuchi, 1962). Perhaps, the most convincing evidence in support of a presynaptic locus for PTP is supplied through quantal analysis where i t was concluded that the number of quanta of evoked transmitter is increased whereas the quantal size remains unchanged (del Cast i l lo and Katz, 1954). Similar to the situation found in the spinal cord, neuromuscular junc-t ion, squid giant synapse and numerous other systems, the hippocampus also exhibits PTP. There is general agreement that the properties and require-ments for induction of PTP and LTP in this system are not the same - 44 -(Dunwiddie and Lynch, 1979; McNaughton, 1982). Since the evidence in sup-port of a presynaptic locus for PTP is f a i r l y strong in other tissues, i t is assumed by many that this phenomenon in the hippocampus is no different . 3 METHODS 3.1 Preparation of slices Transversely sectioned hippocampal slices (500 urn thick) were obtained from male Wistar rats (75-125 g). The animal was i n i t i a l l y anaesthetized with a mixture of halothane (2%) and oxygen. During this process, an ice pack was placed beneath the rat to lower i ts body temperature (rectal temperature prior to surgery was 31-32°C) so that the v i a b i l i t y of the slices could be increased. The skin covering the top of the head was cut with a scalpel in an anterior to posterior direction to expose the skull bones. The plates of the skull and the dura mater covering the brain were carefully removed and the brain was severed from the spinal cord at the pontine l e v e l . The optic nerves were also severed and the brain was taken out and quickly drenched with cold (4°C) standard medium (constitution given in the next section) to reduce metabolic rate and oxygen requirements. The hippocampi from one or both hemispheres were dissected free and transverse slices of this structure were obtained using a Mcllwain tissue chopper. To separate and arrange the s l i ces , the sliced hippocampus was placed in a petridish f i l l e d with cold standard medium which was oxygenated with carbo-gen (95% O2, % CO^). The slices were sandwiched between two nylon meshes to minimize movement during the experiments and quickly transferred - 45 -to the s l ice chamber (see Figure 4 for diagram of chamber). The whole procedure from the start of surgery until the time the slices were inserted into the bath usually took 3 minutes or less. A l l experiments conducted were on slices submerged in the perfusing medium and temperature was main-tained at 32 ± 0 . 5 ° C . The perfusing medium was constantly bubbled with carbogen and an extra l ine carrying humidified carbogen was also fed to the atmosphere above the s l i c e s . During the equilibration period (one hour) before commencement of experiments, the bath was maintained at room tempera-ture and the chamber opening was covered with a piece of parafilm to ensure that the air above the slices was saturated with oxygen. The flow rate of the medium was maintained at 3 ml/min. Only one s l i ce per animal was used i f an experiment involved using media other than the standard medium. In any event, no slices older than four hours were u t i l i z e d because of possible effects of deterioration even though they usually survived for more than 12 hours. 3.2 Constitution of perfusing media The pH of a l l media were maintained at 7.4 while bubbled with carbogen (95% 0 2 , 5% C0 2 ) . Unless otherwise specified, experiments were conduc-ted in standard medium. Standard medium: 120 mM NaCl, 3.1 mM KC1, 1.3 mM NaH 2P0 4, 26 mM NaHC03, 2 mM C a C l 2 , 2 mM MgCl 2 , 10 mM dextrose. ++ Ca -free standard medium: same as standard medium except CaCl 2 was omitted, 0.5 or 1 mM MnCl 2 was added and MgCl 2 was increased to 3.5 or 3 mM, respectively, to compensate for divalent ion content. - 46 -Figure 4 Diagram of s l ice chamber and perfusion system for maintaining hippocampal s l i c e s . AB: aluminium block; Co: extra carbogen l ine ; Gn: ground lead; GW: ground wire; HE: heating element; LM: l ines for medium; LN: lower nylon platform; Mf: manifold; Mn: manipulator; Nz: nozzle; 01: outlet for medium; SC: s l i c e chamber; SL: suction l ine ; SP: sensor probe; UN: upper nylon platform. * - 47 -Picrotoxin medium: 120 mM NaCl, 3.1 mM KC1, 26 mM NaHC03, 4 mM C a C l 2 , 4 mM MgCl 2 , 10 mM dextrose, 10 uM picrotoxin. The divalent cation concentration was increased as compared to the standard medium to minimize epileptiform act ivi ty that can be induced by picrotoxin. NaH2P04 was omitted because of problems with the s o l u b i l i t y of the increased amount of CaCl,,. This change did not s ignif icant ly affect the buffering capacity of the medium and i t was maintained at pH 7.4 while being bubbled with carbogen. Picrotoxin medium was used in the associative LTP experiments and in experiments where a fac i l i ta ted LTP was desired because i t was reported (Wigstrom and Gustafsson, 1983) that GABA antagonists fac i l i ta ted the induction of LTP. ++ ++ Ca -free (Mn ) picrotoxin medium: same as picrotoxin medium except CaC^ was omitted and MnCl 2 (1 mM) was added. MgCl 2 was i n -creased to 7 mM to maintain the divalent cation concentration. Picrotoxin-EDTA medium: same as picrotoxin medium except 200 uM EDTA (ethylenediaminetetraacetic acid, pH 7.4) was added. ++ ++ ++ ++v Ca -free (Mn ) picrotoxin-EDTA medium: same as Ca -free (Mn ) picrotoxin medium except 200 uM EDTA was added. C a + + - f r e e (Co + + ) picrotoxin-EDTA medium: same as above except 1 mM CoCl 2 was used instead of 1 mM MnCl 2 . High K + media: NaCl was reduced by the appropriate amount when KC1 was increased to maintain a constant osmolarity in control and test media. Other drug-containing media: NMDLA, NMLA, NMDA, L-glutamate (Sigma Chemicals) and DL-APV (Cambridge Research Biochemicals) were added to standard medium from concentrated stock solutions. The f inal volume - 48 -of drug-containing medium was altered by less than 0.5% using this method. 3.3 Recording systems Recording microelectrodes ( f i b r e - f i l l e d borosilicate glass, OD 1.5 mm, ID 1.0 mM, Frederick Haer and Co. , t ip 1 urn, f i l l e d with 4 M NaCl, res i s -tance 1-2 MTJ) were pulled using a Narashige microelectrode puller . The s i g -nals recorded by the micropipette or by a NaCl-fi11ed barrel of an ionto-phoresis electrode were amplified by either a DAM-5A (World Precision Instruments) or a Neurolog (Medical Systems Corp.) amplifier and were d i s -played on a DATA 6000 (Data Precision) waveform analyzer. Records of evoked responses were averaged (4-8 sweeps) by the DATA 6000 unit and were plotted on paper by a Hewlett-Packard 7470A graphics plotter. To record population spikes or population EPSPs, the recording electrode was positioned either in the cell body layer of the CA^ or CAg areas or the apical dendritic region of the CA^ neurones (100-200 urn from cell body layer) (Figure 5). The all-or-none action potentials in single CAg cel ls were monitored by placing the recording electrode in the CAg cel l body layer (Figure 6). 3.4 Stimulating -systems A Grass S88 stimulator was used to deliver current pulses through one or two channels using Grass PSIU6 constant current stimulus isolation units. The isolation units were connected to bipolar concentric metal stimulating electrodes (SNEX 100, Rhodes Electronics, resistance 1-2 M )^ for evoking population responses. Population spikes and EPSPs in the CA-^  area were evoked by stimulation of stratum oriens or radiatum and population spikes in the CA^ region were evoked by stimulation of the mossy fibres - 49 -Figure 5 Positioning of stimulating and recording electrodes for synaptically-evoked population responses. Concentric bipolar metal electrodes were used for stimulation and f i b r e - f i l l e d glass micropipettes (t ip 1-2 urn, f i l l e d with 4 M NaCl) were used for recording. To evoke synaptic responses in the CA]_ region, stratum oriens or stratum radiatum were stimulated at 0.2 Hz. Extracellular population spikes were recorded in the CAj ce l l body layer and population EPSPs at the apical dendritic zone. CA3 population spikes were monitored with a microelectrode placed in the CA3 cel l body layer and were evoked by stimulation of the mossy fibres (MF) at 0.2 Hz. The popu-lation responses were a l l evoked using negative stimulus pulses delivered through a constant current unit . - 50 -(Figure 5). Monopolar glass electrodes (similar to the recording elec-trodes), monopolar tungsten electrodes or a NaCl-fi11ed barrel of a 7 barrel iontophoresis electrode were used for stimulation and measuring threshold for activation of Schaffer collateral terminal regions of single CA^ cel ls (Figure 6). Control stimulation frequencies used to evoke synaptic and antidromic responses were 0.1-0.2 Hz. In experiments where both the stratum oriens and stratum radiatum were used, they were evoked alternately at 5 second intervals such that the frequency of stimulation of each input was 0.1 Hz. The polarity of the pulses used for stimulation was always negative. 3.5 Exci tabi l i ty- test ing of Schaffer collateral terminal regions The method employed for exc i tabi l i ty testing of presynaptic terminals is a modification of a procedure f i r s t described by Wall (1958). A record-ing electrode was placed in the CA^ cel l body layer to record an extracel-lular all-or-none spike in a CA^ neurone and a monopolar stimulating electrode was placed at the apical dendritic area of CA^ neurones where the Schaffer collaterals terminate (Figure 6). The amount of injected current required to activate' a CA^ cel l was determined using negative pulses delivered through the stimulating electrode. The position of the stimulating electrode was determined by moving i t ver t i ca l ly into the s l i ce to a location where the threshold for activation of the cel l was minimal. The threshold is taken as the amount of current that is required to d i s -charge an action potential in the cel l in an all-or-none fashion in 50% of the t r i a l s . Stimulation strengths required to f i r e the CA^ neurone ranged from 3-12 uA (0.05-0.2 ms duration). Control thresholds were monitored for at least 15 minutes to ensure s t a b i l i t y before commencing an experiment. In - 51 -Stim. Figure 6 Positioning of stimulating and recording electrodes for exci tabi l i ty testing of the Schaffer collateral terminal regions of single CA3 c e l l s . The extracellular recording electrode which was placed in the ce l l body layer to record all-or-none action potentials in single CA3 ce l ls was a f i b r e -f i l l e d glass micropipette (tip 1-2 pm, f i l l e d with 4 M NaCl). The stimulat-ing electrode which was positioned in the apical dendritic region of CA^ (where the Schaffer collaterals terminate) was either a glass micropipette similar to the recording electrode, a monopolar tungsten electrode or a 4 M NaCl-fi1 led barrel of a 7 barrel iontophoresis electrode. The current pulses used for stimulation were delivered by a constant current unit and were always negative in polari ty . The threshold for activation of the test Schaffer collateral terminal regions was taken as the amount of current required to discharge an action potential in the recorded CA3 neurone in 50 percent of the t r i a l s . - 52 -each experiment, the test f ibre was determined to be antidromic by infusing Ca + + - f ree medium (containing 1 mM M n + + , 3 mM Mg + + ) for at least 10 minutes. 3.6 Iontophoretic applications of drugs Iontophoretic applications of drugs were done using a 6 channel Neuro-phore BH2 iontophoresis unit (Medical Systems Corp.) . NMDLA, L-glutamate and DL-homocysteate (DLH) were applied iontophoretically, the former two compounds in the CA^ apical dendritic area and the latter agent in the CA^ ce l l body layer. The appropriate drug was f i l l e d in 3 side barrels and 4 M NaCl in the remaining barrels of a 7 barrel micropipette (tip 2-3 urn). The NaCl-containing barrels were used for recording, stimulating or current balancing. The concentrations of the NMDLA, L-glutamate and DLH (Sigma Chemicals) solutions were 100 mM each and a l l were ejected from their barrels by the passing of a negative charge. Backing currents (10-15 nA, positive charge) were applied to retain the drugs in their barrels and both backing currents as well as iontophoresis of drug were automatically current balanced. The effects of injected current were examined by passing negative charge of the same magnitude as that used for drug application through N a C l - f i l l e d barrels of the 7 barrel electrode. 3.7 Iontophoretic DLH experiments DL-Homocysteate (DLH, 20-100 nA, 3 min) was applied at the CA 3 ce l l body layer to produce a transient increase in the f i r i n g rate of these neu-rones. The threshold for antidromic activation of single CA3 ce l ls by stimulation at the Schaffer collateral terminal regions was monitored before and after DLH application in standard medium (Figure 7A). In some experi-- 53 -St im. Figure 7 Experimental arrangement for monitoring e x c i t a b i l i t y changes in the Schaffer collateral terminal regions following a high frequency activation of CA3 cell bodies by an iontophoretic application of DL-homocysteate (DLH). The / barrel iontophoresis electrode was positioned in the CA3 ce l l body layer. The central barrel ( f i l l e d with 4 M NaCl) was used for recording the a l l - o r -none CA3 action potential . Three side barrels were f i l l e d with 4 M NaCl and were used for passing backing currents (10—15 nA) and current balancing during iontophoresis. The remaining 3 side barrels were f i l l e d with 100 mM DLH which was applied by passing a negative charge through these barrels . DLH (20-100 nA, 3 min) was applied in standard medium and C a + + - f r e e (containing 1 mM Mn + + ) medium. The stimulating electrode was either a monopolar glass micropipette ( t ip 1-2 urn, f i l l e d with 4 M NaCl) or a mono-polar tungsten electrode and was placed at the CAj apical dendritic zone where the Schaffer collaterals terminate. Threshold for activation of the CA3 neurone by stimulation at the Schaffer collateral terminals was deter-mined as the amount of current (negative pulses) needed to evoke the CA3 cell to discharge an action potential . - 54 -merits, the effects of DLH applied in Ca + + - f ree medium were examined on the antidromic threshold. In these cases, the perfusing medium was switched from standard medium to one containing no C a + + (with 1 mM M n + + , 3 mM ++. Mg ) for 5 minutes. DLH was iontophoresed during the last 3 minutes of this perfusion. The magnitude of the negative charge used for ejecting DLH was determined individually for each experiment, using a sustained increase in the f i r i n g rate of CA^ cel ls as the desirable objective. Upon termina-tion of DLH application in these experiments, threshold for the CA^ cel l was monitored for 30 min. DLH was again applied for 3 minutes but in the ++ standard Ca -containing medium. The post-treatment threshold followed for an additional 30 min. To eliminate the poss ibi l i ty that the 5 minute exposure to C a + + - f r e e medium alone affected the Schaffer collateral termi-nal e x c i t a b i l i t y , in some experiments the threshold was monitored during and for 15 minutes following such an exposure. The recording/iontophoresis electrode was thought to be at the c e l l bodies of the CA^ neurones for several reasons. F i r s t l y , the electrode was placed in the CA^ cel l body layer under visual control with the aid of a microscope. Secondly, the f i r i n g rate of the recorded neurone was i n -creased during DLH application, regardless of whether i t was given in the presence or absence of C a + + . This observation suggests a direct effect on the recorded neurone. Furthermore, neuronal somata rather than axons res-pond to amino acid application with an increase in f i r i n g rate. The stimu-lating electrode was thought to activate the Schaffer collaterals because cel ls could s t i l l be evoked antidromically during Ca + + - f ree medium perfu-sion. Cells that could not be evoked by such a stimulation during C a + + -- 55 -free medium infusion, perhaps because synapses were involved between stimu-lating and recording si tes , were not included in the results . 3.8 APV, NMDA, NMLA and NMDLA experiments NMDLA (10-100 nA, 1-2 min) was iontophoresed onto the apical dendritic area of CA-^  (100-200 urn from cel l body layer) by passing a negative charge through the electrode barrel containing the drug solution. The CA^ popu-lation spike evoked by stratum radiatum stimulation was monitored before, during and after the drug application using a separate recording electrode placed in the cell body layer. It was assumed that NMDLA application was affecting the population of neurones recorded from i f the appearance of the population spike was altered during the iontophoresis. NMDLA (25-400 uM, 1-2 min), NMDA (50 uM, 2 min), NMLA (50 pM, 2 min) and DL-APV (25-100 uM, 4 min) were added to either standard medium or C a + + - f r e e (1 mM M n + + , 3 mM Mg + +) medium and were applied to the whole bath. In experiments where NMDLA was applied during APV infusion, the amino acid was ini t iated 2 minutes after APV application began. To ensure that the population spike was abolished by Ca -free medium application before treatment with NMDLA, the former was infused for 3 minutes before commence-ment of NMDLA application. When tetanic stimulations were delivered to either the mossy f ibre or stratum radiatum inputs during APV infusion, i t was given 30 seconds before termination of the drug application. The ef-fects of NMDLA were examined on the stratum radiatum-CA^ and mossy f i b r e -CAg population spikes as well as Schaffer collateral terminal excitabi-l i t y . Because of technical d i f f i c u l t i e s in stimulating and recording, the effects of drug application on the CA^ population spike and the Schaffer - 56 -collateral terminal exc i tabi l i ty were done separately. In Figures 13A, 13B_, 15A, 15B, 16A and 16JB, six individual sets of experiments were conducted. 3.9 Presynaptic interaction experiments Schaffer collateral terminal exc i tabi l i ty was determined as described previously (Figure 6). The threshold for activation of a CA^ cel l by stimulation at the Schaffer collateral terminals was determined as the amount of current needed to discharge the cel l in 50% of consecutive at-tempts. The influence of the activation of other fibres on the exci tabi l i ty of the experimental f ibre was determined by using a separate conditioning bipolar electrode placed 100-200 urn from the test stimulating electrode (Figure 8). The conditioning stimulation consisted of 5 or 10 pulses at 100 Hz delivered every f ive seconds. The intensity of the conditioning stimula-tion was varied between 60-120 uA and the duration of each of the pulses was held constant at 0.2 ms. For each experiment, i t was determined that the test f ibre was not discharging in response to the conditioning stimulation. In some additional experiments, the conditioning stimulation was delivered through the same electrode used for measuring the test threshold (Figure 9). The stimulus in these experiments consisted of 1, 5 or 10 pulses at 100 Hz delivered every 5 seconds. Furthermore, the stimulation intensities were also varied between 4-80 uA (duration of each pulse was 0.1 or 0.2 ms). For both sets of experiments, the interstimulus interval between the condition-ing and test stimuli was varied between 10 and 300 ms. Ca++-dependence of the conditioning effect was also examined by conducting the same experiments ++ ++ ++ ++ in Ca -free medium (containing 1 mM Mn , 3 mM Mg ). The Ca -free medium application also verif ied that the evoked spikes were a n t i -- 57 -Figure 8 Experimental arrangement for observing the effects of act ivi ty in other pre-synaptic fibres on the exc i tabi l i ty of the test Schaffer collateral termin-a l . The recording electrode (glass, t ip 1-2 urn, f i l l e d with 4 M NaCl) was placed in the CA3 c e l l body layer to monitor all-or-none action- potentials in single CA3 c e l l s . The test stimulating electrode (S2) which was either a glass micropipette similar to the recording electrode or a monopolar tung-sten metal electrode was positioned in the apical dendrites of the CAj area to monitor the threshold for antidromic activation of the test CA3 cell from i ts terminal regions. The conditioning (SI) electrode was a con-centric bipolar metal electrode placed in another region of the CAj apical dendrites (100-200 um away from the test S2 electrode). It was established that the conditioning (5 or 10 pulses at 100 Hz) did not activate the test cel l to discharge an action potential (see record at bottom of figure, nega-t i v i t y is down). The test threshold (S2) was determined at conditioning (Sl)-test (S2) interstimulus intervals between 10-300 ms. - 58 -SI (cond. stim) and S2 ( test stim) SI 10 ms S2 Figure 9 Experimental arrangement for monitoring e x c i t a b i l i t y changes in the Schaffer collateral terminal induced by ac t ivi ty in the same and other nearby f i b r e s . The recording electrode was placed in the CA3 ce l l body layer to record all-or-none action potentials in single CA3 c e l l s . The conditioning/test stimulating electrode was placed in the CA^ apical dendritic area to determine the threshold for activation of the test Schaffer collateral and to deliver the conditioning stimulation (1, 5 or 10 pulses at 100 Hz). The threshold for the test ce l l was determined at conditioning (Sl)-test (S2) interstimulus intervals between 10-300 ms. the records at the bottom of the figure show the all-or-none action potential in the test CA 3 ce l l evoked by suprathreshold conditioning stimulation (SI, 1 pulse) followed by a test (S2) stimulation. Test threshold (S2) was determined as the amount of cur-rent required to f i r e the test cel l in 50 percent of the attempts. Negati-vity is down. - 59 -dromic. The possibi l i ty of GABAergic presynaptic inhibition being respon-sible for the post-conditioning exc i tabi l i ty changes was examined by con-ducting the experiments in the presence of 100 uM picrotoxin. The exc i tabi l i ty changes in the terminal of the experimental f ibre produced by conditioning stimulation in stratum radiatum could be brought about by an activity-induced elevation in extracellular K + content. This poss ibi l i ty was examined by raising the K + concentration in the standard bathing medium from 3.1 mM to 4.5, 6 and 12 mM. In these experiments, NaCl was reduced by the appropriate amount to eliminate the poss ibi l i ty of any osmotic changes that may occur. The media containing elevated K were each infused for 1 minute and the washout time between each dose was 20 minutes. The order of the concentrations used was from the lowest to the highest. Antidromic threshold for the fibre was determined at 30 seconds during the application of the elevated K + medium. Another explanation for the presynaptic interactions could be that released transmitter acts on the Schaffer collateral terminals to produce the observed exci tabi l i ty changes. This poss ibi l i ty was examined by using varying doses of L-glutamate (the suspected transmitter of the Schaffer collaterals) in the perfusing standard medium (0.2-8 mM). Each glutamate application was 1 min in duration and the time between each dose was 20 min. Starting with the lowest dose f i r s t , threshold was determined at 30 seconds after in i t ia t ion of each application. 3.10 Associ ative- short-term - potenti ation -(STP) and LTP-experiments These experiments were conducted in picrotoxin-containing medium to minimize inhibitory influences as well as to f a c i l i t a t e the induction of associative potentiation (Wigstrom and Gustafsson, 1983). To examine - 60 -associative conditioning effect on the population EPSP, a test bipolar stimulating electrode was positioned in stratum radiatum to evoke a "weak" population EPSP (200-600 uV) that was recorded at the apical dendritic area of CAp The conditioning bipolar stimulating electrode was placed either in another region of stratum radiatum, in the stratum oriens or in the alveus (Figure 10). "Strong" responses were evoked through the conditioning electrode: stratum radiatum EPSP was 1-3 mV; stratum oriens response con-sisted of a population EPSP (2-5 mV) that was interrupted by a population spike (0.5-1 mV); alvear stimulation resulted in an antidromic compound action potential of CA^ cells (3-7 mV). A l l of these responses were recorded through the electrode placed at the CA-^  apical dendritic region. Paired pulse f a c i l i t a t i o n to the second stimulus pulse was observed when the test "weak" EPSP was stimulated twice with an interstimulus interval of 50 ms (Figure 10, inset a j . To test for overlap between the "weak" and "strong" inputs, the "strong" input was stimulated 50 ms before the test "weak" input. If there was a f a c i l i t a t i o n of the "weak" input, i t was assumed that there was overlap whereas no f a c i l i t a t i o n or a slight depres-sion suggested no common fibres between the two inputs (Figure 10, inset Jb). The presence of picrotoxin in the bathing medium rules out the possibi-l i t y that GABA-mediated inhibition could play a role in masking or suppres-sing paired pulse f a c i l i t a t i o n (unless i t is due to activation of GABAg receptors). The conditioning stimulation consisted of 1, 5 or 10 trains (100 Hz, 10 pulses in each t ra in , one train every 5 seconds). Each condi-tioning train was either delivered alone (unpaired, with no concurrent a c t i -vation of the test input) or paired with a single stimulation of the test - 61 -S i . Alveus 1/ a. V Figure 10 Experimental arrangement for associative induction of short-term (STP) and long-term potentiation (LTP) of the CA^ population EPSP. The test popu-lation EPSP (weak response, 200-600 uV) was evoked by stimulation of stratum radiatum (S2) and recorded in the CA^ apical dendritic region. Condi-tioning stimulating (SI) electrodes which were used to evoke strong res-ponses were placed either in stratum radiatum, stratum oriens or the alveus and were used for delivering 1, 5 or 10 trains (10 pulses in each train at 100 Hz, one train every 5 seconds). The conditioning trains were either paired with one single stimulation of the test (S2) input during each train or were not paired with the test input and given alone. To establish the separateness of the test (S2) and conditioning (SI) inputs, paired pulse experiments were conducted. Inset _a shows paired pulse f a c i l i t a t i o n i f the test input (S2) is stimulated twice in succession at an interstimulus inter-val of 50 ms. Inset b_ shows a sl ight suppression of the test EPSP (S2) by a prior stimulation (preceded S2 by 50 ms) of a heterosynaptic stratum radia-tum input (SI), suggesting no overlap of the test and conditioning inputs. In a l l experiments (radiatum, oriens or alvear conditioning), the re la t ion-ship between the test and conditioning inputs were examined in a similar fashion. The two inputs were considered to be separate only i f the test EPSP was either not potentiated or s l i g h t l y depressed using the paradigm shown in inset b_. Negativity is downward in records. Calibration bars in inset b_ apply to both insets £ and b_. - 62 -input. The temporal relationship between conditioning and test inputs for the induction of associative potentiation was determined by varying the interstimulus interval between the onset of the conditioning train and a c t i -vation of the test stimulus (- 100 to + 100 ms). For these experiments, the conditioning' ' input was stratum oriens and the number of conditioning trains was fixed at f i v e . Except for the temporal relationship experiments, when the conditioning stimulation was paired with the test EPSP, the latter was always evoked at 1 ms after the onset of each conditioning t ra in . The con-trol stimulus frequency for both the conditioning and test inputs was 0.1 Hz throughout the experiment (except when conditioning was given). Each was stimulated with a single pulse through their respective electrodes and i t was arranged such that the two stimuli would alternate at 5 second intervals . To examine i f any presynaptic exc i tabi l i ty changes accompanied STP and LTP induced by the associative paradigm described above, the threshold (determined as the amount of current required to activate the cell to d i s -charge an action potential in 1-2 of 3 consecutive attempts) for antidromic activation of single CAg neurones by stimulation at the Schaffer col la te -ral terminal regions was determined before and after the unpaired and paired conditioning trains (test stimulus was given at 1 ms after onset of each conditioning train) (Figure 11). The conditioning input was either stratum radiatum or stratum oriens and i t was confirmed that the unpaired condition-ing trains did not activate the test CAg neurone. As in the experiments with the EPSP above, the conditioning stimulus consisted of 1, 5 or 10 trains (each train delivered every f ive seconds, 10 pulses at 100 Hz in each t r a i n ) . - 63 -stratum oriens rCAi neurone ^1 A stratum h \radiatum CA3 neurone Figure 11 Experimental arrangement for the associative induction of Schaffer co l la te -ral terminal exc i tabi l i ty alterations using conditioning stimulations that lead to associative STP and LTP of the population EPSP. A recording micro-electrode was placed in the CA3 c e l l body layer to monitor an all-or-none action potential in the test CA3 c e l l (see inset, negativity is down-wards). The test stimulating electrode (S2) was positioned at the terminal regions of the Schaffer collaterals and was used for determining the thres-hold for antidromic activation of the test c e l l . The conditioning bipolar electrode (SI) was placed either in stratum oriens or stratum radiatum and was used to deliver 1, 5 or 10 conditioning trains (10 pulses at 100 Hz in each t r a i n , one train every 5 seconds). Conditioning trains were either delivered unpaired (with ho concurrent activation of the -test ce l l ) or paired (at 1 ms following the onset of each train) with stimulation of the test c e l l . It was confirmed that during pairing, the stimululation used to evoke the test fibre (S2) was suprathreshold for action potential discharge. - 64 -3.11 Experiments involving interruption of input stimulation As with other experiments conducted, the control frequency of stimula-tion to evoke a population EPSP was 0.2 Hz. The stimulating electrode was placed in stratum radiatum and the recording electrode at the CA^ apical dendrites. After establishing stable controls for at least 30 minutes, the stimulation was stopped for 10 minutes and reinstated following this period. The size of the population EPSP was monitored for 30 minutes after such a treatment. To examine i f any presynaptic exc i tabi l i ty changes were associated with the alterations in the EPSP, the threshold for antidromic activation of Schaffer collateral terminals was determined before and for 30 minutes after a 10 minute "rest" period. Control threshold was stable for at least 15 minutes before the stimulation was interrupted and stimulus strength during the control and following the quiescent period was main-tained at a suprathreshold level for action potential discharge (except when threshold measurements were being taken). 3.12 Elevated K + experiments When K + was increased in the medium, the Na+ content was reduced to maintain a constant molar concentration of these two ions combined. The K + concentration was increased from 3.1 mM to 20 mM in standard medium. + The effects of a 10 minute exposure of the sl ices to high K medium on the stratum radiatum-induced EPSP in the CA^ apical dendrites and the Schaffer collateral terminal exc i tabi l i ty were examined. A potentiation of the EPSP as well as a decrease in exc i tabi l i ty of the Schaffer collateral terminals were observed following the exposure to high K + medium. These observa-tions were similar to those seen during LTP (Sastry, 1982). It was of - 65 -interest to ascertain whether this decrease in terminal exc i tabi l i ty was a + result of Na -inactivation or hyperpolarization. Therefore, during the post-application period when the Schaffer collateral antidromic threshold was elevated (at 7-10 minutes post-application), a medium containing a s l i g h t l y elevated (4.5 mM) K + concentration (which was known in previous experiments to produce a decrease in control threshold) was infused for 1 minute and threshold determined at 30 seconds during the perfusion. In subsequent experiments where elevated K + doses were applied, ++ ++ ++ picrotoxin-EDTA medium and Ca -free (Mn or Co ) picrotoxin-EDTA medium were used as control and test media, respectively. The f i r s t reason for this choice is that picrotoxin fac i l i ta tes LTP induction. If potentia-tion was present following C a + + - f r e e medium application, i t would have been more v is ible i f the magnitude of this potentiation was greater, espe-c i a l l y i f Co or Mn was not completely eliminated during the washout period. Secondly, the EDTA was added to f a c i l i t a t e the washout of C o + + and Mn . Therefore, the entire experiment was conducted with the picro-toxin-EDTA medium as the control perfusing medium. Apical dendritic EPSPs were recorded in response to stratum radiatum stimulation. Dose response curves for elevated K + concentrations were obtained in both control medium (5-80 mM K +) as well as C a + + - f r e e medium (10-80 mM K +) containing either 1 mM M n + + or 1 mM C o + + . Starting with the lowest dose, the K + applications were given in control and Ca + + - f ree media alternately and the interval between termination of one application and commencement of the next + was 30 minutes. In some experiments, for the same concentration of K , ++ the application in Ca -free medium preceded the application in control - 66 -medium and in others, the order was reversed. Two separate series of experiments were conducted for Mn + + -containing and Co + + -containing ++ + Ca - f ree media. Before each K application (whether i t was given in ++ ++ control or Ca -free medium), the appropriate Ca -free medium was infused for 4 minutes to abolish the synaptic response and to wash out the ++ + Ca present in the control medium. The elevated K medium (in control or appropriate C a + + - f r e e medium) was then applied for 3 minutes before returning to C a + + - f r e e medium for 2 minutes (to wash out excess K +) and eventually back to control medium. The Ca + + - f ree medium application + ++ before and after high K in Ca -free medium was to minimize any inter-action with the C a + + contained in control medium. The same treatment was applied to elevated K + application in control medium so that a l l factors except for the C a + + content during K + exposure would be identical . To ascertain i f exposure of the slices to Ca -free medium alone would pro-duce any post-application changes in the population EPSP, a 10 minute i n f u -++ ++ si on of this medium (containing Mn or Co in their respective experi-ments) was given and post-treatment records taken for 30 minutes. 3 3.13 H-Glutamate-binding studies Tetanic stimulations were delivered through a metal bipolar electrode to produce either a post-tetanic LTP (400 Hz, 200 pulses) or homosynaptic depression (20 Hz, 600 pulses) of the CA^ neuronal response. The stimula-tion strength for the tetanic stimulations was fixed (150 uA, 0.2 ms dura-t ion, negative pulses). In these experiments, two sets of 5 slices from the same animal were used. One set was tetanized with either the low or the high frequency stimulation and the other set, which was placed in the same - 67 -bath, were not tetanized and were u t i l i z e d as controls. The f ive slices in the experimental group were tetanized individually and the CA^ population spike was recorded in the cel l body layer of the last s l ice to be tetanized. The post-tetanic response was monitored for 10 minutes to ensure that either LTP or homosynaptic depression was present after the appropriate tetanus + 3 before the slices were removed for Na -independent H-glutamate binding determinations. An increase in these binding sites was observed at 7 and 30 minutes post-tetanus (Baudry et al . - , 1980). In experiments which involved infusion of media other than the standard medium, the control and experimen-tal sets of slices were placed in separate s l ice chambers. Cl~-free medium (all C l ~ in the medium was replaced with nitrate) was infused for 5 minutes. The CA-^  population spike in response to stratum radiatum stimu-lation was recorded in one s l i ce of the treated group prior to, during and for 10 minutes following exposure to this medium. The treated and untreated groups of slices were then removed for Na+-independent binding determina-tions. The result from each test group was paired with the result from its own control group of s l i c e s . The synaptic membrane preparation was obtained using the method of Enna and Snyder (1975). Each group of 5 slices was homogenized (using a glass homogenizer with a teflon pestle) in 1 ml of 0.32 M sucrose by hand (20 strokes). The homogenate was then centrifuged at 1000X g for 10 min-utes. The supernatant was collected, diluted f ivefold with 0.32 M sucrose and recentrifuged at 14,000X g for 20 minutes. The pellet obtained (P^ fraction) was resuspended in 2 ml of cold d i s t i l l e d water and allowed to incubate for 40 minutes. This suspension was then centrifuged at 7000X g - 68 -for 20 minutes after which the supernatant was discarded and the pellet washed gently with cold d i s t i l l e d water in order to collect only the upper layer. This suspension was centrifuged at 40,000X g for 30 minutes and the supernatant was discarded. The pellet was resuspended by sonication in cold d i s t i l l e d water and recentrifuged at 40,000X g for 30 minutes. The f inal pellet was resuspended in 50 mM Tris-HCl buffer (pH 7.4) by sonication (about 0.5-1.0 mg protein/ml). + 3 The Na -independent H-glutamate binding assay was conducted in t r i p l i c a t e . Aliquots of the above synaptic membrane preparation (0.1 ml) were preincubated at 30°C for 10 minutes. Following this period, 0.1 ml of 3 L- H-glutamate (Amersham, 35 Ci/mmol, f inal concentration 100 nM) was added. After f if teen minutes, the binding was terminated by addition of 3 ml cold buffer (50 mM T r i s - H C l , pH 7.4). Each suspension was f i l t e r e d under vacuum suction through a millipore f i l t e r (0.45 urn pore s ize ) . The tube was rinsed with 3 ml cold buffer which was also poured over the f i l t e r . The f i l t e r s were f i n a l l y washed with 4 ml cold Tris-HCl buffer and allowed to drain with suction on for 5 minutes. The f i l t r a t i o n procedure lasted less than 15 seconds and i t was ensured in each experiment that the time taken to f i l t e r both the test sample and its paired control were comparable. Blanks consisted of 0.1 ml Tris-HCl buffer (50 mM, pH 7.4) without tissue and were treated in a similar fashion as the synaptic membrane preparation. Each f i l t e r was placed into a s c i n t i l l a t i o n vial containing 10 ml of aqueous s c i n t i l l a t i o n cocktail (ACS, Amersham). The tritium content of each sample was counted at 30% efficiency with a Packard l iquid s c i n t i l l a t i o n counter. According to Lynch et- a l . (1982), i f the above assay was conducted in the - 69 -presence of 100 uM cold glutamate, the binding was not greater than that obtained for the blanks. Therefore, in this study, as with theirs , specific binding was determined as total counts for each sample minus blanks. 3 3.14 H-Glutamate accumulation studies + 3 As with the Na -independent H-glutamate binding studies, two sets of 5 slices were used for each experiment and LTP or homosynaptic depression was produced using either a 400 Hz, 200 pulses or 20 Hz, 600 pulses tetanic stimulation, respectively, of stratum radiatum (150 uA, 0.2 ms duration, negative pulses). In some experiments, the Ca++-dependence of tetanus-induced uptake changes was examined by delivering either the low or high frequency tetanus to each of the 5 experimental sl ices between 5 to 10 minutes of a 10 minute infusion of Ca -free medium (contains 1 mM Mn , 3 mM Mg ). At the same time, in a separate s l ice chamber, the same tetanus was delivered to each s l i ce in the control group that was being perfused with standard Ca + + -containing medium. Both sets of sl ices were 3 removed simultaneously for H-glutamate uptake measurements 10 minutes following tetanus of the last s l i c e . Data from each test group was paired with results from its own control group for analysis. Immediately upon their removal from the s l ice chamber, each group of 5 slices was incubated in standard medium (which has 147.3 mM Na+) contain-3 ing H-glutamate (Amersham, 35 Ci/mmol, f inal concentration 1 uM) to 3 determine H-glutamate accumulation into whole slices (which was presumed to be due to uptake). Each group of sl ices was incubated for 10 minutes at 32°C with 0.5 ml of the above radioactive solution. Then, the slices were washed with 10 ml cold (4°C) buffer (50 mM T r i s - H C l , pH 7.4) under suction - 70 -through a nitrocellulose f i l t e r (Mill ipore , pore size 0.45 urn). The vacuum was kept on for 2 minutes following the wash to eliminate excess l i q u i d . The sl ices were then homogenized by hand with a glass-teflon homogenizer (20 strokes). An aliquot (0.1 ml) of the homogenate was added to a vial con-taining 10 ml aqueous s c i n t i l l a t i o n cocktail (ACS, Amersham) to be counted by the l iquid s c i n t i l l a t i o n counter. 3.15 Protein determination + Protein analysis was conducted for both the Na -independent binding as well as uptake studies using the method of Lowry et a l . (1951) with bovine serum albumin as the standard. A l l samples were analyzed in t r i p l i -cate and a standard curve was constructed for each experiment. 100 ul each of standard stock solutions containing 20, 40, 60, 80 and 100 ug protein/100 ml as well as 10 ul aliquots of the samples to be analyzed were added to separate test tubes kept on ice . The volume in each test tube was made up to 0.5 ml with d i s t i l l e d water and blanks consisted of 0.5 ml d i s t i l l e d water. 5 ml of Lowry's Solution A (2% Na 2 C0 3 , 0.02% KNa tartrate, 0.4% NaOH) containing 1% CuSO^ was added to each tube. A l l samples were vortexed and lef t to stand at room temperature for 10 minutes. Following this incubation, 0.5 ml Folin-Ciocalteu phenol reagent (Sigma Chemicals) which was diluted 1:1 with d i s t i l l e d water was added. Tubes were vortexed again and the reaction was allowed to proceed for an additional 25 minutes. The protein content for the standard curve and test samples were determined by measuring the absorption of each solution at 750 nm using a Beckman ACTA C2 spectrophotometer. The standard curve of absorption vs concentration of protein was constructed and the unknown protein content in each sample - 71 -extrapolated from its absorption. Since only 10 ul of the sample was used (so that the concentration would f a l l within the range of the standard curve) and 100 pi of each of the standards was used for determining the curve, the actual protein concentration for the test sample expressed as mg protein/ml wi l l be 10X the value obtained from reading off the curve. 3 Radioactivity due to H-glutamate in the binding and uptake studies was corrected appropriately for protein content in the samples. 4 RESULTS 4.1 Effects of DLH application at the CAg c e l l body on Schaffer c o l l a- teral terminal exci tabi l i ty DLH application on CAg cel l bodies (20-100 nA, 3 min) resulted in an increase in f i r i n g rate of these neurones (up to 100 Hz). The increase in action potential discharge was seen both in the presence (n = 19) as well as absence (n = 9) of C a + + (1 mM M n + + , 3 mM Mg + +) in the bathing medium. The rate of cel l f i r i n g promptly returned to control levels following termi-nation of the DLH application. Subsequent to DLH application (the agent was applied during perfusion with standard medium), the response could not be recorded in the majority of cases (14 of 19 cel ls) for about 2 minutes post-treatment. Perhaps, prolonged DLH application produced shunting of the spike at the CAg cel l somata. Upon recovery, the threshold for spike generation from the Schaffer collateral terminals was increased as compared to control . In most cases, the i n i t i a l increase in the threshold f e l l s l i g h t l y at about 7 to 8 minutes post-treatment (but was s t i l l at a level - 72 -higher than control) . At roughly 10-12 minutes post-drug, the threshold started to increase again and maintained at a new level for at least 15 minutes (Figure 12, Table 1). The antidromic threshold was unaltered during or following an applica-++ ++ tion of Ca -free (Mn -containing) medium (Table 1). As with the case of DLH application during application of standard medium, the same applica-++ tion in Ca -free medium resulted in a 2 minute post-treatment suppression of the response. However, the prolonged increase in threshold produced by application in Ca + + -containing medium was not observed. 4.2 Effects of- bath- application of NMDLA on the CA-^  population- spike and  Schaffer collateral terminal exc i tabi l i ty NMDLA (100 uM, 2 min) application to the whole bath resulted in a total abolition of the CA-^  population spike during and immediately follow-ing the drug perfusion. This depression was succeeded by a potentiation of the response 5-7 minutes after termination of NMDLA application (population spike as a % of pre-drug control: 200 ± 15 SEM at 20 min post-NMDLA, n = 8) (Figure 13A). The CA^ antidromic spike resulting from Schaffer collateral terminal region stimulation could not be evoked (presumably because the action potential in the CA^ neuronal soma-dendritic region was shunted) using 5 to 10 times control threshold during and for 5 to 7 minutes follow-ing the same NMDLA application. Subsequent to this period of refrac tor i -ness, the threshold was sustained at a level above control (threshold as a % of pre-drug control: 151 ± 5 SEM at 20 min post-NMDLA, n = 8) (Figure 13j$). The size of the population spike as well as the CA^ antidromic threshold increase usually returned to control levels in 30-40 minutes. The - 73 -I » • • • • • • • 0 10 20 30 40 50 60 70 80 MIN Figure 12 Effect of DL-homocysteate (DLH, 100 nA, 3 min) on the threshold for a n t i -dromic activation of a CA3 neurone, in the presence and absence of extra-cel lular C a + + . The Ca -free (contained 1 mM M n + + , 3 mM Mg + + ) medium was applied for 5 minutes and DLH was iontophoresed during the last 3 minutes of the application. Inset A shows the arrangement of the electrodes for stimulation, recording and iontophoresis. Inset IB shows the antidromic response from a single CA3 neurone. Note the increase in threshold after application of DLH in standard (2 mM C a + + , 2 mM Mg + + ) medium and no change following application in C a + + - f r e e medium. Threshold was measured using 0.2 ms negative pulses (evoked at 0.2 Hz, used a constant current unit) by increasing the stimulation until the cel l responded with an a l l - o r -none action potential in about 50 percent of the t r i a l s . Tabic 1. A. Effects of DLH applied on CA^ neuronal somata on the threshold for antidromic activation of the neurone at Schaffer collaterals in the CAib area. 15 min post-DLH 30 min post-DLH {% of control threshold) (% of control threshold) DLH applied in 143.79 * 3.50 (n = 19) a 143.84 * 3.49 (n - 19) a normal medium Range: 115 - 165 Range: 117 -167 DLH applied in 101.22 * 1.61 (n . 9 ) b 100.22 ± 1.13 (n = 9 ) b Ca + +-free medium Range: 96-110 Range: 97 - 108 (1.0 mM Mn++) B. Effect of exposure to Ca + +-free (1.0 mM Mn++) medium for 5 min on the threshold. 2 min in Ca - free (DLH was applied later) (% of control threshold) 4 min in Ca -free (DLH was not applied) {% of control threshold) 15 min post Ca -free (DLH was not applied) [% of control threshold) 98.77 * 2.38 (n = 9 ) b 99.60 ± 1.92 '(n = 10) c d99.30 * 1.66 (n . 10) C Range: 93 - 110 Range: 91 - 110 Range: 93 - 108 aBoth sets of data were ments; cboth sets were text for more details. from the same 19 experiments; b a l 1 from the same 10 experiments; dall data were from the same 9 experi-values given are mean * SEM. See - 75 -Figure 13 Ef fects of N-methyl-DL-aspartate (NMDLA, 100 uM, 2 min) applied to the bath on the CAi population spike and the threshold for antidromic ac t iva t ion of s ingle CA3 c e l l s from the terminal regions of Schaffer c o l l a t e r a l s . The potent iat ing action of NMDLA on the population spike i s shown in A. Insets show a control CA^ population spike on the l e f t and a potentiate"? response at 20 minutes post-NMDLA appl ica t ion on the r i g h t . C a l i b r a t i o n bars in the r i g h t record apply to both responses. An increase in the antidromic thres -hold of a s ingle Schaffer c o l l a t e r a l terminal i s seen fo l lowing termination of the above appl ica t ion {B). The upward arrows indicate the commencement of NMDLA appl icat ion and the downward arrows show time of termination of the drug. Results shown in A and B are from two separate experiments. SCH ANTIDROMIC THRESHOLD CD o/ > AS % OF CONTROL * CA* POPULATION SPIKE AS /o OF CONTROL - 77 -NMDLA dose of 100 pM was chosen after doing dose-response studies (25-400 pM) and i t seemed to be the concentration that produced the most consistent and re l iable results , although similar observations were made with the other doses. It has to be borne in mind that the flow rate of the medium, the concentration of NMDLA and the time of exposure of the slices to the drug are c r i t i c a l for observation of a potentiation of the population spike. The depression of the response without any subsequent potentiation is more f r e -quently produced by application of this amino acid and i t i s , therefore, important to t i t rate the dose and duration of application for each i n d i v i -dual experimental arrangement. Some experiments were done using 50 pM NMDA or 50 pM NMLA instead of NMDLA. The results for the effects of NMDA on the population spike (212 ± 19% SEM of pre-drug control at 20 min post-NMDA, n = 8) and Schaffer collateral threshold change (158 ± 7% SEM of pre-drug con-trol at 20 min post-NMDA, n = 8) were comparable to those obtained using NMDLA. However, NMLA had no significant effect on the CA^ population spike at this concentration (n = 4). 4.3 Effects-of bath-application of NMDLA on the GA^ population spike As in the case of the Schaffer collateral-CA^ population spike, an application of NMDLA (100 pM, 2 min) to the bath resulted in a complete suppression of the mossy fibre-CA^ population spike during and for 5 to 7 minutes following drug application. Following this period of ref rac tor i -ness, the population spike recovered to a magnitude below that of pre-drug control (population spike as a % of pre-drug control: 77 ± 10 SEM at 20 min post-NMDLA, n = 10) (Figure 14). In a few cases (2 of 10 expts.) , there was a very brief potentiation of the response at 4-5 min post-drug but this - 78 -§ 120 o o o < oo < 100 80 S 60 a . oo o t—t I— < 40 I 20 C O o NMDLA 100 yM J_ 10 15 20 TIME (min) 25 30 35 Figure 14 Ef fects of bath appl i ca t ion of NMDLA (100 pM, 2 min) on the mossy f i b r e -CA3 population spike . Horizontal bar above the graph shows duration of NMDLA a p p l i c a t i o n . Each point on graph represents mean ± SEM (n =10) . - 79 -increase quickly subsided in 2-3 min and the response was maintained at a depressed level thereafter. 4.4 Effects of bath -application of NMDLA in the- presence of Ca + + - f ree  medium on the -GA^ population spike and Schaffer collateral - terminal -exci- t a b i l i t y ++ If NMDLA was applied during an exposure of the s l i ce to a Ca -free ++ ++ (1 mM Mn , 3 mM Mg ) medium, the potentiation could not be induced (population spike as a % of pre-drug control: 98 ± 5 SEM at 20 min post-NMDLA, n = 8) (Figure 15A_). Since the induction of the potentiation of the population spike is blocked by C a + + - f r e e medium, the associated decrease in exc i tabi l i ty of Schaffer collateral terminals, i f i t plays a role in this potentiation, should also be counteracted by this treatment. Consistent with this idea, the application of NMDLA during infusion with Ca + + - f ree medium abolishes the increase in Schaffer collateral terminal threshold (threshold as a % of pre-drug control: 100 ± 4 SEM at 20 min post-NMDLA, n = 9) (Figure 15B). 4.5 Effects of bath application of NMDLA in -the presence of- APV- on- the  CA-^  population spike and Schaffer col lateral terminal excitabil i ty When NMDLA (100 uM, 2 min) was applied in the presence of APV (100 uM, 4 min) in the bathing medium, neither the i n i t i a l depression (population spike as a % of pre-drug control: 89 ± 6 SEM at 3 min post-NMDLA, n = 8) nor the potentiation (population spike as a % of pre-drug control: 102 ± 5 SEM at 20 min post-NMDLA, n = 8) of the population spike produced by NMDLA could be observed (Figure 16A). The same treatment also prevented the increase in the Schaffer collateral threshold increase induced by NMDLA - 80 -Figure 15 Blockade of the e f fects of NMDLA (100 uM, 2 min) on the CAi population spike and the threshold for antidromic ac t iva t ion of a s ingle CA3 c e l l from the terminal regions of Schaffer c o l l a t e r a l s by i t s appl icat ion in the presence of C a + + - f r e e medium ( M n + + 1 mM, M g + + 3 mM). The induction of both the potentiat ion of the population spike (A_) as well as the increase in Schaffer c o l l a t e r a l antidromic threshold (B) are blocked. The upward arrows indicate the commencement of NMDLA appl icat ion and the downward arrows show time of termination of the drug. Horizontal bars above graphs show duration of appl icat ion of Ca - f r e e medium. Results shown in _A and B are from two separate experiments. SCH ANTIDROMIC THRESHOLD CO AS % OF CONTROL CA 1 POPULATION SPIKE AS 1 % OF CONTROL 00 o f -o —r~ O I \ - 82 -Figure 16 Blockade of the ef fects of NMDLA (100 uM, 2 min) on the CA^ population spike and the Schaffer c o l l a t e r a l antidromic threshold by concurrently administered APV (100 yM, 4 min). The resul ts for the CAj population spike are shown in /\ and the resu l t s for the Schaffer c o l l a t e r a l antidromic threshold in B. T f ^ upward arrows indicate the commencement of NMDLA appl ica t ion and- the downward arrows show time of termination of the drug. Horizontal bars above graphs show duration of appl icat ion of APV. Results shown in A and B are from two separate experiments. SCH ANTIDROMIC THRESHOLD CD AS % OF CONTROL CA« POPULATION SPIKE AS > 1 % OF CONTROL 09 O f-o o T |S3 O "I I m T 5' in / \ \ i C O C O - 84 -(threshold as a % of pre-drug control: 98 ± 5 SEM at 20 min post-NMDLA, n = 8) (Figure 16B). 4.6 Effects - of - bath application of APV on- tetanus-induced LTP- of- the -CA^ population spike When a tetanic stimulation (400 Hz, 200 pulses) was delivered to the stratum radiatum input during the last 30 seconds of APV application (100 uM, 4 min) to the bath, LTP could not be induced (population spike as a % of control: 88 ± 6 SEM at 20 min post-tetanus, n = 8). However, when the same tetanic stimulation was given in standard medium without APV 30 minutes after the f i r s t tetanus, LTP could be re l iably e l i c i t e d (population spike as a % of control: 183 ± 10 SEM at 20 min post-tetanus, n = 8) (Figure 17A). Lower doses of APV (25 uM, 4 min) were also effective in blocking the induc-tion of LTP (n = 4). It should be noted that APV (25 or 100 yM) i t s e l f frequently produced a sl ight suppression of the CA-^  population spike (82 ± 3% SEM of control at 1 min in APV, 7 of 10 expts.) and in 4 of 10 experi-ments produced a post-application potentiation of the response (125 ± 4 % SEM of control at 10 minutes post-APV). 4.7 Effects - of bath- application of - APV- on- -tetanus-induced - LTP of- the CAg population spike In contrast to the results obtained in the CA^ region, APV had no significant effect on either the control CAg population spike or the induction of LTP in this area. Following a tetanic stimulation (400 Hz, 200 pulses) to the mossy f ibre input during infusion with APV (100 uM, 4 min), LTP could be induced (population spike as a % of control: 164 ± 7 SEM at 20 min post-tetanus, n = 8) (Figure 17BJ. The magnitude of LTP obtained for - 85 -Figure 17 Ef fec ts of APV (100 uM, 4 min) on the induction of LTP i n the stratum radiatum-CAi and mossy fibre-CA3 population sp ikes . The resul ts for the CAj population spike are shown in A and those for the CA3 population spike are shown in B. The arrows in the graphs indicate times at which te tanic st imulations (400 Hz, 200 pulses) were del ivered to the stratum radiatum (A) or mossy f i b r e (B) inputs . Horizontal bars above the graphs show duration of appl ica t ion o7 APV. The r e s u l t s shown are from two sepa-rate experiments. CA 3 POPULATION SPIKE AS % OF CONTROL CD CA 1 POPULATION SPIKE AS % OF CONTROL > Ul o ~r o © in o 3 5 © s / I •I \ / I / / O Ol o on o o / \ C O cn - 87 -the mossy f ibre -CA 3 system in the absence of APV treatment in a separate set of experiments was not s ignif icantly different (p > 0.2) from that obtained when the tetanus was given in the presence of APV (population spike as a %of control : 161 ± 10 SEM at 20 min post-tetanus, n = 8). 4.8 Iontophoretic applications of NMDLA at the GA-^  apical -dendritic zone Since an application of NMDLA to the bath produced a potentiation of the Schaffer collateral-CA^ population spike, i t was of interest to exa-mine i f this potentiation could be a result of changes at the synaptic leve l . Furthermore, an application of NMDLA to the bath could be mimicking a tetanic stimulation to stratum radiatum by inducing an increase in the f i r i n g rate of CA^ c e l l s . Therefore, i t was decided to apply NMDLA l o c a l -ly to the apical dendritic area of CA-^. An application of NMDLA (100 pM, 1-2 min) at the apical dendritic zone of CA-^  produced a rapid depression of the population spike during i ts iontophoresis (population spike as a % of control: 10 ± 4 SEM at 3 min post-NMDLA, n = 8). The response recovered over 5-6 minutes following termination of the drug. After this period of recovery, there was a potentiation of the population spike (160 ± 4% SEM of pre-drug control at 20 min post-NMDLA, n = 6) which gradually recovered to control size in 30-40 minutes. These findings are in support of those reported earl ier by Collingridge et al.- (1983b). Low doses (10-40 nA) of NMDLA at the synaptic zone were sufficient to produce the i n i t i a l depression without any subsequent potentiation (n = 8) but higher doses (100 nA) were required to e l i c i t the potentiation of the population spike (n = 6). When the effects of iontophoretic application at the CA-^  apical dendritic/ synaptic zone were examined on the threshold for antidromic activation of - 88 -single Schaffer collateral terminals, i t was found that NMDLA (100 nA, 1-2 min) induced a slight drop (10-15%) in threshold during the drug ejection for a brief period of about 10 seconds but was succeeded by a long-lasting post-application elevation of the threshold (146 ± 6 % SEM of pre-drug con-trol at 20 min post-NMDLA, n = 8) (Figure 18). As in the case of inducing potentiation of the response by the NMDLA application in the synaptic zone, a low dose of the amino acid (10-50 nA) fai led to re l iably induce the pre-synaptic change whereas a higher dose (100 nA) was more consistent in pro-ducing this effect . At no time during or after the iontophoretic applica-tion of NMDLA on the Schaffer collateral terminals did the CA 3 ce l l become unevokable by stimulation at the terminals. This observation suggests that NMDLA perfusion, which produces a period of refractoriness of the CA3 ce l l during and for 5-7 minutes following the application, probably does not act on the terminals to render the cel l unevokable but, rather, may be causing some change in the CA^ cel l body so that i t cannot sustain an action potential . 4.9 Interactions - among presynaptic - terminal s in the GA-^  - region 4.9.1 Conditioning through a separate electrode: The experimental arrangement is shown in Table 2A. A conditioning tetanus (SI) of 5 pulses at 100 Hz resulted, in the majority of cases, in a decrease in the test Schaffer collateral terminal threshold (S2) for 10-300 ms post-conditioning (Table 213). An increase in the conditioned threshold was observed in one experiment at the interstimulus delay of 300 ms (179 % of unconditioned con-t r o l , 1 of 13 s l i c e s ) . As the intensity of the conditioning (Si) stimula-tion was increased, the conditioned threshold (S2) of the f ibre was de-- 89 -140 7* 130 So U J OC I Z 120 I* 110 100 < o 90 L R-recording electrode S-stimuTating electrode B > E 10 ms 10 20 30 40 50 60 T I M E (m in) Figure 18 NMDLA-induced increase in the threshold for activation of a single Schaffer collateral from its terminal regions. NMDLA (100 nA, 1 min) was applied iontophoretically at the apical dendritic/synaptic area of CA^. The upward arrow indicates commencement and the downward arrow shows termination of the ejection. Inset A i l lustrates the experimental arrangement. Inset j3 is a record of the all-or-none action potential recorded from one ce l l in the CA3 c e l l body layer evoked by stimulation at the terminal regions of the Schaffer col la terals . Negativity is down. Table 2. E F F E C T S OF VARIOUS CONDITIONING S T I M U L A T I O N S ON THE THRESHOLD FOR A C T I V A T I O N OF S I N G L E S C H A F F E R C O L L A T E R A L TERMINALS S 1 - S 2 t le lay (ms) 10 50 100 300 S 1 - S 2 D e l a y (ms) 10 50 100 300 l / e c r e a s e In S2 Range 72-95 77-94 8 1 - 9 9 91-102 E . D e c r e a s e In S2 Range 6 5 - 9 9 73 -93 79-101 92-105 t h r e s h o l d (X o f Mean*SEM 8 7 . 5 * 2 . 0 * 8 5 . 2 * 1 . 5 * 9 0 . 4 * 1 . 5 * 9 7 . 8 * 1 . 1 t h r e s h o l d X o f Mean*SEM 8 5 . 0 * 1 . 6 * 8 5 . 1 * 0 . 9 * 9 0 . 7 * 0 . 9 * 9 9 . 2 * 0 . 5 u n c o n d . c o n t r o l ) n 13 14 14 12 u n c o n d . c o n t r o l ) n 27 29 29 25 SI I n t e n s i t y ( M A ) 60 80 100 120 SI I n t e n s i t y (I o f c o n t r o l t h r e s h o l d ) 90 110 400 600 S2 t h r e s h o l d a t F . S2 t h r e s h o l d at S1-S2 d e l a y of Range 6 6 - 9 8 50 ms ( I o f Mean*SEM 9 3 . 7 * 2 . 1 * 81 -98 9 1 . 4 * 2 . 3 * 77 -95 8 7 . 5 * 2 . 7 * 78-91 8 4 . 0 * 2 . 4 * * S 1 - S 2 d e l a y o f Range 50 ms (X o f Mean*SEM 96-101 9 8 . 6 * 0 . 4 75-96 67-92 8 6 . 7 * 1 . 4 * b 8 0 . 4 * 2 . 0 * 69-89 7 6 . 7 * 1 . 8 u n c o n d . c o n t r o l ) n 6 7 6 7 u n c o n d . c o n t r o l ) n 17 17 15 15 b A . E x p e r i m e n t a l s e t - u p to I l l u s t r a t e p o s i t i o n i n g o f s t i m u l a t i n g and r e c o r d i n g e l e c t r o d e s f o r t h e d a t a shown In B and C . C o n d i t i o n i n g (S I ) s t i m u l u s c o n s i s t e d of 5 p u l s e s at 100 H z . B . The t h r e s h o l d f o r a n t i d r o m i c a c t i v a t i o n o f a s i n g l e S c h a f f e r c o l l a t e r a l i n r e s p o n s e to the t e s t (S2) s t i m u l u s was d e t e r m i n e d at c o n d i t f o n i n g I n t e r v a l s r a n g i n g f rom 10-300 m s . C . The s t i m u l a t i o n s t r e n g t h o f SI was p r o g r e s s i v e l y i n c r e a s e d and S2 t h r e s h o l d was d e t e r m i n e d at a c o n d i t i o n i n g I n t e r v a l o f 50 ms . 0 . l E x p e r i m e n t a l s e t - u p f o r t h e d a t a shown i n E and f . C o n d i t i o n i n g (SI ) s t i m u l u s c o n s i s t e d of 1 p u l s e . E_. S2 t h r e s h o l d was d e t e r m i n e d a t I n t e r s t i m u l u s i n t e r v a l s r a n g i n g f r o m 10-300 ms f o l l o w i n g a s u p r a t h r e s h o l d ( f o r s p i k e g e n e r a t i o n ) SI s t i m u l u s . F_. As In C a b o v e . * p < 0.01 when compared t o u n c o n d i t i o n e d c o n t r o l u s i n g p a i r e d t - t e s t . 8 The d i f f e r e n c e between t h e two s e t s o f d a t a d e n o t e d by t h i s l e t t e r Is s i g n i f i c a n t t o p < 0 . 0 2 . 0 The d i f f e r e n c e between t h e two s e t s o f d a t a d e n o t e d by t h i s l e t t e r i s s i g n i f i c a n t to p < 0 . 0 1 . - 91 -creased in a graded manner (Table 2G_). In contrast, when the number of pulses in the conditioning train was increased from 5 to 10 (frequency of the train was kept constant at 100 Hz), there were more instances where the S2 threshold was increased instead of decreased (244 ± 31 % SEM of control at 50 ms interstimulus interval , 3 of 7 expts; 230 ± 17% SEM of control at 300 ms interstimulus interval , 5 of 7 expts.) . The results are summarized in Table 3. To examine the Ca++-dependence of the conditioning effect on the ++ ++ ++ test threshold, Ca -free medium (containing 1 mM Mn , 3 mM Mg ) was infused for 10 minutes before repeating the experiment at an interstimulus interval of 50 ms. The results show that the decrease in conditioned (S2) threshold produced by the conditioning (Si) was counteracted (Table 4). 4.9.2 Conditioning through- the test- electrode: The experimental arrangement is shown in Table 2JJ. The conditioning consisted of one pulse that was suprathreshold for action potential discharge in the test f ibre (except in experiments where the number of conditioning pulses was i n -creased). Essentially, the results obtained for this series of experiments were similar to those using a separate electrode for conditioning. The con-ditioned threshold (S2) following the conditioning stimulus (SI) was usually decreased for 10-300 ms (Table 2E) but in two instances exhibited increases at the interstimulus interval of 300 ms (172-288% of unconditioned control, 2 of 27 s l i c e s ) . When the strength of the conditioning stimulus was i n -creased, there was a graded decrease in the conditioned threshold (Table 2F). When the number of conditioning pulses was increased to 10 (given at a frequency of 100 Hz), however, there was a tendency towards an increase rather than a decrease in threshold (210 ± 16% SEM of unconditioned control - 92 -Table 3 . Effect of Increasing the Number of Pulses in Each Condit-ioning Train on Schaffer Collateral Terminal E x c i t a b i l i t y .  SI = f i v e or ten pulses at 100 Hz Number of condit ioning (SI) pulses at 100 Hz 5 10 S2 threshold at S1-S2 188-294 none delay of 50 ms & of unconditioned 3 of 7 expts . control ) S2 threshold at S1-S2 delay of 300 ms (% of unconditioned control ) 179 1 of 12 expts. 183-278 5 of 7 expts . - 93 -Table 4. Ef fec t of Ca -Free Medium Exposure on the Condit-ioned Threshold. S1-S2 Delay (ms) 50 Range 73-95 S2 threshold Mean ± SEM 84 .7±1 .8* (% of control) n 8 10 min i n Ca-free Range 87-102 medium (with ImM Mn, Mean * SEM 98.8*2.0* 3mM Mg)(% of control) n 8 * The dif ference between the two sets of data denoted by t h i s symbol is s i g n i f i c a n t to p < 0 .01 . - 94 -at 50 ms conditioning interval , 7 of 8 s l i c e s ) . The conditioning effect on the Schaffer collateral threshold was also determined to be Ca+ +-dependent because the decrease in conditioned threshold (S2) was counteracted in C a + + - f r e e medium (98 ± 2 % SEM of unconditioned control at 50 ms condition-ing interval , n = 10). Upon s ta t is t ica l analysis using the paired t-test , results obtained in C a + + - f r e e medium were not s ignif icant ly different from unconditioned controls (p > 0.2) but were significant (p < 0.01) when com-pared to corresponding conditioned thresholds (at 50 ms interstimulus inter-val) of the same fibres when the experiment was conducted in standard C a + + -containing medium. To examine i f GABAergic presynaptic inhibition plays a role in the presynaptic interactions, the experiment was repeated in the pre-sence of 100 uM picrotoxin (added to standard medium). The results suggest that conventional presynaptic inhibition is not involved (Figure 19, n = 4). 4.10 Elevated - extrace-l-luTar- -K +--and - g-ltrtamate- on - -the • Schaffer- -collateral  antidromic-threshold In most cases, the threshold for the test f ibre was decreased with the lowest concentration (4.5 mM) of elevated K + (76 ± 4% SEM of control at 30 seconds during K + application, 5 of 6 expts.) and was increased in al l experiments at the highest concentration (12 mM) (242 ± 28% SEM of control + at 30 seconds during K application, 7 of 7 expts.) . At the intermediate dose (6 mM), however, both effects could be seen (81 ± 7% SEM of control at 30 seconds during K + application, 2 of 7 expts.; 207 ± 30% SEM of control at 30 seconds during K + application, 5 of 7 expts.; when al l the results obtained for 6 mM K + application were pooled, the value was not s ignif icant ly different from control) . These results are summarized in Table 5. The lower concentrations of glutamate (0.2-0.5 mM) - 95 -110 • O r-Z o 0 100 200 INTERSTIMULUS INTERVAL (ms) Figure 19 Fa i lure of p icrotoxin (100 uM) to counteract the increase in Schaffer c o l l a t e r a l terminal e x c i t a b i l i t y produced by paired pulse s t imula t ion . The condit ioning stimulus pulse was always suprathreshold for action potent ial discharge in the test c e l l . The test threshold was measured between 5-200 ms fol lowing the condit ioning pulse . The curve with f i l l e d c i r c l e s i l l u s -trates resul ts obtained for the experiment conducted in standard medium and the graph with f i l l e d diamonds represents the resul t s of the same experiment repeated in the presence of 100 uM picrotoxin in standard medium. Each point on graph i s mean ± SEM (n = 4 ) . Table 5. Effects of Raising Extracellular K + Concentration on the Schaffer Col lateral ' Terminal Exci tabi 1 i ty.  St1m. Concentration of K + in medium 4.5 6 12 • ' to I Decrease in threshold 6 4 - 8 9 7 4 - 8 8 (Xof pre-treatment control) 5 of 6 expts. 2 of 7 expts. Increase in threshold (% of pre-treatment control) 122 -315 5 of 7 expts. 155 - 400 7 of 7 expts. - 97 -produced no change whereas higher concentrations used (1-6 mM) resulted in decreases in threshold (Table 6). Doses higher than 6 mM could not be used for bath application because of shunting of the spike at the CAg cel l body. When this occurred, the response could not be recorded. To overcome this problem, glutamate was iontophoresed at the stimulating site (200 nA, 3 min). Results obtained were similar to those of bath application (threshold as a % of pre-drug control: 70 ± 4 SEM at 1 min during glutamate applica-t ion, n = 19; 58 ± 4 SEM at 3 min during glutamate application, n = 19). With the bath application of glutamate, i t appears that the f i r i n g rate of CAg cel ls is increased at concentrations of 200 uM (n = 6) whereas in most cases, the Schaffer collateral threshold exc i tabi l i ty is not affected until the concentration attains a minimum of 1 mM (n = 12). 4.11 Assoeiative-short-term-Potentiation (STP) and-LTP 4.11.1 Stratum- radiatum conditioning. Conditioning trains delivered to stratum radiatum that were not paired with the test EPSP (1, 5 or 10 trains, 10 pulses at 100 Hz in each t ra in , one train every 5 seconds) often resulted in a depression of the test EPSP (85 ± 4 % SEM of control at 60 seconds post-10 train unpaired tetanus of radiatum, 7 of 8 expts.) (Figure 20A). However, i f the test EPSP was evoked once (at 1 ms following the onset of conditioning train) during each conditioning t ra in , a short-term potentiation (STP, population EPSP as a % of control at 60 seconds post-10 train paired conditioning by radiatum: 184 ± 8 SEM, 6 of 8 expts.) followed by LTP (population EPSP as a % of control at 15 min post-10 train paired conditioning by radiatum: 162 ± 5 SEM, 6 of 8 expts.) of the test response could be observed. Usually, with 1 to 5 paired trains , STP, which lasted 2 - 98 -Table 6. Effects of Bath Application of Glutamate on Schaffer Collateral Terminal E x c i t a b i l i t y . Glutamate concentration 0.5 mM 1.0 mM 2.0 mM 4.0 mM 6.0 mM Schaffer collateral threshold as a % of pre-drug control 99 ± 1.5 96 ± 0.8 94 ± 1.5 90 ± 2.2 89 ± 2.3 n 3 12 9 6 3 P NS < 0.01 < 0.01 < 0.01 < 0.05 *Values are expressed as mean ± SEM. - 99 -Figure 20 Associative induction of STP, LTP and the reduction in the Schaffer c o l l a -teral terminal e x c i t a b i l i t y . (A_) The schematic diagram on the lef t i l l u s -trates the experimental arrangement. A bipolar test stimulating electrode (S2) was positioned in the stratum radiatum and a bipolar conditioning stimulating electrode (Si) was positioned in another area of the stratum radiatum. A recording microelectrode (containing 4 M NaCl) was positioned in the apical dendritic area of CAi neurones to monitor the test EPSP evoked at 0.2 Hz (stimulation strength was adjusted to obtain a response between 300-600 uV). The conditioning stimulation strength was adjusted to evoke a population EPSP of 1-3 mV in s ize . If a twin stimulation of So (50 ms interval) resulted in a f a c i l i t a t i o n of the second population EPSP (see inset, left ) and i f a stimulation of Si preceding S2 stimulation by 50 ms resulted in no f a c i l i t a t i o n of the second population EPSP (see inset , r ight) , then the Si and $2 stimulations were presumed to activate sepa-rate input f ibres . In a l l experiments, the effect of unpaired conditioning trains (UC; i . e . , the test stimulation was off during the conditioning; each conditioning train contained 10 pulses at 100 Hz) and of paired conditioning trains (PC; i . e . , the test stimulation was on 1 ms after the onset of each train) were examined on the test population EPSP. During the f i r s t 3 min-utes after UC or PC, the response was monitored every 15 seconds and at a l l other times at 30 second intervals . The graph on the right shows results from one experiment. Note STP after 1 and 5 PCs, and LTP after 10 PCs. (B) Effects of the conditioning on the e x c i t a b i l i t y of the terminal region of a Schaffer c o l l a t e r a l . A monopolar test stimulating electrode (S 2) was positioned in the apical dendritic area of the CAi neurones to activate (0.2 ms negative pulses, 3-10 yA, 0.2 Hz) the terminal regions of Schaffer collaterals so that antidromic all-or-none action potentials (see inset) could be recorded from the CA3 c e l l bodies. A conditioning stimu-lation electrode (S^) was positioned in the stratum radiatum and the unpaired (UC) and paired (PC) conditioning trains were applied as described in A. It was confirmed that the conditioning stimulation did not activate the test Schaffer c o l l a t e r a l . During the PC, the stimulation strength to antidromically activate the test Schaffer collateral was increased to 2 times control to make sure that the f ibre was activated during PC. A s imi-lar activation of the test f ibre without the presence of the conditioning produced no changes in the exc i tabi l i ty of the test f ibre (results not shown). The amount of current required to produce an all-or-none action potential was taken as that which induced a spike in 1-2 of 3 consecutive attempts. In the graph to the right of the schematic diagram, recordings taken at 30 second intervals were plotted. Note that 1 and 5 PCs induced a 3 minute decrease while 10 PCs induced a prolonged decrease in the exci tabi-l i t y of the test f ibre terminal. Results in (A) and (Bj were from different experiments. AMOUNT OF CURRENT TO ACTIVATE SCHAFFER COLLATERAL TERMINAL JEST POPULATION EPSP AS A % OF CONTROL AS A % OF CONTROL - 001 -- 101 -to 3 minutes could be induced repeatedly without observing of LTP whereas increasing the number of paired trains to 10 resulted in STP followed by LTP (Figure 20A). The induction of LTP appeared not to be an all-or-none pheno-menon but, rather, a graded one. LTP could usually be induced using between 5 to 10 paired trains and the magnitude as well as duration of the potentia-tion increased with an increasing number of paired trains. The results from experiments using 10 paired trains were quantified because this paradigm appeared to be the optimal one for producing maximal LTP and further increasing the number of trains often produced a depression rather than f a c i l i t a t i o n of the test EPSP. The exc i tabi l i ty of the Schaffer collateral terminals was determined following 1, 5 or 10 paired and unpaired conditioning trains to stratum radiatum (Figure 20B). It was confirmed that the test neurone was not a c t i -vated by the conditioning trains . The unpaired conditioning trains did not produce any significant change in the test threshold (98 ± 3% SEM of control at 60 seconds post-5 unpaired trains, 7 of 7 expts.; 99 ± 2% SEM of control at 60 seconds post-10 unpaired trains, 7 of 7 expts.) . Paired conditioning trains, however (amount of current to stimulate test cel l was set at a suprathreshold level for action potential discharge, pairing occurred at 1 ms following onset of each conditioning t ra in) , resulted in a post-condi-tioning decrease in exc i tabi l i ty , which is reflected by an increase in threshold (threshold as a % of control at 60 seconds post-5 paired trains: 185 ± 6 SEM, 6 of 7 f ibres ) . Increasing the number of paired conditioning trains eventually led to a long-lasting decrease in the Schaffer collateral terminal exc i tabi l i ty (threshold as a % of control at 15 min post-10 paired trains : 154 ± 5 SEM, 5 of 6 fibres) (Figure 20 B_). - 102 -4.11.2 Stratum oriens--conditioning. Unpaired conditioning trains (10 pulses at 100 Hz, one train every 5 seconds) delivered to stratum oriens produced a post-tetanic depression of the test stratum radiatum-induced population EPSP (Figure 21). When the conditioning trains were paired with the test EPSP (at 1 ms after the onset of each t ra in) , a post-conditioning potentiation could be seen (population EPSP as a % of control at 60 seconds post-5 paired trains : 121 ± 4 SEM, 5 of 5 expts.) . The temporal re lat ion-ship between the test EPSP and the conditioning train for the induction of STP was determined. In order to e l i c i t STP, the test stimulation could not precede the onset of the conditioning train by greater than 50 ms or f a l l more than 80-90 ms after the beginning of the conditioning tetanus (Figure 21). The decrease in exc i tabi l i ty of Schaffer collateral terminals observed with stratum radiatum conditioning was also seen following stratum oriens conditioning (threshold as a % of control at 60 seconds post-5 paired trains by oriens: 138 ± 6 SEM, 5 of 6 expts.) . As before, unpaired conditioning trains did not produce any threshold changes (threshold as a % of control at 60 seconds post-5 unpaired trains of oriens: 99 ± 3 SEM, 6 of 6 expts.) 4.11.3 A-lveus conditioning; Conditioning stimulation to the alveus produced similar effects to stratum radiatum or stratum oriens conditioning on the test EPSP (Table 7). 4.12 Effects - of- a -transient interruption- of -input - stimulation on the EPSP  and Schaffer col lateral terminal excitabi1ity Control stimulation frequency of stratum radiatum was 0.2 Hz. A 10 minute period of non-stimulation of this input resulted in a potentiation of - 103 -120 110 o o u_ o a * < CO <C o. co 2 100 -J Z3 • D_ O 90 L (8) : (3) (3) (15)/ ; (17) i l\(10)(12) ( 1 3 i I - 3 , ( ^ f (23) * P < 0.05 ; P < 0.01 JL •100 -80 -60 -40 -20 0 20 40 60 INTERSTIMULUS INTERVAL (ms) 80 100 Figure 21 The l i m i t s of the temporal re la t ionship between condi t ioning and test st imu-l i for the induction of associat ive p o t e n t i a t i o n . The test EPSP was evoked by s t imulat ion of stratum radiatum and the condi t ioning was achieved through st imulat ion of stratum oriens (5 t r a i n s , 10 pulses in each t r a i n at 100 Hz, one t r a i n every 5 seconds). One test stimulus was paired with each of the f i v e condit ioning t r a i n s at every inters t imulus in terva l examined. The condi t ioning- tes t i n t e r v a l was varied berween -100 to +100 ms. A negative delay indicates that the test stimulus preceded the onset of the condi t ion-ing t r a i n and a p o s i t i v e delay indicates the time at which the test popula-t ion EPSP was evoked fo l lowing the onset of the condi t ioning tetanus. Each point on the graph ( f i l l e d c i r c l e s ) represents mean ± SEM of the test popu-l a t i o n EPSP magnitude measured at 1 minute post-5 paired t r a i n s . The one point represented by the f i l l e d diamond shows a s i g n i f i c a n t depression of the test EPSP at 1 minute post-5 unpaired t r a i n s of stratum oriens ( i . e . , test EPSP was not evoked during condi t ion ing) , n i s shown in parentheses above each point on the graph. - 104 -Table 7. Post-Conditioning Potentiation Induced by Pairing Tetanic Trains of the Alveus with a Single Stimulation of the Test Input. . Alveus Time post -paired conditioned tetanus 60 s 15 min Test population EPSP 146 ±. 7 SEM 133 ± 5 SEM as a % of contro l n = 6 n = 4 - 105 -the CAj_ population EPSP which lasted 15 to 20 minutes upon return to the control stimulation (Figure 22A). To ascertain i f this potentiation was associated with any changes in Schaffer collateral terminal exc i tabi l i ty , the threshold for antidromic activation of CA^ cel ls (Figure 6) was monitored before and after the 10 minute "rest" period. Results indicate that the potentiation of the EPSP is not accompanied by any exci tabi l i ty changes of the Schaffer collateral terminals (Figure 22J3). 4.13 Effects - of elevated extracellular K + on-- the- EPSP and Schaffer  collateral terminal exc i tabi l i ty An exposure of the slices to a medium containing 20 mM K + for 10 minutes resulted in an i n i t i a l potentiation of the response (at 30 seconds to 1 minute during application) followed by a rapid suppression that lasted for the duration of the application and for about 1 minute following i ts termination. This period of non-responsiveness was succeeded by a potentia-tion of the population EPSP that lasted roughly 20 minutes (Figure 23AJ. The same elevated K exposure caused an i n i t i a l drop in the Schaffer collateral terminal threshold during the application at a time that corres-ponded to the increase in the EPSP size . This increase in exc i tabi l i ty lasted for about 30 seconds and very soon after , the f ibre was not evokable with stimulation strengths of 3 to 5 times control. As with the EPSP, the spike in the CA 3 cel l did not begin to recover until 1 minute after return to standard medium. The i n i t i a l phase of recovery was characterized by a delay to onset of the antidromic spike of 2-3 ms but this effect was short-lived as the action potential regained i ts original pre-drug latency in 1-2 minutes, presumably because the excess K + was eliminated. The - 106 -Figure 22 Effects of a 10 minute interruption of input stimulation on the magnitude of the test population EPSP and Schaffer collateral terminal e x c i t a b i l i t y . The control rate of stratum radiatum stimulation to evoke a population EPSP was 0.2 Hz. Following an interruption of this control stimulus frequency with a 10 minute "rest" period, the EPSP was potentiated (A). This potentiation was not associated with any changes in Schaffer collateral terminal excita-b i l i t y (B). The bars above the graphs show the time at which the input was not stimulated and each point on graphs represent mean ± SEM (n = 10 for each set of experiments). SCH ANTIDROMIC THRESHOLD DO AS A % OF CONTROL CAj POPULATION EPSP AS A % OF CONTROL Cn C O O O O ro U 3 cp o o o ro o 4^  O cr> o C O ro —I O 3 —t—» 3 cn ro O ro cn U O - 108 -Figure 23 Effects of elevated e x t r a c e l l u l a r K + (20 mM, 5 min) on the CAi popula-t ion EPSP and Schaffer c o l l a t e r a l terminal e x c i t a b i l i t y . Following the high K + exposure, the population EPSP was potentiated (A) and th i s potent iat ion was associated with an increase in the Schaffer c o l l a t e r a l terminal thres -hold ( i . e . , a decrease in e x c i t a b i l i t y ) (B_). The bars in the graphs repre-sent the duration of 20 mM K + - conta in ing medium exposure to the whole bath. Each point on graphs is mean ± SEM ( n = 10 for each set of e x p e r i -ments). - 109 -a- 80 L <_> • CO I _ J I I 1 .1 0 5 10 15 20 25 TIME (min) - 110 -post-treatment threshold was at a level above control for about 20 minutes (Figure 23B). Since high K + treatment results in a post-application potentiation of the EPSP and, l ike LTP, is associated with a decrease in presynaptic terminal exc i tabi l i ty , these two potentiations may share a common mechanism. The presynaptic exc i tabi l i ty change could be due to several factors and i t was of interest to determine i f either Na + -inactivation or a hyperpol-arization could account for i t . It was determined in previous experiments (Table 5) that an elevation of K + in the bathing medium from 3.1 mM to 4.5 mM produces an increase in exc i tabi l i ty (reflected as a decrease in thres-hold) of Schaffer collateral terminals. If the prolonged decrease in exci-t a b i l i t y of Schaffer collateral terminals which is associated with the potentiation of the EPSP is due to Na + - inactivation, one would expect the depolarizing effect of raising extracellular K to produce more inactiva-tion and, hence, a further elevation of the threshold. In contrast, i f the exc i tabi l i ty decrease is a result of a hyperpolarization of the terminals, then i t is reasonable to assume that the K -induced depolarization w i l l cause a decrease in threshold. The results show a f a l l in the threshold by treatment of the s l i ce with 4.5 mM K + -containing medium during the period of decreased exci tabi l i ty (threshold as a % of pre-K + control at 30 seconds during 4.5 mM K + perfusion: 73 ± 7 SEM, 8 of 10 expts.) . This observation, although not confirming that the decrease in exci tabi l i ty is due to hyperpolarization, certainly supports this premise and rules out the p o s s i b i l i t y of Na + -inactivation as a mechanism. - I l l -++ + 4.14 Ca -dependence of K -induced potentiation-of-the EPSP In preliminary studies using standard medium, i t was found that synap-t i ca l ly -dr iven responses never recovered to control size following exposure of the slices to Ca + + - f ree (Mn + + or Co + + -containing) media, presumably because these Ca + +-antagonists were not tota l ly eliminated. Therefore, in this series of experiments, picrotoxin (10 pM) and EDTA (200 pM) were added to a l l media, the former to f a c i l i t a t e LTP induction (Wigstrom and Gustafsson, 1983) and the latter to hasten the removal of M n + + and C o + + . Elevation of extracellular K + (10-80 mM) during perfusion with Ca -free (Mn ) medium resulted in a dose-dependent potentiation of the population EPSP at 15 minutes following termination of the application. The same treatment (5-80 mM K + ) when administered in control Ca + + -containing medium, f a i l e d to produce any potentiation at 15 minutes post-treatment (enhancement of the response was seen, however, prior to 15 minutes but this did not appear to be sustained). Rather, the response exhibited a tendency towards depression, especially with the higher doses (Figure 24). Exposure of the slices to a 9 minute infusion of Ca -free (Mn ) medium (with normal K + concentration of 3.1 mM) resulted in a sl ight post-treatment potentiation of the response (population EPSP as a % of control at 15 min post-Mn + + : 111 ± 3 SEM, n = 8). To establish that the K +-induced potentiation was due to the Ca + + -blocking effect of M n + + and not due to a property peculiar to that ion, the experiment was conducted with C a + + -free medium containing 1 mM C o + + . Similar results were obtained with medium containing this Ca + + -antagonist at a high K + dose (80 mM) (popu-lation EPSP as a% of control at 15 min post -Co + + : 142 ± 14 SEM, n = 5) - 112 -Figure 24 The post-treatment potentiat ion and depression of the population EPSP i n -duced by appl ica t ion of elevated K + in the absence (1 mM M n + + , 7 mM Mg ) and presence (4 mM C a + + , 4 mM Mg + + ) of e x t r a c e l l u l a r C a + + , r e s p e c t i v e l y . The graph for K + exposure j n C a + + - f r e e medium i s shown by the f i l l e d diamonds and the graph for K appl i ca t ion in Ca + + - conta in ing standard medium i s shown by the f i l l e d c i r c l e s . The e x t r a c e l l u l a r K + was increased from 3.1 to 5-80 mM in C a + + - c o n t a i n i n g medium and 10-80 mM i n Ca free medium. A l l K + appl icat ions (3 minutes duration) were pre-ceded by (4 minutes) and followed by (2 minutes) infusions of C a + + - f r e e (1 mM M n + + , 7 mM Mg + + ) medium. Note that there i s a d e f i n i t e increase in the population EPSP s ize for increasing K* concentrations when applied i n Ca - f ree medium and, although not s t a t i s t i c a l l y s i g n i f i c a n t , a trend towards a decrease with increasing K + concentrations applied in C a + + -containing medium. Each point on graph represents mean ± SEM of population EPSP s ize measured at 15 minutes post-exposure to elevated K + . n i s shown by the number in parentheses above each p o i n t . - 113 -i ' ' ' I 1 1 — — - i _ 1 0 10 20 30 40 50 60 70 80 K + CONCENTRATION (mM) - 114 -but lower K + concentrations (10-40 mM) fa i led to produce consistent poten-tiation of the EPSP. Furthermore, Ca + + - f ree (Co + + ) application for 9 minutes (containing normal K + of 3.1 mM) also fa i led to produce the post-application potentiation exhibited following C a + + - f r e e (Mn + +) adminis-tration (population EPSP as a % of control at 15 min post Co : 93 ± 3 SEM, n = 4). The recovery rate of the EPSP appeared to be much faster f o l -lowing M n + + than C o + + application. Perhaps, EDTA chelates M n + + better than C o + + and, therefore, preferentially fac i l i ta tes the washout of the former. If C o + + is not eliminated rapidly and completely, i t is not surprising that post-application potentiation of the EPSP may not be ob-served. However, i f the potentiation is suff ic ient ly large to overcome the depression caused by the residual Ca + + -antagonist , then one may be able to see i t . + 3 4.15 Na -independent- H-glutamate binding Following a high frequency (400 Hz) tetanic stimulation to stratum radiatum, LTP of the population spike is present. However, there seems to + 3 be no change in the Na -independent H-glutamate binding (presumed to be due to glutamate receptors by Baudry and Lynch [1980a]) at 10 minutes post-tetanus (Table 8). However, when a low frequency (20 Hz) tetanus is delivered to the input, a post-tetanic homosynaptic depression of the popu-3 lation spike accompanied by an increase in the H-glutamate binding results (Table 8). Similarly, a depression of the population spike ten minutes following a transient exposure of slices to Cl~-free medium is associated with an increase in binding (Table 9). Note that the population spike is markedly potentiated during infusion with Cl~-free medium but - 115 -+ o Table 8. Na -Independent H-Glutamate Binding Following Tetanic Stimulation cf Stratum Radiatum. 10 min post- 10 min post-20 Hz, 600 pulses 400 Hz, 200 pulses Population spike Range 3 4 - 5 9 210 - 335 ( %of control) Mean * SEM 42 * 5* 293 * 15* n 10 si ices 10 sl ices Na+-independent 3 H-glutamate bind-Range 110 - 135 92 - 103 ing ( % of unstimu- Mean * SEM 121 * 2* 9 8 * 2 lated control) n 10 groups + 10 groupst * p < 0.01 when compared to untetanized control using paired t - test . Each group has 5 s l i c e s . - 116 -Table 9. Na+-Independent ^H-Glutamate Binding Following A Transient Exposure to Cl _-Free Medium. 4 min in CI--free medium 10 min post-C l - free medium Population spike Range 205 - 456 43 - 70 (% of pre-treatment Mean * SEM 356 * 30* 58 * 3* control) n 10 s l i c e s 10 s l i c e s 3 H-Glutamate binding following exposure to Range 123 - 144 Cl~-free medium (%of Mean * SEM 131'* 2 + untreated control) n 10 groups^ * p < 0.01 when compared to pre-treatment control using paired t-test. + p < 0.01 when compared to untreated control using paired t - t e s t . £ Each group has 5 s l i c e s . - 117 -that this increase is not sustained following the termination of the a p p l i -cation (Table 9). In some experiments (n = 4 s l i c e s ) , the stimulating site was kept constant and the recording electrode was moved to various positions between the CA2 and CA^ g regions along the cel l body layer of the s l i c e . It was discovered that responses of substantial size could be detected from a l l the recording s i tes . This observation suggests that postsynaptic activation due to the present method of stimulation is extensive and i t is probably un-necessary to use multiple stimulating sites to activate an "adequate" number of input fibres as was done by Lynch et a l . (1982). Furthermore, the demon-3 stration of a rel iable increase in H-glutamate binding following a low frequency tetanus but not following a high frequency tetanus (although the stimulating electrode, stimulation sites and stimulation strengths were the same in both cases), indicates that the lack of change in binding following the high frequency tetanus was not due to an inadequate activation of input f ibres . 3 4.16 H-Glutamate-accumulation into whole si ices 3 H-Glutamate accumulation into slices following a high frequency (400 Hz) LTP-inducing tetanus of stratum radiatum is decreased at 10 minutes post-tetanus (Table 10). Wieraszko (1983) and I, in this study, presumed that the accumulation of radioactivity into slices reflects the act ivi ty of a Na+-dependent glutamate uptake process. Since the induction of LTP appears to be Ca + +-dependent (Dunwiddie and Lynch, 1979; Wigstrom et a l . , 1979), I decided to examine i f the reduction in presumed uptake can account for LTP by delivering the high frequency tetanus in the absence of extracel-- 118 -Table 10. °H-Glutamate Accumulation Into S l i c e s Following A High Frequency (400 Hz, 200 pulses) Tetanic St imulat ion of Stratum Radiatum. 10 min post-400 Hz, 200 pulses A. ^H-Glutamate accumulation Range 25 - 80 into s l i ces (% of Mean ± SEM 60 * * 5 unstimulated contro l ) n 10 groups^ 3 B. H-Glutamate accumulation f o l -lowing tetanus i n Ca-free (lmM Mn, Range 16 - 86 3mM Mg) medium (% of control Mean ± SEM 53 * 8 + tetanized i n normal medium) n 10 groupst In A, H-glutamate accumulation in to tetanized s l i c e s was expressed as a 3> of accumulation into untetanized contro l s l i c e s . In _B, 3 ++ H-glutamate accumulation in to s l i c e s te tanized i n Ca - f ree medium was expressed as a % of accumulation in to s l i c e s tetanized i n . normal medium. * p < 0.01 when compared to unstimulated contro l using paired t - t e s t . + p < 0.01 when compared to control tetanized i n normal medium using paired t - t e s t . + Each group has 5 s l i c e s . - 119 -4-4" ~H* 4*+ lular Ca (medium contained 1 mM Mn , 3 mM Mg ). The results obtained from this experiment indicate that there is even a further reduc-tion in the uptake when the tetanus is given in C a + + - f r e e medium (Table 10). When a low frequency (20 Hz) tetanus is given to the stratum radiatum input in the presence of standard medium, there is in most cases an increase 3 in H-glutamate accumulation (210 ± 23% SEM of unstimulated controls at 10 minutes post-tetanus, 7 of 12 groups of 5 sl ices each). There was no change in 1 of 12 and a decrease in 4 of 12 groups (68 ± 10% SEM of unstimulated controls at 10 minutes post-tetanus). 5 DISCUSSION 5.1 Ca -dependence of-LTP It is widely believed that the induction of LTP requires the presence of extracellular C a + + (Dunwiddie and Lynch, 1979; Wigstrom et a l ; , 1979). The evidence provided by the two studies quoted above involve delivering tetanic stimulations to an input while synaptic transmission is blocked by ++ the removal of Ca from the bathing medium. It has been established in the present study on the associative induction of LTP that a postsynaptic activation is required during input stimulation to successfully e l i c i t the potentiation (Figure 20). Therefore, in the studies of Dunwiddie and Lynch (1979) and Wigstrom et a l ; (1979), LTP induction may have been counteracted because of lack of postsynaptic depolarization during the tetanic stimula-tion (due to blockade of evoked transmitter release during perfusion with - 120 -++ ++ Ca -free medium) and not, as the authors claim, due to Ca -dependence. The experiments in the present study involving elevated K + in C a + + - f r e e medium (Figure 24) support the hypothesis that LTP induction does not re-quire C a + + . The K + applied to the whole bath produces depolarization of both the presynaptic terminal and the postsynaptic neurones to f u l f i l l the requirements for triggering LTP. In contrast to the potentiation of the EPSP following treatment of the s l ice with high K + in Ca + + - f ree medium, the same doses of K + applied in Ca + + -containing medium appear to produce a depression (Figure 24). It was previously suggested that homo- and heterosynaptic depression is a postsynaptic phenomenon that is brought about by Ca influx into CA-^  neurones (Chirwa et a l ; , 1983; Sastry et a l ; , 1984a). It is possible that the progressive depression seen following + ++ administration of increasing K concentrations in the presence of Ca is similar to that observed during tetanus-induced homo- and heterosynaptic depression. Perhaps, the induction of LTP simply requires concurrent pre-and postsynaptic depolarization to activate a voltage-sensitive process that is independent of C a + + . The observation that elevated K + application in the absence of extracellular Ca results in a greater post-treatment potentiation of the EPSP than i ts application in the presence of C a + + may be explained simply by postulating that the Ca + + -induced depression of the EPSP masks or overshadows any potentiation that may have developed. Al ter -nately, the presence of C a + + may actually retard the development of LTP ++ i t s e l f and i t is possible to induce this phenomenon in standard Ca -con-taining medium, not because of the a v a i l a b i l i t y of C a + + but rather, in spite of i t . - 121 -A transient elevation of extracellular C a + + from 2 to 4 mM results in a post application potentiation of the CA^ population spike and EPSP (Turner e t -a l ? , 1982). The authors, who suggest that this potentiation is similar to that seen during LTP also show that there is a prolonged increase in the C a + + levels in the slices following such a treatment (Baimbridge and M i l l e r , 1981). The potentiating effect of C a + + in their studies may have been due to slow elimination of the ion from the s l i c e s . It is possi-ble that Ca + produced charge screening of membranes or persisted in the tissue to result in increased transmitter release. The observed effects may, therefore, be due to the residual Ca rather than a Ca -triggered alteration which in i t s e l f does not require elevated C a + + for maintenance. Their observations on the magnitude of the CA^ antidromic population spike due to alvear stimulation and on the responsiveness of CA-^  neurones to iontophoretically applied glutamate suggests that increased levels of the ion are present up to 30 minutes post-treatment. Furthermore, intracellular Ca concentrations can s t i l l be higher than control at 60 minutes post-treatment. The treatment used by Turner et - a-1: (1982) did not always pro-duce a depression of the synaptic response, although they reported that this occasionally occurred. Their study differs from the present one in that ++ + they raised extracellular Ca and not extracellular K . As mentioned previously, they raised extracellular C a + + from 2 to 4 mM whereas the ++ Ca concentration in the standard medium used in my study was 4 mM. Per-++ haps, Ca entry into postsynaptic neurones to produce the depression is greatly fac i l i ta ted by activation of voltage-sensitive channels. This could explain the apparent discrepancy between their results and the observations reported here. - 122 -The slight potentiation of the population EPSP observed following ++ ++ exposure of the s l i ce to Ca -free (Mn -containing) medium with "nor-mal" K + concentration (3.1 mM) resembles in magnitude and time course that seen after interruption of control input stimulation (discussed in a later section). Since synaptic transmission is blocked during the Ca + + - f ree medium perfusion, i t would be quite analogous to the situation of non-stimu-lat ion. Perhaps, these two potentiations share similar mechanisms. 5.2 NMDA receptor -involvement-in LTP Since Collingridge et a l ; (1983b) suggested that NMDA receptors are involved in the induction of LTP, NMDLA and the purported NMDA "antagonist", APV, were used in the present study to examine this p o s s i b i l i t y . There were essentially no differences between the actions of NMDLA and NMDA on the stratum radiatum-CA^ neuronal system so i t is probably safe to assume that the results obtained for NMDA would have been similar to those obtained using NMDLA here. Bath application as well as iontophoretic application at the CA-^  apical synaptic zone of NMDLA or NMDA induced a post-application potentia-tion of the population spike as well as a decrease in the exc i tabi l i ty of Schaffer collateral terminal regions, suggesting that the Schaffer col l a -teral-CA^ neuronal synaptic zone is a site of action for the amino acid. NMDLA application in the absence of extracellular C a + + (medium contains 1 mM M n + + , 3 mM Mg + + ) resulted in no post-application poten-t iat ion of the population spike and also no increase in the Schaffer c o l l a -teral antidromic threshold. The most obvious explanation i s , of course, that the induction of these changes is Ca -dependent. However, other - 123 -considerations have to be taken into account. It was reported that elevated Mg concentrations produce a voltage-dependent block of ion channels activated by NMDA (Mayer e t - a l ; , 1984; Nowak e t - a l : , 1984). By raising the Mg concentration in the extracellular medium from 2 mM in standard medium to 3 mM in C a + + - f r e e medium, one could produce a greater degree of blockade of the NMDA receptor-coupled channel. In addition, M n + + also interferes with the NMDA-induced conductance increase (Dingledine, 1983a, 1983b). Apparently, the NMDA-induced depolarization of neurones does not result from an increase in C a + + conductance (Mayer et a l . , 1984; Nowak et a l . , 1984) because C d + + does not interfere with the inward current. As postulated previously, NMDA-induced potentiation, l ike elevated K +-induced ++ potentiation could be triggered by some Ca -independent voltage-sensitive process. If this is the case, then activation of NMDA receptors per se is not necessary for the potentiation and the amino acid could just serve as an agent that produces the necessary depolarization. As evidenced by the different NMDLA doses (iontophoretic) required to e l i c i t depression (low doses) compared with potentiation of the population spike and the associated decrease in presynaptic terminal exc i tabi l i ty (larger doses), i t seems that the early depression can be more easily pro-duced. Perhaps, there are either a fewer number of NMDA-preferring recep-tors present or the a f f i n i t y of these receptors may be different on the terminals than on CA-^  neurones so that higher doses of NMDLA may be required to activate a sufficient number to trigger events leading to poten-t i a t i o n . Another explanation is that the postsynaptic depolarization pro-duced by NMDLA when coupled with presynaptic depolarization (as in the case - 124 -of elevated K + medium exposure) or input activation induces the poten-tiation of the population spike and the associated presynaptic exci tabi l i ty change. Based on the present results , i t is proposed that either presynaptic NMDA receptors can mediate an LTP-like state or activation of postsynaptic NMDA receptors results in sufficient postsynaptic depolarization to trigger the associative interactions leading to the presynaptic change. This pro-posal regarding a presynaptic mechanism for NMDLA-induced potentiation is supported by Lynch et - a l . (1985). The decrease in presynaptic terminal exc i tabi l i ty that is associated with LTP may bring about an enhancement of stimulus-evoked transmitter release. The exact mechanism of this presynap-t i c change has yet to be determined. The p o s s i b i l i t i e s that exist to pro-duce a decrease in exci tabi l i ty include a hyperpolarization, Na + - inactiva-t ion, an increase in capacitance, an increase in resting conductance or an increase in membrane resistance associated with a hyperpolarization. Of the above suggestions, only a hyperpolarization or an enhancement of membrane resistance resulting in a hyperpolarization would be conducive for increas-ing transmitter release from presynaptic terminals. Evidence provided (discussed later) suggests that Na + -inactivation is not responsible for the presynaptic change and supports the hyperpolarization theory. Although the present results suggest a presynaptic component in NMDLA-induced poten-tiation of the CA-^  population spike, a possible additional postsynaptic involvement cannot be' excluded. It is possible that the trigger for the long-lasting change in synaptic efficacy occurs at the subsynaptic level and this event then somehow influences the presynaptic terminal to result in the - 125 -sustained increase in transmission. Conversely, the opposite could be true in that a presynaptic trigger causes a long-lasting postsynaptic alteration to maintain LTP. It was hypothesized (Wigstrom and Gustafsson, 1984; Wigstrom et a l , 1985)) that a prolonged extracellular negativity (in the order of 100 ms) recorded in the apical dendrites of CA-^  neurones due to input activation plays a role in LTP induction because picrotoxin which fac i l i ta tes LTP deve-lopment (Wigstrom and Gustafsson, 1983) also potentiates this wave. They suggested that this wave is a result of activation of NMDA receptors because APV appears to counteract i t . Collingridge (1985) hypothesized that the voltage-dependent block of NMDA receptors by M g + + during normal synaptic transmission is reduced during depolarization resulting from tetanic stimu-lation of the input. This temporary "unblocking" of the NMDA system can then trigger changes leading to LTP. It has been established that the induction of synaptic potentiation does not require extracellular C a + + . If this wave is involved in LTP induction, then i t is probably not due to a C a + + conductance. The a b i l i t y of acetylcholine to produce a slow excita-tion of hippocampal neurones and neocortical neurones due to activation of muscarinic receptors has been reported (Biscoe and Straughan, 1966; Krnjevic et a l . , 1980; Krnjevic and P h i l l i s , 1963; Krnjevic and Ropert, 1981; Straughan, 1975). The sensi t ivi ty of the negative wave to muscarinic anta-gonists l ike atropine should be examined. Furthermore, i t wi l l be interest-ing to investigate the effect of muscarinic agonists and antagonists on the development of LTP. - 126 -The observed effects of APV on applied NMDLA in the CA-^  region sug-gests that the "antagonist" counteracted the actions produced by the amino acid. It has been reported that D-APV is a selective "antagonist" at NMDA receptors in the hippocampus (Collingridge et - a l ; , 1983a, 1983b, 1984; Harris et a l ; , 1984). According to Collingridge et a l ; (1984), D-APV does not affect the size of the synaptically evoked response whereas L-APV sup-presses i t . However, a recent report (King and Dingledine, 1985) suggests that D-APV may f i r s t activate and then block NMDA receptors. A partial agonist property of either the D- or the L- isomer of APV could explain why the drug suppresses the population spike, albeit weakly, during application and also why a post-application potentiation of the population spike occa-sionally results (Sastry et a l . , 1984b). These effects are also seen with the agonists NMDLA and glutamate (Chirwa et a l ; , 1984). Furthermore, i t has been observed that a tetanic stimulation delivered to stratum radiatum during application of the agonist, glutamate, produces a profound suppres-sion of the population spike (Chirwa et - a-!-., 1984). Similarly , a tetanus given in the presence of APV also produces a depression of the population spike (Sastry et a l . , 1984b). It would have been preferable to use D-APV in the present studies but, unfortunately, I did not have access to this drug. The term "antagonist" when applied to APV has to be used with caution because the agent could very well be a partial agonist. In a previous study, i t was reported that LTP could not be seen i f tetanic stimulation of stratum radiatum was given during iontophoretic application of APV in the apical dendritic area of CA-^  neurones and that a weak LTP could, in fact , be observed i f verapamil was applied along with APV (Sastry et a l ; , 1984b). - 127 -LTP induced in the presence of verapamil appears to be quantitatively larger than in control situations (Sastry et- - a l : , 1984a). Presuming that ionto-phoretic application of APV did not reach a l l sites of generation of LTP, i t is possible that verapamil potentiated LTP at these s i tes . The results concerning the action of APV on the induction of LTP con-firm earl ier reports (Col 1ingridge e t - a l . , 1983b; Harris e t - a l . , 1984) that this drug blocks the development of tetanus-induced LTP in the CA-^  region (Figure 17). It is interesting, however, that 25 pM APV was effective in counteracting the induction of LTP in the CA^ region but 100 pM was insuf-f i c i e n t to do so in the CAg area. Furthermore, the observation that NMDLA application does not produce a post-treatment potentiation of the CA3 population spike suggests different mechanisms for LTP in the CA-^  and CA3 areas. Based on reports in l i terature , i t appears reasonable to suggest that LTP in the CA-^  region is mediated by NMDA receptors. However, one has to bear in mind that there are certain inconsistencies between the characteris-tics of tetanus-induced LTP and NMDLA-induced potentiation of the population spike. For instance, a profound depression of the population spike follow-ing NMDLA application always occurs prior to any observable potentiation. In contrast, LTP induced by high frequency tetanic stimulation does not exhibit any appreciable post-tetanic depression (Sastry et - al . - , 1984a). It is possible that the effects caused by NMDLA are dose-dependent and careful adjustment of the concentration could, conceivably, result in a state that more closely resembles that obtained during LTP. NMDLA could have produced a profound shunting of the spike due to excessive depolarization. Other - 128 -arguments against the NMDA theory are supplied by the results obtained with elevated K + applications in C a + + - f r e e medium (Figure 24). F i r s t l y , the C a + + - f r e e medium in these experiments contained 7 mM M g + + and 1 mM Mn which probably results in a more effective blockade of NMDA receptor channels than C a + + - f r e e medium used in the NMDLA experiments (contained 3 mM M g + + , 1 mM Mn + + ) and yet the former treatment resulted in a potentia-tion of the synaptic response whereas the latter did not. Secondly, assum-ing that NMDA receptor activation takes place following transmitter (gluta-mate) release, i t is improbable that NMDA receptors are involved in elevated + ++ K -induced potentiation because the absence of Ca in the medium would necessarily mean a severely reduced or an abolition of both evoked and rest-ing release of neurotransmitter. Al l previous studies in support of the NMDA theory for LTP were con-ducted in the CA^ region (Collingridge et a l ; , 1983b; Harris et- al.- , 1984). The results in the present study indicating that the induction of LTP in the CAg region is not counteracted by APV and the fai lure of NMDLA to produce a potentiation of the CAg population spike suggest that the NMDA theory is not viable here either. In view of the arguments against the involvement of NMDA receptors in LTP, i t is unclear why APV blocks tetanus-induced LTP in the CA^ area. Perhaps, this drug has effects other than NMDA receptor blockade which can account for the observed results . In conclusion: 1) NMDLA produces a potentiation of the population spike, an effect that is associated with and has a similar time course to a reduction in the presynaptic Schaffer c o l l a -teral terminal exc i tabi l i ty ; 2) both the potentiation and the presynaptic - 129 -exc i tabi l i ty change are blocked i f NMDLA is applied in Ca + + - f ree medium; 3) the actions of NMDLA on the CA^ population spike and Schaffer c o l l a -teral terminal are counteracted by APV; 4) the induction of LTP produced by a tetanic stimulation to stratum radiatum in the CA^ region but not the mossy fibres in the CA^ area is blocked by APV; 5) NMDLA does not produce a potentiation of the CA^ population spike. In the present study direct evidence is provided for a presynaptic change during NMDLA-induced poten-tiation of the CA^ neuronal response. It is proposed that this poten-tiation produced by NMDLA is a result of a voltage-sensitive process which triggers a sustained increase in transmitter release from Schaffer c o l l a -teral terminals. 5.3 Interactions-among-presynaptic-terminals in the-GA^ region The results indicate that presynaptic fibres in the CA^ apical den-d r i t i c region "communicate" with each other (Table 2). Since the test f ibre does not have to be activated to see a threshold change, the interaction is not necessarily a post-spike phenomenon in the test f ibre but, rather, could be an indirect interaction with surrounding f ibres . The results obtained using two different experimental paradigms ( i . e . , separate conditioning electrode to activate other fibres or conditioning delivered through the test electrode to activate the test f ibre and other surrounding fibres) were essentially similar , suggesting that the mechanisms for mediating excitabi-l i t y changes in the test f i b r e , whether or not i t was discharging during conditioning were the same. The decreases in conditioned threshold were more marked with increasing stimulus strengths. However, when the stimula-tion intensity was kept constant and the number of conditioning pulses was - 130 -increased, the test threshold exhibited a tendency towards an increase rather than a further decrease (Table 2, Table 3). This apparently para-doxical observation can be explained i f we postulate that the interaction among fibres is due to an activity-induced elevation in extracellular K + . Table 5 i l lustrates the effects of various concentrations of extracellular K + on the Schaffer collateral terminal threshold. The lowest dose pro-duces predominantly a decrease in threshold whereas the highest dose results in an increase in threshold. Perhaps, progressive increases in extracel-lular K + levels result in a graded depolarization of fibres until a cer-tain point where further elevation of the ion produces a depolarization-induced decrease in exc i tabi l i ty of the terminals (e .g . , Na + -inactivation or hyperpolarization due to activation of a Na+-pump). The most intense depolarization resulting from the K + buildup should be at an area closest to the site of release and the depolarizing effect should diminish as one moves away from this region. Repetitive activation of the same fibres by increasing the number of conditioning pulses could result in a marked ele-vation of K + local ly at the stimulating s i t e . However, increasing the stimulus intensity would tend to activate more fibres over a wide area and, presumably, the K + buildup is not as great. This explanation would be consistent with the observed effects of elevated K + on the Schaffer c o l l a -teral threshold. If released K + is the cause for the observed presynaptic exc i tabi l i ty changes, the releasing source for this ion is unknown. The p o s s i b i l i t i e s include the presynaptic fibres themselves, postsynaptic CA^ neurones or g l ia l cel ls (MacVicar, 1984). - 131 -The time course for the increase in terminal exc i tabi l i ty following conditioning is remarkably similar to that of GABA-mediated primary afferent depolarization in the spinal cord (Eccles, 1964; Eccles et aT. , 1963a, 1963b). To speculate that GABAergic presynaptic inhibition may be responsi-ble for these hippocampal interactions is an attractive proposition but i t is unlikely because 100 LIM picrotoxin does not counteract this process (Figure 19) and also presynaptic inhibition has not been demonstrated in the hippocampus. Even i f the conventional type of "presynaptic inhibit ion" involving interneurones may not be present, there is no reason to rule out other modes of transmitter-mediated inhibi t ion . For instance, i t is possi-ble that transmitter liberated by neuronal discharge acts direct ly on the terminals of the active as well as other passive fibres nearby to alter their e x c i t a b i l i t y . As mentioned above, an alternate hypothesis for the cause of the activity-induced Schaffer collateral terminal exc i tabi l i ty changes is that neurotransmitter released by neuronal discharges acts on the active as well as surrounding passive f ibres . The results for the experiments involving bath application and iontophoretic application of glutamate suggest that this i s , indeed a p o s s i b i l i t y . It is puzzling, however, that at a l l doses used, the amino acid produced an increase in exc i tabi l i ty of the presynaptic terminals and decreases in exc i tabi l i ty were never seen. Perhaps, the concentrations of glutamate used were not as high as concentrations that would be attained in the synaptic c lef t during synaptic transmission. It is also possible that the transmitter-induced decrease in exc i tabi l i ty has a very short time course and precedes the decrease in exc i tabi l i ty or the - 132 -receptors involved are those that desensitize (e .g . , NMDA receptors [Fagni et - a l ; , 1983; Murali Mohan and Sastry, 1985]). Using the experimental methods in the present study, the earliest time at which threshold was determined was at 30 seconds following in i t ia t ion of the glutamate appl i -cation. If the duration of the glutamate-induced increase in threshold was in the order of milliseconds to several seconds, then the paradigm used would not have recognized i t . The prolonged negative wave of Wigstrbm et  a l . (1985) evoked by input stimulation which has a time course of roughly 100 ms may be responsible for the decreases in threshold of Schaffer c o l l a -teral terminals seen following conditioning. Since the wave is thought to be due to NMDA receptor activation, there is a requirement for transmitter release to occur before i t can be e l i c i t e d . There may be a depolarization in the terminal that is set-up due to presynaptic NMDA receptor activation and i t wil l be interesting to see i f APV blocks the effects of conditioning on the Schaffer collateral terminal exc i tabi l i ty . Omission of extracellular C a + + counteracts the decrease in condi-tioned threshold (Table 4). The reason for this is not entirely clear. F i r s t l y , i t is possible that the process is Ca + +-dependent. Secondly, i t could be that synaptic transmission, or at least postsynaptic depol-arization, is required. Removing C a + + from the bathing medium could have blocked several processes: 1) K + release from the postsynaptic cel l due to synaptic transmission (which could be a major source of the increase in ++ + extracellular concentration of the ion); 2) Ca -dependent K conduc-tance from pre- or postsynaptic elements (Krnjevic and Lisiewicz, 1972; Sastry, 1979a) or g l i a l cel ls (MacVicar, 1984); 3) glutamate release from - 133 -presynaptic terminals so that there is no feedback of the amino acid on these s i tes . 5.4 Associative STP-and-LTP The associative nature of LTP has been established in the hippocampus (Barrionuevo and Brown, 1983; McNaughton, 1982; McNaughton et a l ; , 1978; Robinson and Racine, 1982). In the previous section, i t was established that presynaptic terminals in the CA-^  region "communicate" with each other. Perhaps, these interactions could be involved in the associative induction of STP and LTP. The results obtained indicate that both STP and LTP can be induced without a tetanic stimulation to the test input but i t is essential that an activation of the test input occurs during tetanic stimu-lation of a heterosynaptic input or during an antidromic tetanus of the postsynaptic c e l l s . It is widely believed that post-tetanic potentiation (PTP) in various systems, including the hippocampus is a presynaptic phenomenon (del Cast i l lo and Katz, 1954; Eccles and Krnjevic, 1959b; Hubbard and W i l l i s , 1962; Lloyd, 1949; Magleby and Zengel, 1975, 1976; McNaughton, 1982; Takeuchi and Takeuchi, 1962; Wall and Johnson, 1958). During PTP in the spinal cord, the presynaptic terminal exc i tabi l i ty is decreased (Wall and Johnson, 1958) and this change has been attributed to a hyperpolarization (Eccles and Krnjevic, 1959b; Sastry, 1979a, Wall and Johnson, 1958). In the present study, the presynaptic terminal exc i tabi l i ty is decreased not only during LTP but also during STP (which may be the same phenomenon as PTP). The close association of the decrease in presynaptic exc i tabi l i ty with the potentiation of the EPSP (Table 20) suggests a cause and effect relationship. Therefore, i t is - 134 -not unreasonable to postulate that both STP and LTP are quantitatively d i s -similar but share the same mechanisms of induction. Both the presynaptic exc i tabi l i ty change as well as the potentiation of the EPSP are absent following unpaired conditioning, but are seen follow-ing paired conditioning trains . This finding suggests that the locus res-ponsible for the potentiation is the presynaptic test input but the mechanism to trigger this change is postsynaptic. It is possible that input activation has to be paired with suffic ient postsynaptic depolarization to produce the observed potentiation. Consistent with this idea, picrotoxin, which blocks postsynaptic inhibition (and causes a depolarization of the postsynaptic c e l l ) , fac i l i ta tes the induction of LTP (Wigstrom and Gustafsson, 1983). Reducing postsynaptic depolarization during tetanic stimulation of the input by iontophoretic application of GABA, tetrodotoxin or pentobarbital onto CA^ cel l bodies (Scharfman and Sarvey, 1985) or by injecting intracel lular hyperpolarizing currents (Malinow and M i l l e r , 1986) resulted in a blockade of LTP development. The communication between pre-synaptic terminals and postsynaptic neurones could be achieved through ephaptic interactions, electrotonic coupling by gap junctions, a buildup in extracellular K + or some other agent. Since i t has been determined that the induction of LTP does not require extracellular C a + + , the likelihood that released transmitter plays a role in the interaction is minimal. Activation of the presynaptic f ibre could e l i c i t a "subliminal f a c i l i -tatory process" in the terminal which cannot trigger changes leading to fac i l i ta ted transmitter release in i t s e l f but is capable of doing so i f i t is paired with adequate postsynaptic depolarization. Sastry et a l . (1986) - 135 -has demonstrated that pairing intracellular depolarizing pulses in CA-^  neurones with activation of stratum radiatum can produce LTP of the stratum radiatum-CA^ population EPSP. The prolonged dendritic negativity that occurs following input stimulation (Wigstrom and Gustafsson, 1984; Wigstrom et a l . , 1985) could be the "subliminal process" in the presynaptic terminals or i t could be the conditioning postsynaptic depolarization. The duration of the negative wave is several times longer than that of the population EPSP. The increase in exc i tabi l i ty of Schaffer collateral terminals caused by conditioning of other fibres is roughly 200 ms in duration (Table 2) and could very well be due to this same process. In order to induce STP in the stratum radiatum EPSP by pairing i t with the stratum oriens conditioning t ra in , the test EPSP has to be evoked between 50 ms before and 80 ms after the onset of the conditioning train (Figure 21). I propose that STP can only be e l i c i ted i f the "subliminal process" in the test presynaptic terminal overlaps the depolarization of the postsynaptic neurones caused by the conditioning tetanus t rain . Assuming that the "subliminal process" is about 100 ms in duration, the observation that associative STP can be produced i f the test EPSP evoked at 50 ms prior to the onset of the conditioning train is easily explained. The population EPSP evoked by a single stratum oriens stimulation has a time course of roughly 20 ms. It is possible that a conditioning train delivered to stratum oriens (10 pulses at 100 Hz) wil l produce spatial and temporal sum-mation of successive EPSPs to result in a summated EPSP of larger size and prolonged time course. Perhaps, this summated EPSP provides the adequate postsynaptic depolarization that lasts in the order of 80-100 ms to account - 136 -for the observation that STP can be e l i c i t e d even though the stratum radia-tum test input is stimulated 80 ms following the onset of the stratum oriens conditioning t r a i n . Figure 25 is a schematic i l lus t ra t ion of this hypo-thesis . Bear in mind that the temporal requirements for induction of STP were determined using a synaptic input as the conditioning stimulus. In this case, even though the time course of the stratum oriens EPSP is pro-longed by the tetanus, i t wil l inevitably decay to baseline eventually. If one could maintain an adequate level of depolarization of the postsynaptic neurones (e .g . , by intracellular current injection) , perhaps, i t wi l l be possible to extend the temporal l imit indefinitely so that the determining factor for the time course is the duration of the depolarization. Of course, this would be possible only in the case of the test stimulus occur-ing following onset of the conditioning stimulation as the duration of the "subliminal process" wil l determine the limits for the opposite paradigm (when the test stimulus precedes the conditioning). 5.5 Presynaptic involvement in-LTP It has been demonstrated that LTP is associated with an increase in evoked neurotransmitter release (Bliss et al -;, 1985; Dolphin ' e t - a l - ; , 1982; Skrede and Malthe-S^rrensen, 1981). LTP of the perforant path-granule ce l l synapse is associated with a reduction in the presynaptic terminal excita-b i l i t y (Sastry, 1982). In the present study, the relationship between potentiation of the synaptic response and the presynaptic exc i tabi l i ty change is further examined. A previous study has shown that application of an excitatory agent local ly on the cel l bodies of CA-^  neurones produces a post-application potentiation of the CA^ population spike (Goh and Sastry, - 137 -Figure 25 Schematic i l l u s t r a t i o n of the hypothetical mechanism responsible for determin-ing the temporal requirements for induction of associative STP. The top trace is a record of the summated EPSPs due to the stratum oriens conditioning t r a i n . The bottom records denoted by A and B_ represent the "subliminal process" in the presynaptic terminal for the induction of associative STP. The position of the waves in A and B_ show the maximal intervals (before and after the onset of the conditioning t ra in , respectively) at which the "subliminal process" can occur for successful induction of associative STP. - 138 -1983). The experiments involving high frequency activation of CAg cel ls by iontophoresis of DLH onto CAg cel l bodies (which mimics a tetanic stimulation to the Schaffer collaterals) produced a decrease in the Schaffer collateral terminal e x c i t a b i l i t y . In previous experiments (Sastry, 1982), a tetanic stimulation was delivered through the stimulating electrode and i t was possible that electrode properties were altered to result in erroneous interpretations of threshold changes. The present study rules out this poss ibi l i ty because no high frequency stimulation was given through the electrode used for monitoring threshold. It is interesting that DLH appli -cation in C a + + - f r e e medium did not produce any prolonged changes in pre-synaptic e x c i t a b i l i t y . One could invoke Ca + +-dependence, but in view of the findings that LTP induction does not require C a + + but does require concurrent postsynaptic depolarization (discussed e a r l i e r ) , this observation is easily explained. Obviously, in the absence of synaptic transmission during infusion of Ca + + - f ree medium, there is no activation of the post-synaptic c e l l s . Therefore, i f the decrease in presynaptic terminal excita-++ b i l i t y is to be responsible for LTP, i t should not be inducible in Ca -free medium using the experimental paradigm above. The experiments involving elevated extracellular K + effects on the population EPSP and Schaffer collateral threshold (Figure 23) i l lust ra te the close relationship with regard to magnitude and time course of potentiation of the EPSP and decrease in Schaffer collateral terminal exc i tabi l i ty . Therefore, i t is believed that the presynaptic exc i tabi l i ty change is some-how responsible for the potentiation of the EPSP. As postulated ear l ier , the increased extracellular K + probably produces a depolarization in both - 139 -pre- and postsynaptic elements to trigger long-term changes in the proper-ties of the presynaptic terminal which eventually translates to increased transmitter release. The nature of the presynaptic exc i tabi l i ty change is largely unknown. The present studies, however, have eliminated the possi-b i l i t y of Na + -inactivation because an elevation of extracellular K + (which produces a depolarization) during the phase of increased threshold ( i . e . , decreased exci tabil i ty) results in a decrease in the threshold rather than a further increase. There are several possible mechanisms that can account for the pre-synaptic change leading to LTP. It is generally thought that post-tetanic potentiation (PTP) of the monosynaptic reflex in the spinal cord is mediated by a hyperpolarization of the primary afferent terminals (Eccles and Krnjevic, 1959b; Lloyd, 1949; Sastry, 1979a; Wall and Johnson, 1958). The reduction in exc i tabi l i ty of presynaptic terminals that is associated with LTP (Sastry, 1982) may be a reflection of a hyperpolarization (Wall, 1958) of these terminals. Short-term post-tetanic potentiation and LTP in the hippocampus appear to be two separate phenomena whose properties and requirements for induction are dissimilar (Dunwiddie and Lynch, 1979; McNaughton, 1982). It is possible that the events leading to the develop-ment of PTP and LTP are different . However, they may both s t i l l be ex-plained by a common end result , that i s , a hyperpolarization of the termi-nals. If LTP, l ike PTP, is sustained by a hyperpolarization of the termi-nals, then this presynaptic change would have to increase transmitter re-lease. It has been reported that a hyperpolarization of presynaptic termi-nals leads to an increase in the size of the action potential (Eccles and - 140 -Krnjevic, 1959a, 1959b; Lloyd, 1949) which in turn results in an enhancement of evoked release of transmitter (Hubbard and W i l l i s , 1962; Takeuchi and Takeuchi, 1962). The underlying mechanism for generating a hyperpol-arization could be the activation of an electrogenic Na+-pump (McDougal and Osborn, 1976), an increase in resting potassium conductance or some other unknown cause. However, an increase in K + conductance may not result in increased transmitter release. It has been reported that gluta-mate, the suspected transmitter between Schaffer collaterals and CA-^  neu-rones (Storm-Mathisen, 1977a) when applied on hippocampal neurones, produces a post-application hyperpolarization of neuronal somata that is brought about by an increase in Na+-pump act ivi ty (Segal, 1981). Perhaps, the transmitter released during a tetanic stimulation acts on the presynaptic terminal to enhance the act ivi ty of a Na+-pump located here. Aside from a hyperpolarization of the presynaptic terminals, an i n -crease in transmitter release during LTP may be accomplished by other long-lasting alterations. An increase in the resting membrane resistance asso-ciated with a hyperpolarization can produce a larger spike in the terminal. In primary afferent terminals in the spinal cord, i t appears that the driving force for C l ~ ions is in an outward (depolarizing) direction (Curtis e t - a l : , 1977; Eccles et a l , 1963b; Gmelin, 1978). If a similar situation exists in the hippocampal presynaptic terminals, then a blockade of a resting C l ~ conductance could account for increased membrane res is -tance accompanied by a hyperpolarization. For this hypothesis to be viable, i t is necessary to assume that the neuronal membrane is freely permeable to C l ~ ions during steady state conditions. It follows, therefore, that in - 141 -order to maintain a gradient for C l ~ between the inside and outside of the terminal, there has to be some energy requiring process at play, possibly a Cl~-pump, that moves the ion in an inward direction (Nishi e t - a l ; , 1974). It has been demonstrated that C a + + influx in the presynaptic termi-nal occurs during an action potential in primary afferent fibres (Sastry, 1979c). Presuming that a Ca component exists in the terminals of hippo-campal neurones, an enhancement or prolongation of this component during LTP can result in an increase in evoked transmitter release due to greater C a + + entry per impulse. A similar mechanism was suggested for PTP (Sastry, 1979c). In fact , i t is possible that a blockade of the rectifying potassium conductance that follows an action potential can account for a more prolonged C a + + component in the spike at the presynaptic terminal (Sastry, 1979c). An increase in the synthesis and/or the presence of more transmitter in the releasable pool can lead to enhanced transmitter release. Although i t has been determined that during LTP of the CA-^  population spike pro-duced by a tetanic stimulation of the Schaffer collateral input, the resting synthesis of glutamate is not altered (Benjamin et a l . , 1983), i t has been reported by others (Corradetti et a l ; , 1983) that stimulus-evoked release of this amino acid is accompanied by an increase in neosynthesis. Since there is an enhancement of evoked transmitter release during LTP (Skrede and Malthe-Stfrrensen, 1981), perhaps the act ivi ty of the synthesis process is increased proportionally during evoked act ivi ty to compensate for the dimi-nution of the releasable amino acid pool. If this is the case, one has to postulate that at least some, i f not al l of the additional transmitter pro-- 142 -duced by an electrical activation of the input during LTP, is available for release in subsequent stimulations. Besides increasing synthesis, another p o s s i b i l i t y for increasing transmitter release is to f a c i l i t a t e the transfer of transmitter from the storage pool to the releasable pool so that a new equilibrium level is maintained between the two due to an increase in the rate constant for this process. The rate of transmitter synthesis has to be greater than the above transfer rate so that the storage pool wil l not be eventually depleted. It has been suggested that increased methylation of presynaptic com-ponents can lead to increased transmitter release due to an altered membrane f l u i d i t y (Benjamin et - a l : , 1984). Perhaps, an increase in methylation of the presynaptic terminal can account for LTP. Alternatively, the observed change can be occurring at the level of the postsynaptic neurone to result in homo- and heterosynaptic depression. Whether an increase in methylation is responsible for LTP or homo- and heterosynaptic depression should be examined. A recent report provides evidence for the involvement of a Ca + +-dependent kinase (protein kinase C) in the induction of LTP (Malenka e t - a l ; , 1986). Phorbol esters, which selectively activate protein kinase C, can produce a post-application potentiation of the synaptic response. Their results suggest that the locus of the phorbol ester's actions is presynaptic to f a c i l i t a t e transmitter release. Other possible presynaptic changes that can be responsible for LTP include an increase in the number of presynaptic boutons, an increased safety factor along the axon and i ts branches so that the probability of action potential propagation to the terminals is enhanced, an increase in - 143 -the conduction velocity of presynaptic fibres to cause a more synchronous release of transmitter resulting in a better summation of the EPSP, an increase in the size of the presynaptic bouton so that transmitter released has a possibi l i ty of activating more surface area subsynaptically, a rear-rangement of synapses to f a c i l i t a t e transmission and modulation of either transmitter release or the postsynaptic response by liberation of an unknown substance due to input activation (e .g . , noradrenaline). A reduction in the synaptic gap, i f present during LTP, can lead to an increase in the concen-tration of the released transmitter in the c lef t leading to an enhancement of the subsynaptic response. 5.6 Potentiation of-the-EPSP-following-interruption-of -input-stimulation A stable control EPSP in the CA^ area can be obtained by stimulating stratum radiatum at a frequency of 0.2 Hz. When this stimulation is inter-rupted for 10 minutes and then reinstated, the resulting EPSP is potentiated and recovers over 10 minutes to a level that is comparable to the pre-quies-cent period. This observation suggests that the control stimulation i t s e l f results in some process that suppresses the EPSP. Since the potentiation of the EPSP is not accompanied by any changes in the presynaptic Schaffer c o l -lateral terminal exc i tabi l i ty , the mechanism underlying this phenomenon may not be the same as that responsible for LTP. The potentiation produced by transiently interrupting the input stimulation, however, is quantitatively smaller than that observed following tetanus-induced, elevated K +-induced or NMDLA-induced enhancement of population responses (Figures 3, 13, 22 and 23). Perhaps, the decrease in exc i tabi l i ty of the presynaptic terminals was present but could not be detected re l iably with the experimental method due to i ts small magnitude. In view of the necessity of test input activation to induce LTP in that pathway using the associative paradigm (Figure 20), i t - 144 -is unlikely that the mechanism underlying the potentiation following an interruption of input stimulation is similar to that seen following activa-tion of the afferent f ibres . Of course, this interpretation does not pre-clude a presynaptic mechanism, but i t does suggest an alternate one. Per-haps, transmitter release is fac i l i ta ted through a greater avai labi l i ty in the releasable pool. This may arise from increased transmitter synthesis or a mobilization of neurotransmitter from the storage pool to the releasable pool. In addition, there may be a gradual depletion of transmitter due to depletion using the control stimulus frequency of 0.2 Hz. At some time, following the in i t ia t ion of stimulation, the supply vs demand of neuro-transmitter could be in equilibrium so no further reduction in transmitter output results . Presumably, the EPSP eventually settles to a stable level with a smaller magnitude than at the beginning of stimulation. Postsynaptic mechanisms could also play a role in the potentiation of the EPSP induced by non-stimulation. It is possible that there is a tonic suppression of the EPSP during input stimulation due to desensitization of NMDA receptors (Fagni e t a ! ; , 1983; Murali Mohan and Sastry, 1985) (perhaps, there is an NMDA component in the EPSP) or due to a homosynaptic depression caused by C a + + influx into CA^ neurones (Chirwa et- - a l . , 1983; Sastry et  a l . , 1984a). It is possible that the response adapts to a particular f re -quency and that an interruption of this entraining leads to an unsettling of the response which takes about 10 minutes to reequi1ibrate. 3 5.7 H-Glutamate-binding and-uptake- studies Baudry and Lynch (1980a) hypothesized that LTP is due to an increase in the number of subsynaptic glutamate receptors. It is apparent from the - 145 -Na+-independent binding results that an increase in the number of these binding sites is not a necessity for the observation of LTP (Table 8). Similar ly , B a + + which was reported not to produce an increase in glutamate binding (Baudry and Lynch, 1979), has been shown to be capable of inducing a post-application potentiation of the population spike that lasted for pro-++ ++ longed periods of time (Ca in normal medium was substituted with Ba ) (Maretic et a l . , 1984). In contrast, i t appears that there is an increase in glutamate binding associated with a depression of the population spike. The stimulation parameters for producing kindled burst discharges in hippo-campal neurones (Douglas and Goddard, 1975; Goddard et a l . , 1969; Savage et  a l . , 1982) are quite similar to the tetanic stimulations used to induce homo- and heterosynaptic depressions of the population spike (Sastry et al •, 1984a). Perhaps, the increase in glutamate binding sites observed with kindling (Savage et a l ; , 1982) is similar to that seen accompanying homo-synaptic depression. Since homosynaptic depression and LTP can co-occur (Sastry et a l ; , 1984a) the observation of one or the other of these pheno-mena wil l be determined by their relative magnitudes.. I have shown in the present study that a single 400 Hz tetanus consisting of 200 pulses, which produces minimal homosynaptic depression (Sastry et a l ; , 1984a), is capable of producing LTP of the population spike with no associated increase in glutamate binding. In the studies of Baudry et a l ; (1980) and Lynch et a l ; , (1982), the stimulations used to produce LTP were several trains of tetani at 100-300 Hz. Although they could observe LTP with such stimulation para-meters, i t is possible that the activation of postsynaptic neurones due to the repetitive trains of tetani produced an underlying depression that was - 146 -overshadowed by LTP. It is possible that the increase in glutamate binding observed by them, therefore, was associated with the homosynaptic depression rather than LTP. Although Baudry and Lynch (1980a) presume that the gluta-mate binding sites are receptors, they show no physiological or pharmaco-logical evidence to support their proposal. It was suggested by the same authors (Baudry et - a l ; , 1980) that the a f f i n i t y of the glutamate binding sites did not change following the induction of LTP. The dissociation constant (K^) for high a f f i n i t y Na+-independent binding was reported to be 750 nM (Baudry and Lynch, 1981). Based on this observation, one would suspect that the binding sites are not receptors because concentrations of hundreds of pM glutamate are required to produce any measurable physiologi-cal response. On the other hand, i t is possible that alterations in binding properties due to the biochemical procedures occurred. Furthermore, a recent paper by Garthwaite (1985) implicates cel lular uptake of glutamate in disguising the true potency of this amino acid on receptors. Perhaps, in intact l ive tissue, the apparent potency of glutamate is much lower than in isolated membrane preparations because the uptake mechanism would be present in the former but not the latter instance. If an increase in the number of subsynaptic receptors is responsible for LTP, then one has to postulate that the existing receptors are f u l l y occupied during synaptic transmission so that the postsynaptic response for each given neurone is always maximal when activated. Presuming that the neurotransmitter released is in excess of that required for synaptic transmission, then an increase in the number of subsynaptic receptors would necessarily mean a corresponding increase in the quantal unit . Quantal analysis of LTP in various other systems has shown - 147 -that quantal content is increased with no change in the quantal unit (Baxter et a l . , 1985; Briggs et a l ; , 1985; Koyano e t - a l ; , 1985). Perhaps, the same situation applies to the stratum radiatum-CA^ system in the hippocampus. Exposure of hippocampal membranes to elevated Ca concentrations (Baudry and Lynch, 1979) as well as low frequency tetanic stimulation of hippocampal s l i c e s , which causes C a + + influx into CA^ cel ls (Chirwa et a l . , 1983), produce an increase in Na+-independent glutamate binding. On the other hand, a high frequency tetanus does not cause as much elevation in CA^ intracel lular C a + + (Chirwa et a l ; , 1983) and there is also no correspond-ing increase in glutamate binding (Table 8). Therefore, i t is possible that the increase in glutamate binding sites is a Ca -mediated event in the CA^ neurones. Recent observations (Murali Mohan and Sastry, 1985) suggest that there is a Ca + +-dependent desensitization and supersensitivity of glutamate receptors in CA^ neurones. The increase in glutamate binding seen in the present study and those of others (Baudry and Lynch, 1979; Baudry et a l ; , 1980; Lynch et a l 1 9 8 2 ) may be due to an increase in gluta-mate receptors responsible for the Ca+ +-dependent increase in responsive-ness of CA^ neurones to glutamate seen by Murali Mohan and Sastry (1985). Since these "receptors" are extrasynaptic (Fagni et a l ; , 1983; Murali Mohan and Sastry, 1985), i t is questionable whether they play a role in normal synaptic transmission. If they are, indeed, functional extrasynaptic recep-tors, then the action of exogenous drugs on these receptors wil l vary depending on the level of the Ca + +-mediated changes. The increase in the number of receptors to be associated with a depression rather than an increase in the response of CA^ neurones to the inputs is similar to the - 148 -situation in denervation supersensitivity where the number of receptors appear to increase because of a disuse of the synapses. Baudry et a l . (1980) reported that the Na+-dependent uptake system did not change following tetanic stimulations. In a previous publication (Sastry and Goh, 1984), i t was presumed that these authors were correct and, therefore, we did not account for any changes in Na+-dependent uptake to influence the results for Na+-independent binding. As can be seen in the present study, tetanic stimulations to an input do, in fact , change the uptake of glutamate dras t ica l ly . Furthermore, Wieraszko (1983) reported that a high frequency tetanus of the Schaffer collaterals produces a de-crease in stimulus-evoked uptake of the neurotransmitter. In the present study, i t is demonstrated that the "resting" glutamate uptake is decreased. Since LTP was associated with a decrease in the uptake of glutamate, i t was of interest to examine i f this could be a mechanism for LTP. This possibi-l i t y was ruled out because unlike LTP, the induction of this uptake change could not be blocked by delivering the tetanic stimulation during infusion of Ca -free medium. In fact , there was a further reduction in the up-take, the reason for which is unknown. The reduction in uptake appears to be attributable to elements other than the postsynaptic c e l l . However the increase in the uptake could be due to an activation of postsynaptic neu-rones because there is a further decrease in uptake due to a 400 Hz tetanus given in C a + + - f r e e medium (which abolished synaptic transmission) as com-pared to a 400 Hz tetanus given in normal medium. Also, in the majority of cases following a 20 Hz tetanus (which produces frequency f a c i l i t a t i o n of the population spike during the tetanus [Sastry e t a l : , 1984a]), an increase - 149 -in uptake results . There may be a balance between the increase and decrease in the uptake following the 20 Hz tetanus so that the more predominant effect of the two should be seen. Perhaps, this is why in some instances the uptake was increased and in others decreased following the 20 Hz teta-nus. It is possible that the changes in the uptake are not peculiar to different elements but are a function of the tetanus frequency, high f re -quency favourable to induce a decreased uptake and a low frequency to induce an increased uptake. It may also be that the induction of the increase in uptake is Ca + +-dependent. Since a decrease in the uptake does not appear to increase synaptic transmission, perhaps this uptake alteration is located nonsynaptically and does not affect synaptically released glutamate. It was reported by Baudry and Lynch (1981) that Na+-dependent uptake is saturable with a dissociation constant of 2.4 uM. If the concentration of the neuro-transmitter released during synaptic transmission markedly exceeds that required for saturation of this system, then a slight decrease or increase in the uptake should have minimal implications for synaptic transmission. In this connection, i t was shown by Curtis et a l ; (1976) that nipecotic acid, an uptake blocker for GABA, fai led to enhance GABAergic inhibition of Purkinje c e l l s . The significance of the changes in the uptake of trans-mitters following tetanic stimulations is unclear at present. Moreover, 3 since H-glutamate accumulation into whole slices was being measured, i t is uncertain as to how much of this accumulation was through specific gluta-mate uptake systems. It was shown by Wieraszko (1983) that stimulus-evoked uptake is decreased following a high frequency (100 Hz) tetanus. This uptake change was evident at stimulus frequencies between 1 and 20 Hz and - 150 -was not s ignif icantly different from non-tetanized controls at both of these extremes. Therefore, one would expect that there be no change in stimulus-evoked uptake following induction of LTP at test frequencies of < 1 Hz or > 20 Hz. Since in studies of LTP, the normal stimulation frequency used to evoke the control population spike is 0.1-0.2 Hz, i t is probable that stimulus-evoked uptake is not altered during LTP using these stimulation parameters. 5.8 Physiological significance of-studies on f i e l d potentials A synchronous activation of a population of input fibres to evoke a population spike or a population EPSP in the postsynaptic neurones is proba-bly not a common occurrence under physiological conditions. One may think that the studies on LTP as conducted in the present investigation are just an anomaly created in a laboratory setting and are irrelevant for natural processes. This may not be entirely true because i t has been shown that high frequency asynchronous act ivi ty in CA^ neurones (which can be seen in physiological situations) results in LTP of the CA^ population spike (Goh and Sastry, 1983). Even though the population spike is used for quantita-tive purposes in these experiments, i t is not d i f f i c u l t to imagine that a transient increase in the f i r i n g rate of presynaptic cel ls due to activation of a sensory input causes a subsequent increase in synaptic transmission at each of the activated synapses. In fact , according to Andersen and Langmoen (1981), only 60 synapses have to be activated simultaneously to discharge an action potential in a pyramidal c e l l . This roughly corresponds to 60 pre-synaptic fibres because an all-or-none EPSP evoked by stimulation of a single f ibre has a magnitude of roughly 0.1 mV (Andersen and Langmoen, - 151 -1981). Perhaps, after LTP is induced, activation of a fewer number of synapses can cause the cell to discharge. It is quite plausible that 60 synapses can be activated simultaneously because each postsynaptic pyramidal ce l l is thought to have 10,000 synaptic contacts (Hamlyn, 1963). Since some evidence has given me reason to doubt the involvement of the hippocampus and LTP in conscious learning and memory (Isseroff et a l . , 1976; Jarrard, 1983, 1985; Laroche, 1985; Stein e t - a l . , 1969), there is no reason to exclude the poss ibi l i ty that LTP is responsible for "subconscious" memory. As mentioned ear l ier , LTP has been observed in other systems (sympathetic ganglia, neocortex and crayfish neuromuscular junction), some of which cannot be involved in learning and memory as we know i t ( i . e . , the f a c i l i t a t i o n of synaptic transmission is not contingent upon conscious awareness and thought processes). The "memory" in such systems could improve functioning of the organism at a subconscious level . For instance, LTP at the crustacean neuromuscular junction could result in fac i l i ta ted reflexes for muscle contraction and LTP at autonomic ganglia could result in more eff ic ient autonomic control in the organism. It may be possible that one could "learn" to improve their muscular reflexes by training. Perhaps, the mechanism for voluntary control of autonomic functions via the phenome-non of biofeedback is through LTP of autonomic ganglia. 6 CONCLUSIONS 1. The induction of LTP is not Ca -dependent but does appear to i n -volve voltage-sensitive processes. - 152 -2. There is a requirement for depolarization or activation of the pre-synaptic terminals in the presence of adequate postsynaptic depol-arization to successfully e l i c i t LTP. 3. Both STP and LTP can be induced through an associative conditioning paradigm that does not involve prior tetanization of the test input. The temporal constraints governing the induction of associative STP of the test stratum radiatum input by conditioning delivered through the stratum oriens input were determined. To induce associative STP, i t is necessary to activate the test input not more than 50 ms before or 80 ms after the onset of the conditioning t r a i n . 4. Since STP and LTP are consistently associated with a decrease in pre-synaptic terminal exc i tabi l i ty , perhaps the presynaptic change is responsible for increasing evoked transmitter release leading to potentiation of the EPSP. 5. Interactions among presynaptic f ibre terminations in the CA^ region may play a role in the associative induction of STP and LTP. 6. The mechanism underlying the potentiation of the EPSP due to interrup-tion of input stimulation differs from that responsible for STP and LTP. 7. It is unlikely that NMDA receptors are involved in the induction of LTP. 8. An increase in the number of subsynaptic glutamate receptors is not a prerequisite for the observation of LTP. 9. A decrease in the uptake of neurotransmitter is not a mechanism for LTP. - 153 -10. The evidence provided strongly supports a presynaptic locus for STP and LTP. However, an additional postsynaptic mechanism cannot be excluded. 7 REFERENCES ALGER, B. E. and NICOLL, R. A. (1979). GABA mediated biphasic inhibitory responses in hippocampus. Nature-(Lond.) 281: 315-317. ALGER, B. E. and NICOLL, R. A. (1980). Epileptiform burst hyperpolariza-t ion : calcium dependent potassium potential in hippocampal CA^ pyramidal c e l l s . Science 210: 1122-1124. ALGER, B. E. and NICOLL, R. A. (1982). Feed-forward dendritic inhibition in rat hippocampal pyramidal cel ls studied in v i t r o . J.- • • Physiol•;  (Lond.) 328: 105-123. ALGER, B. E. and TEYLER, T. 0. (1976). Long-term and short-term p l a s t i c i t y in the CA^, CA3 and dentate regions of the rat hippocampal s l i c e . Brain Res: 110: 463-480. ANDEN, N-E . , DAHLSTROM, A . , FUXE, K . , LARSSON, K . , OLSON, L. and UNGERSTEDT, U. (1966). Ascending monoamine neurons to the telencephalon and dien-cephalon. Acta physiol . Scand. 67: 313-326. ANDERSEN, P. (1960a). Interhippocampal impulses II. Apical dendritic a c t i -vation of CAi neurons. Acta physiol : Scand; 48: 178-208. ANDERSEN, P. (1960b). Interhippocampal impulses III . Basal dendritic a c t i -vation of CA3 neurons. Acta physi01; Scand.- 48: 209-230. ANDERSEN, P. (1975). Organization of hippocampal neurons and their inter-connections. In: R. L. Isaacson and K. H. Pribam (Eds) The  Hippocampus. V o l : 1. Plenum Press, New York. pp. 155-175. ANDERSEN, P . , BLACKSTAD, T. W. and L0MO, T. (1966a). Location and i d e n t i f i -cation of excitatory synapses on hippocampal pyramidal c e l l s . Exp;  Brain Res. 1_: 236-248. ANDERSEN, P. , BLISS, T. V. P. and SKREDE, K. K. (1971). Lamellar organiza-tion of hippocampal excitatory pathways. Exp.- Brain - - Res-; 13: 222-238. — ANDERSEN, P . , ECCLES, J . C. and L0YNING, Y. (1964a). Location of post-synaptic inhibitory synapses on hippocampal pyramids. J ;  Neurophysiol. 27: 592-607. - 154 -ANDERSEN, P . , ECCLES, J . C. and L0YNING, Y. (1964b). Pathway of post-synaptic inhibit ion in the hippocampus. J ; Neurophysiol; 27: 608-619. ANDERSEN, P . , HOLMQVIST, B. and V00RH0EVE, P. E. (1966b). Entorhinal a c t i -vation of dentate granule c e l l s . Acta physiol ; Scand; 66: 448-460. ANDERSEN, P. and LANGMOEN, I. A. (1981). Synaptic interaction on hippocampal pyramids. In: E . Grastyan and P. Molnar (Eds) Sensory- Functions. Akademiai Kiado, Budapest, pp. 269-273. ANDERSEN, P. , SILFVENIUS, H . , SUNDBERG, S. H . , and SVEEN, 0. (1980a). A comparison of distal and proximal dendritic synapses on CA} pyramids in guinea-pig hippocampal slices in v i t r o . J . - Physiol.- -(Lond.) 307: 273-299. ANDERSEN, P . , SUNDBERG, S. H . , SVEEN, 0., SWANN, J . W. and WIGSTROM, H. (1980b). Possible mechanisms for long-lasting potentiation of synaptic transmission in hippocampal slices from guinea pigs. J ; Physiol.  (Lond.) 302: 463-482. ANDERSEN, P. , SUNDBERG, S. H . , SVEEN, 0. and WIGSTROM, H. (1977). Specific long-lasting potentiation of synaptic transmission in hippocampal s l i c e s . Nature (Lond.) 266: 736-737. AZMITIA, E. C. and MAROVITZ, W. F. (1980). In vitro hippocampal uptake of •^H-serotonin ( 3H-5-HT); a morphological, biochemical and pharmaco-logical approach to s p e c i f i c i t y . J . Histochem.- Cytochem; 28: 636-644. AZMITIA, E. C. and SEGAL, M. (1978). An autoradiographic analysis of the differential ascending projections of the dorsal and median raphe nuclei in the rat . J . Comp.-Neurol; 179: 641-668. BABB, T. L . , WILSON, C. L . , HALGREN, E. and CRANDALL, P. H. (1980). Evidence for direct lateral geniculate projections to hippocampal formation in man. Soc- Neurosci. -Abstr.- 6_: 838. BAIMBRIDGE, K. G. and MILLER, J . J . (1981). Calcium uptake and retention during long-term potentiation of neuronal act ivi ty in the rat hippo-campal s l i ce preparation. Brain -Res; 221: 299-305. BALCAR, V. J . and JOHNSTON, G. A. R. (1972). The structural spec i f ic i ty of the high a f f i n i t y uptake of L-glutamate and L-aspartate by rat brain s l i ces . J ; Neurochem. 19: 2657-2666. BARRIONUEVO, G. and BROWN, T. H. (1983). Associative long-term potentiation in hippocampal s l i ces . Proc; Natl ; Acad; S c i . USA 80: 7347-7351. BAUDRY, M. , BUNDMAN, M . , SMITH, E. and LYNCH, G. (1981a). Micromolar levels of calcium stimulate proteolytic ac t ivi ty and glutamate receptor binding in rat brain synaptic membranes. Science 212: 937-938. - 155 -BAUDRY, M . , KRAMER, K. and LYNCH, G. (1983). I r revers ibi l i ty and time course of calcium stimulated [3H]glutamate binding to rat hippocampal mem-branes. Brain Res. 270: 142-145. BAUDRY, M. and LYNCH, G. (1979). Regulation of glutamate receptors by cations. Nature (Lond.) 282: 748-750. BAUDRY, M. and LYNCH, G. (1980a). Hypothesis regarding the cel lular mechan-isms responsible for long-term synaptic potentiation in the hippocampus. Exp.-Neurol. 68: 202-204. BAUDRY, M. and LYNCH, G. (1980b). Regulation of hippocampal glutamate recep-tors: evidence for the involvement of a calcium-activated protease. Proc. Natl ; Acad. Sci .• (U.S.A.) 77: 2298-2302. BAUDRY, M. and LYNCH, G. (1981). Characterization of two [Vjglutamate binding sites in rat hippocampal membranes. J;-Neurochem; 36: 811-820. BAUDRY, M . , OLIVER, M. , CREAGER, R., WIERASZKO, A. and LYNCH, G. (1980). Increase in glutamate receptors following repetitive electr ical stimu-lation in hippocampal s l i ces . Life S c i : 27: 325-330. BAUDRY, M. , SMITH, E. and LYNCH, G. (1981b). Influences of temperature, detergents and enzymes on glutamate receptor binding and i ts regu-lation by calcium in rat hippocampal membranes. MoT.- Pharmacol; 20: 280-286. ~ BAXTER, D. A . , BITTNER, G. D. and BROWN, T. H. (1985). Quantal mechanism of long-term synaptic potentiation. Proc- NatT.- - Acad: - -Sci.- - USA 82: 5978-5982. — BENJAMIN, A. M . , GOH, J . W. and SASTRY, B. R. (1983). Some neurochemical sequel of long-term potentiation (LTP) of synaptic transmission in the hippocampus (HP) of the rat . J . Neurochem. 41 -(Suppl): S63. BENJAMIN, A. M . , GOH, J . W. and SASTRY, B. R. (1984). Methylation of hippocampal phosphatidylethanolamine and proteins during long-lasting potentiation. Experienti a 41: 278-280. BENNINGER, C , KADIS, J . and PRINCE, D. A. (1980). Extracellular calcium and potassium changes in hippocampal s l i ces . Brain Res.- 187: 165-182. BERGER, T. W. (1984). Long-term potentiation of hippocampal synaptic trans-mission affects rate of behavioral learning. Science 224: 627-630. BERGER, T. W., ALGER, B. E. and THOMPSON, R. F. (1976). Neuronal substrates of classical conditioning in the hippocampus. Science 192: 483-485. BERGER, T. W. and THOMPSON, R. F. (1978). Neuronal p l a s t i c i t y in the limbic system during classical conditioning of the rabbit nict i tat ing membrane response. I. The Hippocampus, firain-Res: 145: 323-346. - 156 -BEST, M. R. and BEST, P. J . (1976). The effects of latent inhibition on hippocampal unit ac t ivi ty in the rat . Exp.-Neurol; 51: 546-573. BISCOE, T. J . and STRAUGHAN, D. w. (1966). Micro-electrophoretic studies on neurones in the cat hippocampus. J . Physiol. (Lond;) 183: 341-359. BJORKLUND, A . , NOBIN, A. and STENEVI, U. (1973). The use of neurotoxic dihydrotryptamines as tools for morphological studies and localized lesioning of central indolamine neurones. Z; Zel lforsch. mikrosk; Anat; 145: 579-601. BLACKSTAD, T. W. (1956). Commissural connections of the hippocampal region in the rat , with special reference to their mode of termination. J ;  Comp. Neurol. 105: 417-537. BLACKSTAD, T. W. (1958). On the termination of some afferents to the hippo-campus and fascia dentata. Acta Anat. 35: 202-214. BLACKSTAD, T. W., BRINK, K . , HEM, J . and JEUNE, B. (1970). Distribution of hippocampal mossy fibers in the rat : an experimental study with si lver impregnation methods. J ; Comp. Neurol. 138: 433-450. BLACKSTAD, T. W., FUXE, K. and H0KFELT, T. (1967). Noradrenaline nerve terminals in the hippocampal region of the guinea pig . Z. Zellforsch;  mikrosk. Anat. 78: 463-473. BLACKSTAD, T. W. and KJAERHEIM, A. (1961). Special axo-dendritic synapses in the hippocampal cortex: electron and light microscopic studies on the layer of mossy f ibers . J ; Comp;-Neurol. 117: 133-159. BLAND, B. H . , K0ST0P0UL0S, G. K. and PHILLIS, J . W. (1974). Acetylcholine sensi t iv i ty of hippocampal formation neurons. Can. - - J . Physiol ;  Pharmac. 52: 966-971. BLISS, T. V. P. , DOUGLAS, R. M . , ERRINGTON, M. L. and LYNCH, M. (1985). Correlation between release of endogenous glutamate and long-term potentiation in the dentate gyrus of anaesthetized rats. J . -Physiol;  (Lond.) 361: 50P. BLISS, T. V. P. and GARDNER-MEDWIN, A. R. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the unanaesthetized rabbit following stimulation of the perforant path. 3. Physiol;  (Lond.) 232: 357-374. BLISS, T. V. P. , GODDARD, G. V. and RIIVES, M. (1983). Reduction of long-term potentiation in the dentate gyrus of the rat following selective depletion of monoamines. J . Physiol . (Lond.) 334: 475-491. - 157 -BLISS, T. V. P . , GODDARD, G. V . , ROBERTSON, H. A. and SUTHERLAND, R. J . (1981). Noradrenaline depletion reduces long-term potentiation in the rat hippocampus. In: 0. Feher and F. Joo (Eds) Cellular Analogues-of Conditioning and Neural P l a s t i c i t y . Akademiai Kiado, Budapest. pp. 175-185. ~ ~ BLISS, T. V. P. and LiJMO, T. (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J . Physiol: (Lond.) 232: 331-356. BOWERY, N. G . , DOBLE, A . , HILL, D. R., HUDSON, A. L . , SHAW, T. S. and TURNBULL, M. J . (1979). Baclofen: a selective agonist for a novel type of GABA receptor. Br.-J.-Pharmac. 67: 444. BOWERY, N. G . , HILL, D. R., HUDSON, A. L . , DOBLE, A . , MIDDLEMISS, D. N . , SHAW, T. and TURNBULL, M. (1980). Baclofen decreases neurotransmitter release in the mammalian CNS by an action at a novel GABA receptor. Nature (Lond.) 283: 92-94. BOWERY, N. G . , DOBLE, A . , HILL, D. R., HUDSON, A. L . , SHAW, T. S. , TURNBULL, M. J . and WARRINGTON, R. (1981). Bicuculline-insensitive GABA receptors on peripheral autonomic nerve terminals. Eur; J.-Pharmac; 71: 53-70. BRIGGS, C. A . , BROWN, T. H. and MCAFEE, D. A. (1985). Neurophysiology and pharmacology of long-term potentiation in the rat sympathetic ganglion. J . Physiol. (Lond.) 359: 503-521. BROWN, T. H. and JOHNSTON, D. (1983). Voltage-clamp analysis of mossy fiber synaptic input to hippocampal neurons. J;-Neurophysiol; 50: 487-507. BROWN, T. H. and MCAFEE, D. A. (1982). Long-term synaptic potentiation in the superior cervical ganglion. Science 215: 1411-1413. BRUCKE, F . , GOGOLAK, G. and STUMPF, C. (1963). Midroelectrodenumterschung der Reizantwort und der Zel l ta t igkei t im Hippocampus bei Septumreizung. Pflugers Arch; ges. - Physiol; 276: 456-470. BUZSAKI, G. and EIDELBERG, E. (1981). Commissural projection to the dentate gyrus of the rat : evidence for feed-forward inhibi t ion . Brain Res:  230: 346-350. BUZSAKI, G. and EIDELBERG, E. (1982). Direct afferent excitation and long-term potentiation of hippocampal interneurons. J ; Neurophysiol; 48: 597-607. ~ BUZSAKI, G . , LEUNG, L. S. and VANDERWOLF, C. H. (1983). Cellular bases of hippocampal EEG in the behaving rat . Brain -Res; Rev.- 6_: 139-171. CASTELLUCCI, V. and KANDEL, E. R. (1976). Presynaptic f a c i l i t a t i o n as a mechanism for behavioral sensitization in Aplysia. Science 194: 1176-1178. - 158 -CASTELLUCCI, V . , PINSKER, H . , KUPFERMANN, I. AND KANDEL, E. R. (1970). Neu-ronal mechanisms of habituation and dishabituation of the gill-withdrawal reflex in Aplysia. Science 167: 1745-1748. CHANG, F-L. F. and GREENOUGH, W. T. (1984). Transient and enduring morpho-logical correlates of synaptic ac t ivi ty and efficacy change in the rat hippocampal s l i c e . Brain Res. 309: 35-46. CHIRWA, S., GOH, J . , MARETIC, H . , MURALIMOHAN, P. and SASTRY, B. (1984). Effects of L-glutamate (G) on rat hippocampal CA^ population spike (PS). IUPHAR Ninth-Int.Congr.-Pharmacol. (Abstract) No: 136. CHIRWA, S. S. , GOH, J . W., MARETIC, H. and SASTRY, B. R. (1983). Evidence for a presynaptic role in long-term potentiation in the rat hippo-campus. J . Physiol ; (Lond;) 339: 41P. CHRONISTER, R. B. and ZORNETZER, S. F. (1973). The fiber connections of allocortex: the subiculum hippocampi. Anat; Rec. 175: 292. CHUNG, S-H. (1977). Synaptic memory in the hippocampus. Nature-(Lond;)  266: 677-678. COLEMAN, J . R. and LINDSLEY, D. B. (1977). Behavioral and hippocampal electr ical changes during operant learning in cats and effects of stimulating two hypothalamic-hippocampal systems. Electroenceph.- C l i n ;  Neurophysiol. 42: 309-331. COLLINGRIDGE, G. L. (1985). Long term potentiation in the hippocampus: mechanisms of in i t ia t ion and modulation by neurotransmitter. Trends  Pharmacol; Sci• 6: 407-411. COLLINGRIDGE, G. L . , KEHL, S. J . and MCLENNAN, H. (1983a). The antagonism of amino-acid induced excitations of rat hippocampal CAi neurones in  v i t r o . J ; Physiol. (Lond.) 334: 19-31. COLLINGRIDGE, G. L . , KEHL, S. J . and MCLENNAN, H. (1983b). Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. J ; Physiol . (Lond;) 334: 33-46. COLLINGRIDGE, G. L . , KEHL, S. J . and MCLENNAN, H. (1984). The action of some analogues of the excitatory amino acids in the dentate gyrus of the rat . Can; J ; Physiol. Pharmac; 62: 424-429. CORRADETTI, R., MONETI, G . , MORONI, F . , PEPEU, G. and WIERASZKO, A. (1983). Electr ical stimulation of the stratum radiatum increases the release and neosynthesis of aspartate, glutamate and y-aminobutyric acid in rat hippocampal s l i ces . J . Neurochem.; 41: 1518-1525. CRAGG, B. G. and HAMLYN, L. H. (1957). Some commissural and septal connec-tions of the hippocampus in the rabbit. J.- - Physiol. (Lond.) 135: 460-485. - 159 -CRAWFORD, I. L. and CONNOR, J . D. (1973). Localization and release of glutamic acid in relation to the hippocampal mossy f ibre pathway. Nature (Lond.) 244: 442-443. CRUTCHER, K. A . , MADISON, R. and DAVIS, J . N. (1981). A study of the rat septohippocampal pathway using anterograde transport of horseradish peroxidase. Neuroscience _6: 1961-1973. CURTIS, D. R., FELIX, D. and MCLENNAN, H. (1970). GABA and hippocampal i n h i b i t i o n . Br.- J . Pharmac. 40: 881-883. CURTIS, D. R., GAME, C. J . A. and LODGE, D. (1976). The in vivo inactiva-tion of GABA and other inhibitory amino acids in the cat central nervous system. Exp. Brain-Res. 25: 413-428. CURTIS, D. R., LODGE, D. and BRAND, S. J . (1977). GABA and spinal afferent terminal exc i tabi l i ty in the cat. Brain Res. 130: 360-363. DAVIES, L . P. and JOHNSTON, G. A. R. (1976). Uptake and release of D- and L-aspartate by rat brain s l i c e s . J . Neurochem; 26: 1007-1014. DEADWYLER, S. A . , WEST, J . R., COTMAN, C. W. and LYNCH, G. S. (1975). A neurophysiological analysis of commissural projections to dentate gyrus of the rat . J . Neurophysiol. 38: 167-184. DEL CASTILLO, J . and KATZ, B. (1954). Stat is t ical factors involved in neuromuscular f a c i l i t a t i o n and depression. J . Physiol.- (Lond:) 124: 574-585. DESMOND, N. L. and LEVY, W. B. (1981). Ultrastructural and numerical alterations in dendritic spines as a consequence of long-term poten-t i a t i o n . Anat Rec. 199: 68A-69A. DESMOND, N. L. and LEVY, W. B. (1984). LTP-associated changes in post-synaptic density length and the length of the synaptic interface in the dentate,gyrus. Soc. Neurosci.- Abstr; 10: 77. DINGLEDINE, R. (1983a). Excitatory amino acids: modes of action on hippo-campal pyramidal c e l l s . Fed. Proc: 42: 2881-2885. DINGLEDINE, R. (1983b). N-Methyl-aspartate activates voltage-dependent c a l -cium conductance in rat hippocampal pyramidal c e l l s . J ; -Physiol.  (Lond.) 343: 385-405. DI LAURO, A . , SCHMID, R. W. and MEEK, J . L. (1981). Is aspartic acid the neurotransmitter of the perforant pathway? Brain Res. 207: 476-480. DOLPHIN, A. C , ERRINGTON, M. L. and BLISS, T. V. P. (1982). Long-term potentiation of the perforant path in vivo is associated with increased glutamate release. Nature (Lond.) ~7$T. 4"96-497. - 160 -DOUGLAS, R. M. and GODDARD, G. V. (1975). Long-term potentiation of the perforant path-granule cel l synapse in the rat hippocampus. Brain Res. 86: 205-215. DOUGLAS, R. M . , GODDARD, G. V. and RIIVES, M. (1982). Inhibitory modula-tion of long-term potentiation: evidence for a postsynaptic locus of control . Brain Res. 240: 259-272. DUDAR, J . D. (1974). In vitro excitation of hippocampal pyramidal c e l l dendrites by glutamic acid. Neuropharmacology 13: 1083-1089. DUDAR, J . D. (1975). The effect of septal nuclei stimulation on the release of acetylcholine from the rabbit hippocampus. Brain Res; 83: 123-133. ~ DUNWIDDIE, T. and LYNCH, G. (1978). Long-term potentiation and depression of synaptic responses in the rat hippocampus: localization and f r e -quency dependency. J ; Physiol. (Lond.) 276: 353-367. DUNWIDDIE, T. and LYNCH, G. (1979). The relationship between extracellular calcium concentration and the induction of hippocampal long-term poten-t i a t i o n . Brain Res. 169: 103-110. ECCLES, J . C. (1964). Presynaptic inhibit ion in the spinal cord. In: J . C. Eccles and J . P. Schade (Eds) Physiology-of Spinal Neurons. Elsevier, Amsterdam, pp. 65-91. ECCLES, J . C. and KRNJEVIC, K. (1959a). Potential changes recorded inside primary afferent fibres within the spinal cord. J ; Physiol.- (Lond;)  149: 250-273. ECCLES, J . C. and KRNJEVIC, K. (1959b). Presynaptic changes associated with post-tetanic potentiation in the spinal cord. J ; Physiol. - (Lond.) 149: 274-287. ECCLES, J . C , SCHMIDT, R. F. and WILLIS, W. D. (1963a). Depolarization of central terminals of group lb afferent fibers of muscle. J ;  Neurophysiol. 26: 1-27. ECCLES, J . C , SCHMIDT, R. F. and WILLIS, W. D. (1963b). Pharmacological studies on presynaptic inhibi t ion . J ; Physiol;-(Lond;) 168: 500-530. ENNA, S. J . and SNYDER, S. H. (1975). Properties of y-aminobutyric acid (GABA) receptor binding in rat brain synaptic membrane fractions. Brain Res. 100: 81-97. FAGNI, L . , BAUDRY, M and LYNCH, G. (1983). Classif icat ion and properties of acidic amino acid receptors in hippocampus. I. Electrophysiologi-cal studies of an apparent desensitization and interactions with drugs which block transmission. J . Neurosei; 3: 1538-1546. - 161 -FIFKOVA, E. and VAN HARREVELD, A. (1977). Long-lasting morphological changes in dendritic spines of dentate granular cel ls following stimu-lation of the entorhinal area. J . Neurocytol; 6_: 211-230. FINN, R. C , BROWNING, M. and LYNCH, G. (1980). Trifluoperazine inhibits long-term potentiation and the phosphorylation of a 40,000 dalton pro-te in . Neurosci: Lett . 19: 103-108. FONNUM, F. and WALAAS, I. (1978). The effect of intrahippocampal kainic acid injections and surgical lesions on neurotransmitters in hippo-campus and septum. J . Neurochem: 31: 1173-1181. FOX, S. E. and RANCK, J . B. JR., (1981). Electrophysiological character-is t i cs of hippocampal cortex-spike cel ls and theta c e l l s . Exp. - Brain  Res. 41: 399-410. FRITZ, L. C. and GARDNER-MEDWIN, A. R. (1976). The effect of synaptic activation on the extracellular potassium concentration in the hippo-campal dentate area, in -vi tro ; Brain Res; 112: 183-18. FUXE, K. (1965). Evidence for the existence of monoamine containing neurons in the central nervous system. IV. The distribution of mono amine terminals in the central nervous system. Acta physiol.- - Scand; 64 (Suppl) 247: 39-85. — FUXE, K . , HOKFELT, T . , JONSSON, G. and UNGERSTEDT, U. (1970). Fluorescence microscopy in neuroanatomy. In: W. J . H. Nauta and S. 0. E. Ebbesson (Eds) Contemporary - Research Methods in Neuroanatomy: Springer Verlag, Berl in , pp. 2/5-314. FUXE, K . , HOKFELT, T. and UNGERSTEDT, U. (1969). Distribution of monoamines in the mammalian central nervous system by histochemical studies. In: G. Hooper (Ed) Metabolism of-Amines in the Brain. Macmillan, London, pp. 10-22. FUXE, K. and JONSSON, G. (1974). Further mapping of central 5-hydroxy-tryptamine neurons: studies with the neurotoxic dihydroxy-trypta-mine. Adv.- -Biochem. Psychopharmac; 10: 1-12. GARTHWAITE, J . (1985). Cellular uptake disguises action of L-glutamate on N-methyl-D-aspartate receptors. Br. J . Pharmac. 85: 297-307. GMELIN, G. W. (1978). Electrophoretic studies on presynaptic inhibition in the mammalian spinal cord. In: R. W. Ryall and J . S. Kelly (Eds) Iontophoresis and- - Transmitter- Mechanisms-^ -in- - the - Mammalian -Central  Nervous System: Elsevier, Amsterdam, pp. 267-269. GODDARD, G. V . , MCINTYRE, D. C. and LEECH, C. K. (1969). A permanent change in brain function resulting from daily electr ical stimulation. Exp.-Neurol. 26: 295-330. - 162 -GOH, J . W. and SASTRY, B. R. (1983). Effects of localized applications of L-glutamate on the population spike in the hippocampal s l ice prepara-t ion. Life S c i . 33: 1673-1678. GOH, J . W. and SASTRY, B. R. (1984). Long-lasting potentiation in hippo-campus is not due to an increase in glutamate receptors. Life S c i . 34: 1497-1501. GOLGI, C. (1886). Sulla f ina anatomia degli organi centrali-del sistema  nervoso; Milano. GOTTLIEB. D. T. and COWAN, W. M. (1973). Autoradiographic studies of the commissural and ipsi lateral association connections of the hippocampus and dentate gyrus of the rat . J ; Comp; Neurol; 149: 393-422. GRAY, J . A. (1977). Drug effects on fear and frustration: possible limbic si te of action of minor t ranqui l l i sers . In: L . Iversen, S. Iversen and S. Snyder (Eds) Handbook of Psychopharmacology; V o l ; - 8 ; Plenum Press, New York. pp. 433-515. GREEN, J . D. (1964). The Hippocampus. Physiol; -Rev; 44: 561-608. GREEN J . D. and ARDUINI, A. (1954). Hippocampal electr ical act ivi ty in arousal. J;-Neurophysiol; 17: 533-557. GRIFFITH, W. H . , BROWN, T. H. and JOHNSTON, D. (1986). Voltage-clamp analy-sis of synaptic inhibition during long-term potentiation in hippocampus. J . Neurophysiol; 55: 767-775. GUSTAFSSON, B. and WIGSTROM, H. (1981). Evidence for two types of after-hyperpolarization in CAi pyramidal cel ls in the hippocampus. Brain Res. 206: 462-468. HAAS, H. L. and ROSE, G. (1982). Long-term potentiation of excitatory synaptic transmission in the rat hippocampus: the role of inhibitory processes. J ; -Physiol ; - (Lond.) 329: 541-552. HAAS, H. L. and ROSE, G. (1984). The role of inhibitory mechanisms in hippo-campal long-term potentiation. Neurosci - Lett ; 47: 301-306. HABLITZ, J . J . (1981). Effects of intracellular injections of chloride and EGTA on postepileptiform-burst hyperpolarizations in hippocampal neurons. Neurosci. Lett ; 22: 159-163. HAGIWARA, S. and TASAKI, I. (1958). I. A study on the mechanism of impulse transmission across the giant synapse of the squid. J ; -Physiol.- (Lond.) 143: 114-137. HAMBERGER, A . , CHIANG, G . , NYLEN, E. S. , SCHEFF, S. W. and COTMAN, C. W. (1978). Stimulus evoked increase in the biosynthesis of the putative neurotransmitter glutamate in the hippocampus. Brain Res; 143: 549-555. - 163 -HAMBERGER, A. C , CHIANG, G. H . , NYLEN, E. S. , SCHEFF, S. W. and COTMAN, C. W. (1979a). Glutamate as a CNS transmitter. I. Evaluation of glucose and glutamine as precursors for the synthesis of preferen-t i a l l y released glutamate. Brain Res: 168: 513-530. HAMBERGER, A . , CHIANG, G. H . , SANDOVAL, E. and COTMAN, C. W. (1979b). Glutamate as a CNS transmitter. II . Regulation of synthesis in the releasable pool. Brain-Res. 168: 531-541. HAMLYN, L. H. (1961). Electron microscopy of mossy f ibre endings in Amnion's horn. Nature (Lond;) 190: 645-646. HAMLYN, L. H. (1962). The fine structure of the mossy f ibre endings in the hippocampus of the rabbit. J . Anat:-(Lond.) 96: 112-126. HAMLYN, L. H. (1963). An electron microscope study of pyramidal neurones in the Ammon's horn of the rabbit. J ; Anat. (Lond.) 97: 189-201. HARRIS, E. W., GANONG, A. H. and COTMAN, C. W. (1984). Long-term potentia-tion in the hippocampus involves activation of N-methyl-D-aspartate receptors. Brain-Res; 323: 132-137. HEBB, D. 0. (1949). The-Organization of Behavior; Wiley, New York. HERZ, A. and NACIMIENTO, A. C. (1965). Uber die Wirkung von Pharmaka auf Neurone des Hippocampus nach microelectrophoretischer Verabfolgung. Naunyn-Sehmiedeberg's Arch.- Pharmacol: 251: 295-314. HJORTH-SIMONSEN, A. (1973). Some intr insic connections of the hippocampus in the rat : an experimental analysis. J ; -Comp: Neurol; 147: 145-162. HJORTH-SIMONSEN, A. and JEUNE, B. (1972). Origin and termination of the hippocampal perforant path in the rat studied by si lver impregnation. J . -Comp; Neurol; 144: 215-232. HJORTH-SIMONSEN, A. and LAURBERG, S. (1977). Commissural connections of the dentate area in the rat . J . Comp. Neurol. 174: 591-605. HORWITZ, B. (1981). Neuronal p l a s t i c i t y : how changes in dendritic archi-tecture can affect the spread of postsynaptic potentials. Brain-Res-.-224: 412-418. HOTSON, J . R. and PRINCE, D. A. (1980). A calcium activated hyperpolari-zation follows repetitive f i r i n g in hippocampal neurons. <h Neurophysiol. 43: 409-419. HUBBARD, J . I. and WILLIS, W. D. (1962). Reduction of transmitter output by depolarization. Nature (bond.) 193: 1294-1295. ISAACSON, R. L. (1974). Thebimbic System; Plenum Press, New York. - 164 -ISSEROFF, A . , LEVETON, L . , FREEMAN, G . , LEWIS, M. and STEIN, D. G. (1976). Differences in the behavioral effects of single-stage and serial lesions of the hippocampus. Exp.-Neurol; 53: 339-354. IVERSEN, L. L. and BLOOM, F. E. (1972). Studies on the uptake of [3H]GABA and [ 3H]glycine in slices and homogenates of rat brain and spinal cord by electron microscopic autoradiography. Brain Res; 41: 131-143. IVERSEN, L. L. and STORM-MATHISEN, J . (1976). Uptake of [3H]glutamic acid in excitatory nerve endings in the hippocampal formation of the rat . Acta physio1. Scand. 96: 22A-23A. JARRARD, L. E. (1983). Selective hippocampal lesions and behavior: effects of kainic acid lesions on performance of place and cue tasks. Benav.- Neurosci; 97: 873-889. JARRARD, L. E. (1985). Is the hippocampus real ly involved in memory? In: B. E. W i l l , P. Schmitt and J . C. Dalrymple-Alford (Eds) Brain  P l a s t i c i t y ; Learning-and Memory; Plenum Press, New York. pp. 363-372. JOHNSTON, D. and BROWN, T. H. (1983). Interpretation of voltage-clamp measurements in hippocampal neurons. J ; Neurophysiol; 50: 464-486. JOHNSTON, D. , HABLITZ, J . J . and WILSON, W. A. (1980). Voltage clamp d i s -closes slow inward current in hippocampal burst - f i r ing neurones. Nature (Lond.) 286: 391-393. KANDEL, E. R. and SPENCER, W. A . (1961). Electrophysiology of hippocampal neurons. II. After potentials and repetitive f i r i n g . J.-Neurophysiol; 24: 243-259. KANDEL, E. R., SPENCER, W. A. and BRINLEY, F. J . JR. (1961). Electrophysio-logy of hippocampal neurons. I. Sequential invasion and synaptic organization. J;-Neurophysiol. 24: 225-242. KASAMATSU, Y . , TOYAMA, K, MAEDA, J . and SAKAGUCHI, H. (1981). Long-term potentiation investigated in a s l ice preparation of striate cortex of young kittens. Neurosci.- Lett ; 26: 269-274. KIMURA, H. MCGEER, P. L . , PENG, J . H. and MCGEER, E. G. (1981). The central cholinergic system studied by choline acetyltransferase immunohisto-chemistry in the cat. J;-Comp.-Neurol; 200: 151-200. KING, G. L. and DINGLEDINE, R. (1985). Micromolar and nanomolar concentra-tions of D-2-amino-5-phosphonovalerate have opposite effects on epilep-tiform burst - f i r ing in the rat hippocampal s l i c e . Soc. Neurosci.- Abstr. Ij.: 103. KNOWLES, W. D. and SCHWARTZKROIN, P. A. (1981). Local c i r c u i t synaptic interactions in hippocampal brain s l i c e s . J ; -Neurosci ; 1: 318-322. - 165 -KOERNER, J . F. and COTMAN, C. W. (1982). Response of Schaffer c o l l a t e r a l -CA^ pyramidal ce l l synapses of the hippocampus to analogues of acidic amino acids. Brain Res; 251: 105-115. KOHLER, C. and STEINBUSCH, H. (1982). Identification of serotonin containing neurons in the midbrain raphe projecting to the entorhinal area and the hippocampal formation: a combined immuno-histochemical and fluorescent retrograde tracing study in the rat brain. Neuroscience 7_: 951-975. KOYANO, K . , KUBA, K. and MINOTA, S. (1985). Long-term potentiation of trans-mitter release induced by repetitive presynaptic ac t ivi t ies in bul l - f rog sympathetic ganglia. J : Physiol. (Lond.) 359: 219-233. KRNJEVIC, K. and LISIEWICZ, A. (1972). Injections of calcium ions into spinal motoneurones. J ; Physiol; (Lond.) 225: 363-390. KRNJEVI6, K . , MORRIS, M. E. and REIFFENSTEIN, R. J . (1980). Changes in extracellular Ca^ + and K + ac t ivi ty accompanying hippocampal d i s -charges. Can; J ; Physiol; Pharmac. 58: 579-583. KRNJEVIC, K . , MORRIS, M. E. and REIFFENSTEIN, R. J . (1982). Stimulation-evoked changes in extracellular K + and C a + + in pyramidal layers of the rat ' s hippocampus. Can; J.-Physiol:-Pharmac: 60: 1643-1657. KRNJEVIC, K. and PHILLIS, J . W. (1963). Iontophoretic studies on neurones in the mammalian cerebral cortex. J ; -Physiol;-(Lond;) 165: 274-304. KRNJEVIC, K . , REIFFENSTEIN, R. J . and ROPERT, N. (1980). Disinhibitory action of acetylcholine in the hippocampus. J-.• Phys-ioT. - (Lond.) 308: 73P-74P. KRNJEVIC, K. and ROPERT, N. (1981). Septo-hippocampal pathway modulates hippocampal act ivity by a cholinergic mechanism. Can.- -J.- • Phys-iol.- Pharmac. 59: 911-914. KUHAR, M. J . , AGHAJANIAN, G. K. and ROTH, R. H. (1972). Tryptophan hydroxylase act ivi ty and synaptosomal uptake of serotonin in discrete brain regions after midbrain raphe lesions: correlations with serotonin levels and histochemical fluorescence. Brain -Res: 44: 165-176 KVAMME, E. (1981). Evidence for glutamate compartmentation in the synapto-somal fraction using the glutamate sensi t ivi ty of phosphate activated glutaminase as functional test . Eighth- ISN Meeting Abstr; Nottingham. 117. LANGMOEN, I. A . , ANDERSEN, P, GJERSTAD, L, MOSFELDT-LAURSEN, A. and GANES, T. (1978). Two separate effects of GABA on hippocampal pyramidal cel ls in  v i t r o . Acta physiol . Scand. 102: 28A. - 166 -LAROCHE, S. (1985). What can the long-term potentiation procedure t e l l us about the neural, mechanisms of learning and memory? In: B. E. W i l l , P. Schmitt and J . C. Dalrymple-Alford (Eds) Brain P l a s t i c i t y ; Learning- and  Memory; Plenum Press, New York. pp. 139-155. LEE, K. S. (1983a). Cooperativity among afferents for the induction of long-term potentiation in the CA^ region of the hippocampus. J .  Neurosci. 3_: 1369-1372. LEE, K. S. (1983b). Sustained modification of neuronal act ivi ty in the hippocampus and neocortex. In: W. Seifert (Ed) Neurobiology of- the  Hippocampus; Academic Press, London, pp. 265-272. LEE, K. S. , SCHOTTLER, F . , OLIVER, M. and LYNCH, G. (1980). Brief bursts of high frequency stimulation produce two types of structural change in rat hippocampus. J.-Neurophysiol; 44: 247-258. LEWIS, P. R. and SHUTE, C. C. D. (1967). The cholinergic limbic system: projections to hippocampal formation, medial cortex, nuclei of the ascending cholinergic reticular system and the subfornical organ and supraoptic crest. Brain 90: 521-540. LEWIS, P. R., SHUTE, C. C. D. and SILVER, A. (1967). Confirmation from choline acetylase analyses of a massive cholinergic innervation to the rat hippocampus. 3 ; -Physiol ; (Lond.) 191: 215-224. LINDVALL, 0. and BJORKLUND, A. (1974). The organization of the ascending catecholamine neuron systems in the rat brain. Acta - physio!. Scand; 412: 1-48. LLOYD, D. P. C. (1949). Post-tetanic potentiation of response in mono-synaptic reflex pathways of the spinal cord. J . gen. Physiol. 33: 147-170. — LJ0MO, T. (1971). Potentiation of monosynaptic EPSPs in the perforant path-dentate granule cel l synapse. Exp. Brain -Res; 12: 46-63. LORENTE DE NO, R. (1934). Studies on the structure of the cerebral cortex. II. Continuation of the study of the Ammonic system. J_i Psychol; Neurol. 46: 113-177. LOWRY, 0. H . , ROSEBROUGH, N. J . , FARR, A. L. and RANDALL, R. J . (1951). Protein measurement with the Folin phenol reagent. J . B i o l . Chem. 193: 265-275. LOY, R., KOZIELL, D. A . , LINDSEY, J . D. and MOORE, R. Y. (1980). Noradren-ergic innervation of the adult rat hippocampal formation. J . - -Comp. Neurol; 189: 699-710. - 167 -LYNCH, M. A . , ERRINGTON, M. L. and BLISS, T. V. P. (1985). Long-term potentiation and the sustained increase in glutamate release which follow tetanic stimulation of the perforant path are both blocked by D(-)aminophosphonovaleric acid. Soc;-Neurosci; Abstr. 11: 834. LYNCH, G. S. , GRIBKOFF, V. K. and DEADWYLER, S. A. (1976). Long-term potentiation is accompanied by a reduction in dendritic responsiveness to glutamic acid. Nature (Lond.) 263: 151-153. LYNCH, G . , HALPAIN, S. and BAUDRY, M. (1982). Effects of high-frequency synaptic stimulation on glumate receptor binding studied with a modified in vitro hippocampal s l i ce preparation. Brain-Res; 244: 101-111. LYNCH, G . , LARSON, J . , KELSO, S., BARRIONUEVO, G. and SCHOTTLER, F. (1983). Intracellular injections of EGTA block induction of hippocampal long-term potentiation. Nature (Lond.) 305: 719-721. LYNCH, G . , ROSE, G. and GALL, C. (1978). Anatomical and functional aspects of the septo-hippocampal projections. In: Functions- of the Septo- Hippocampal - - System. Ciba Foundation Symposium^ No": 5TL Elsevier-North Holland, Amsterdam, pp. 5-20. MACVICAR, B. A. (1984). Voltage-dependent calcium channels in g l i a l c e l l s . Science 226: 1345-1347. MACVICAR, B. A. and DUDEK, F. E. (1981). Electrotonic coupling between pyramidal c e l l s : a direct demonstration in rat hippocampal s l i c e s . Science 213: 782-785. MACVICAR, B. A. and DUDEK, F. E. (1982). Electrotonic coupling between granule cel ls of rat dentate gyrus: physiological and anatomical evidence. J ; Neurophysiol. 47: 579-592. MAGLEBY, K. L . and ZENGEL, J . E. (1975). A quantitative description of tetanic and post-tetanic potentiation of transmitter release at the frog neuromuscular junction. J . Physiol. (Lond.) 245: 183-208. MAGLEBY, K. L. and ZENGEL, J . E. (1976). Augmentation: a process that acts to increase transmitter release at the neuromuscular junction of the frog. J ; Physiol. (Lond.) 257: 449-470. MALENKA, R. C , MADISON, D. V. and NICOLL, R. A. (1986). Potentiation of synaptic transmission in the hippocampus by phorbol esters. Nature  (Lond.) 321: 175-177. MALINOW, R. and MILLER, J . P. (1986). Postsynaptic hyperpolarization during conditioning reversibly blocks induction of long-term potentiation. Nature (Lond;) 320: 529-530. - 168 -MALTHE-S0RENSSEN, D . , SKREDE, K. K. and FONNUM, F. (1979). Calcium depen-dent release of D-[ 3H]-aspartate evoked by selective electrical stimulation of excitatory afferent fibres to hippocampal pyramidal cells in v i t ro . Neuroscience 4_: 1255-1265. MARETIc", H . , GOH, J . W., CHIRWA, S. S. and SASTRY, B. R. (1984). B a + + induces a long-lasting potentiation (LLP) of the hippocampal CA\ population spike. Pacific Cascade Soc. Neurosci:-Abstr. MAYER, M. L . , WESTBROOK, G. L. and GUTHRIE, P. B. (1984). Voltage-depen-dent block by Mg^+ of NMDA responses in spinal cord neurones. Nature  (Lond.) 309: 261-263. MCDOUGAL, D. B. JR. and OSBORN, L. A. (1976). Post-tetanic hyperpolariza-t ion, sodium-potassium activated adenosine triphosphatase and high energy phosphate levels in garfish olfactory nerve. J . Physiol. (Lond;)  256: 41-60. MCGINTY, J . F . , HENRICKSEN, S. J . , GOLDSTEIN, A . , TERENIUS, L . and BLOOM, F. E. (1983). Dynorphin is contained within hippocampal mossy f ibers : immunochemical alterations after kainic acid administration and colchicine-induced neurotoxicity. Proc. - Natl : Acad.-- S c i . - USA 80: 589-593. MCNAUGHTON, B. L. (1982). Long-term synaptic enhancement and short-term potentiation in rat fascia dentata act through different mechanisms. J . Physiol ; (Lond.) 324: 249-262. MCNAUGHTON, B. L . , DOUGLAS, R. M. and GODDARD, G. V. (1978). Synaptic enhancement in fascia dentata: co-operativity among coactive afferents. Brain Res. 157: 277-293. MELLGREN, S. I. and SREBRO, B. (1973). Changes in acetylcholinesterase and the distribution of degenerating fibres in the hippocampal region after septal lesions in the rat . Brain Res. 75: 172-176. MILLER, J . P . , RALL, W. and RINZEL, J . (1985). Synaptic amplification by active membrane in dendritic spines. Brain Res. 325: 325-330. MISGELD, U . , SARVEY, J . M. and KLEE, M. R. (1979). Heterosynaptic post-activation potentiation in hippocampal CA3 neurons: long-term changes of the postsynaptic potential . Exp: Brain-Res; 37: 217-229. MODY, I . , BAIMBRIDGE, K. G. and MILLER, J . J . (1984). Blockade of tetanic-and calcium-induced long-term potentiation in the hippocampal s l ice preparation by neuroleptics. Neuropharmacology 23: 625-631. MOORE, R. Y. and HALARIS, A. E. (1975). Hippocampal innervation by sero-tonin neurons of the midbrain raphe in the rat . J . Comp. Neurol; 152: 163-174. - 169 -MURALI MOHAN, P. and SASTRY, B. R. (1985). Calcium and unit response decrement to local ly applied glutamate on rat hippocampal CAi neurons. Eur. J . Pharmac. 114: 335-342. NADLER, J . V . , COTMAN, C. W. and LYNCH, G. S. (1974). Biochemical p l a s t i -c i ty of short-axon interneurones: increase glutamate decarboxylase act ivi ty in the denervated area of rat dentate gyrus following ento-rhinal lesion. Exp. Neurol; 45: 403-413. NADLER, J . V . , VACA, K. W., WHITE, W. F . , LYNCH, G. S. and COTMAN, C. W. (1976). Aspartate and glutamate as possible transmitters of excitatory hippocampal afferents. Nature-(Lond.) 260: 538-540. NADLER, J . V . , WHITE, W. F . , VACA, K. W., PERRY, B. and COTMAN, C. W. (1978). Biochemical correlates of transmission mediated by glutamate and aspar-tate. J . Neurochem. 31: 147-155. NAFSTAD, P. H. J . (1967). An electron microscope study on the termination of the perforant path fibers in the hippocampus and the fascia dentata. Zei tschri f t - fur Zel1forschung 76: 532-542. NEUMAN, R. S. and HARLEY, C. W. (1983). Long-lasting potentiation of the dentate gyrus population spike by norepinephrine. Brain- Res. 273: 162-165. NEWBERRY, N. R. and NICOLL, R. A. (1984a). A bicucul1ine-resistant i n h i b i -tory post-synaptic potential in rat hippocampal pyramidal cel ls in  v i t r o . J . Physiol.- (Lond.) 348: 239-254. NEWBERRY, N. R. and NICOLL, R. A. (1984b). Direct hyperpolarizing action of baclofen on hippocampal pyramidal c e l l s . Nature (Lond.) 308: 450-429. NICOLL, R. A. and ALGER, B. E. (1981). Synaptic excitation may activate a calcium-dependent potassium conductance in hippocampal pyramidal c e l l s . Science 212: 957-958. NISHI, S. , MINOTA, S. and KARCZMAR, A. G. (1974). Primary afferent neurones: the ionic mechanism of GABA-mediated depolarization. Neuropharmacology IJ : 215-219. NOWAK, L . , BREGESTOVSKI, P . , ASCHER, P . , HERBET, A. and PROCHIANTZ, A. (1984). Magnesium gates glutamate-activated channels in mouse central neurones. Nature (Lond;) 307: 462-465. O'KEEFE, J . (1983). Spatial memory within and without the hippocampal system. In: W. Seifert (Ed) Neurobiology - of - the - Hippocampus. Academic Press, London, pp. 375-403. O'KEEFE, J . and NADEL, L. (1978). The Hippocampus as a Cognitive Map. Oxford University Press, New York. - 170 -OLDS, J . , DISTERHOFT, J . F . , SEGAL, M . , KORNBLITH, C. L. and HIRSH, R. (1972). Learning centers of rat brain mapped by measuring latencies of conditioned unit responses. J ; Neurophysiol. 35: 202-219. OLPE, H-R., BAUDRY, M . , FAGNI, L. and LYNCH, G. (1982). The blocking action of baclofen on excitatory transmission in the rat hippocampal s l i c e . J .  Neurosci. 2: 698-703. PASQUIER, D. A. and REINOSO-SUAREZ, F. (1978). The topographic organiza-tion of hypothalamic and brain stem projections to the hippocampus. Brain Res. B u l l . 3; 373-389. PENFIELD, W. and MILNER, B. (1958). Memory d e f i c i t produced by bilateral lesions in the hippocampal zone. Arch;-Neurol;- Psychiat; 79: 475-497. RAMON Y CAJAL, S. (1911). HistolIogie-du-Systgme nerveux de 1'Homme et-des  Vertebres. V o l ; - 2 ; Paris: Maloine. RAMON Y CAJAL, S. (1968). The -Structure of Amnion1 s -Horn; (Transl. by L. M. Kraft ) . Springfield, Charles C. Thomas. RIBAK, C. E . , VAUGH, J . E. and SAITO, K. (1978). Immunocytochemical l o c a l -ization of glutamic acid decarboxylase in neuronal somata following colchicine inhibition of axonal transport. Brain Res. 140: 315-332. RICHARDSON, T. L . , TURNER, R. W. and MILLER, J . J . (1984). Extracellular f i e l d s influence transmembrane potentials and synchronization of hippo-campal neuronal ac t iv i ty . Brain -Res.- 294: 255-262. ROBERTS, P. J . and WATKINS, J . C. (1975). Structural requirements for the inhibition of L-glutamate uptake by gl ia and nerve endings. Brain Res; 85: 120-125. ROBINSON, G. B. and RACINE, R. J . (1982). Heterosynaptic interactions between septal and entorhinal inputs to the dentate gyrus: long-term potentiation effects . Brain Res; 249: 162-166. ROSS, R. A. and REIS, D. J . (1974). Effects of lesions of locus coeruleus on regional distribution of dopamine-B-hydroxylase act ivi ty in rat brain. Brain -Res; 73: 161-166. SALMOIRAGHI, G. C. and STEFANIS, C. N. (1965). Patterns of central neurons , responses to suspected transmitters. Arch; i t a l ; B i o l ; 103: 705-724. SASTRY, B. R. (1979a). A calcium dependent post-tetanic hyperpolarization of primary afferent terminals. Life S c i ; 25: 1179-1188. SASTRY, B. R. (1979b). y-Aminobutyric acid and primary afferent depolari-zation in fel ine spinal cord. Can; - J ; -Physiol ; Pharmac; 57: 1157-1167. - 171 -SASTRY, B. R. (1979c). Calcium and action potentials in primary afferent terminals. Life S c i ; 24: 2193-2200. SASTRY, B. R. (1982) Presynaptic change associated with long-term poten-t ia t ion in hippocampus. Life S c i . 30: 2003-2008. SASTRY, B. R., CHIRWA, S. S., GOH, J . W., MARETIC, H. and PANDANABOINA, M. M. (1984a). Verapamil counteracts depression but not long-lasting potentia-tion of the hippocampal population spike. Life S c i . 34: 1075-1086. SASTRY, B. R. and GOH, J . W. (1984). Long-lasting potentiation in hippo-campus is not due to an increase in glutamate receptors. Life S c i . 34: 1497-1501. ~ ~ SASTRY, B. R., GOH, J . W. and AUYEUNG, A. (1986). Associative induction of post-tetanic and long-lasting potentiation in CAi neurons of rat hippocampus. Science 232: 988-990. SASTRY, B. R., GOH, J . W. and PANDANABOINA, M. M. (1984b). Verapamil coun-teracts the masking of long-lasting potentiation of hippocampal popula-tion spike produced by 2-amino-5-phosphonovalerate. Life S c i ; 34: 323-329. ~ SAVAGE, D. D. , WERLING, L. L . , NADLER, J . V. and MCNAMARA, J . 0. (1982). Selective increase in L-[ 3H]-glutamate binding to a quisqualate-sensitive site on hippocampal synaptic membranes after angular bundle kindling. Eur.-J;-Pharmac; 85: 255-256. SCHAFFER, K. (1892). Beitrag zur Histologie der Ammonshornformation. Arch; mikrosk;-Anat; 39. SCHARFMAN, H. E. and SARVEY, J . M. (1985). Postsynaptic f i r i n g during repetitive stimulation is required for long-term potentiation in hippo-campus. Brain Res. 331: 267-274. SCHWARTZKROIN, P. A. (1975). Characteristics of CA^ neurons recorded intracel lular ly in the hippocampal in vitro s l i ce preparation. Brain  Res. 85: 423-436. SCHWARTZKROIN, P. A. (1977). Further characteristics of hippocampal CA^ cel ls in v i t r o . Brain Res; 128: 53-68. SCHWARTZKROIN, P. A. and ANDERSEN, P. (1975). Glutamic acid sensi t ivi ty of dendrites in hippocampal slices in vi-tro. In: G. W. Kreutzberg (Ed) Advances in Neurology; Vol.- 12.- Raven Press, New York. pp. 45-51. SCHWARTZKROIN, P. A . , SKREDE, K. K . , STORM, J . and ANDERSEN, P. (1974). Glutamic acid excitation of pyramidal cell dendrites in hippocampal s l i c e s . IUPS 26 t h Int. Congr. Physiol ; Vol -. 11; pp. 391. - 172 -SCHWARTZKROIN, P. A. and STAFSTROM, C. E. (1980). Effects of EGTA on the calcium-activated afterhyperpolarization in hippocampal CA3 pyramidal c e l l s . Science 210: 1125-1126. SCHWARTZKROIN, P. A. and WESTER, K. (1975). Long-lasting f a c i l i t a t i o n of a synaptic potential following tetanization in the in vitro hippocampal s l i c e . Brain Res. 89: 107-119. SCHWERDTFEGER. W. K. (1984). Structure and fiber connections of the hippo-campus. A comparative study. Adv; Anat. Embryo!. Cell B i o l ; 83: 1-69. SCOVILLE, W. B. and MILNER, B. (1957). Loss of recent memory after b i -lateral hippocampal lesions. J ; Neurol. Psychiat; 20: 11-21. SEGAL, M. (1975). Physiological and pharmacological evidence for a seroton-ergic projection to the hippocampus. Brain Res. 94: 115-131. SEGAL, M . , (1976). 5-HT antagonists in rat hippocampus. Brain-Res; 103: 161-166. SEGAL, M. (1980). The action of serotonin in the rat hippocampal s l i ce preparation. J ; Physiol. (Lond;) 303: 423-439. SEGAL, M. (1981). The actions of glutamic acid on neurons in the rat hippo-campal s l i c e . In: G. T. DiChiara and G. L. Gessa (Eds) Glutamate as-a  Neurotransmitter. Raven Press, New York. pp. 217-225. SEGAL, M. and BLOOM, F. E. (1974a). The action of norepinephrine in the rat hippocampus. I. Iontophoretic studies. Brain-Res; 72: 79-97. SEGAL, M. and BLOOM, F. E. (1974b). The action of norepinephrine in the rat hippocampus. II . Activation of the input pathway. Brain Res; 72: 99-114. ~ ~ SEGAL, M. and OLDS, J . (1973). The act ivi ty of units in the hippocampal c i rcui t of the rat during differential classical conditioning. J.- - Comp;  Physiol ; Psychol. 82: 195-204. SHUTE, C. C. D. and LEWIS, P. R. (1961). The use of cholinesterase tech-niques combined with operative procedures to follow nervous pathways in the brain. Bibl . -Anat . 2: 34-49. SIEGEL, A. and TASSONI, J . P. (1971). Differential efferent projections from the ventral and dorsal hippocampus of the cat. Brain Behav.- Evol; 4: 185-200. SIMON, H . , LEMOAL, M. and CALAS, A. (1979). Efferents and afferents of the ventral tegmental-AlO region studied after local injection of i^i) leucine and horseradish peroxidase. Brain -Res. 178: 17-40. - 173 -SKREDE, K. K. and MALTHE-S0RENSSEN, D. (1981). Increased resting and evoked release of transmitter following repetitive electr ical tetanization in the hippocampus: a biochemical correlate to long-lasting synaptic potentiation. Brain Res; 208: 436-441. SKREDE, K. K. and WESTGAARD, R. H. (1971). The transverse hippocampal s l i c e : a well defined cortical structure maintained in v i t r o . Brain  Res. 35: 589-593. SMITH, C. M. (1972). The release of acetylcholine from rabbit hippo-campus. Br. J . Pharmac- 45: 172P. SMITH, C. M. (1974). Acetylcholine release from the cholinergic septo-hippocampal pathway. Life S c i . 20: 2159-2166. SPENCER, W. A. and KANDEL, E. R. (1961a). Electrophysiology of hippo-campal neurons. III. Firing level and time constant. J .  Neurophysiol. 24: 260-271. SPENCER, W. A. and KANDEL, E. R. (1961b). Electrophysiology of hippo-campal neurons. IV. Fast prepotentials. J.- -Neurophysiol-. 24: 272-285. SPENCER, W. A. and KANDEL, E. R. (1961c). Hippocampal neuron responses to selective activation of recurrent collaterals of hippocampal axons. Exp. Neurol. 4: 149-161. STEFANIS, C. (1964). Hippocampal neurons: their responsiveness to micro-electrophoretically administered endogenous amines. Pharmacologist 6: STEIN, D. G . , ROSE, J . J . , GRAZIADEC, J . , MISHKIN, D. and BRINK, 0. (1969). Central nervous system: recovery of function. Science 166: 528-530. STEINBUSCH, H. (1981). Distribution of serotonin immunoreactivity in the central nervous system of the rat -cel l bodies and terminals. Neuroscience 6: 557-618. STEINER, F. A. (1968). Influence of microelectrophoretically applied acetylcholine on the_ responsiveness of hippocampal and lateral geni-culate neurones. Pfl tigers-Arch, ges. Physiol; 303: 173-180. STORM-MATHI SEN, J . (1972). Glutamate decarboxylase in the rat hippocampal region after lesions of the afferent f ibre systems. Evidence that the enzyme is localized in int r ins ic neurones. Brain-Res. 40: 215-235. STORM-MATHISEN, J . (1974). High a f f i n i t y uptake of GABA in presumed GABA-ergic nerve endings in rat brain. Brain -Res.- 84: 409-427. - 174 -STORM-MATHISEN, J . (1976). Distribution of the components of the GABA system in cerebellum and hippocampus. Effects of axotomy. In: E. Roberts, T. N. Chase and D. Tower (Eds) GABA- in Nervous - -System  Function; Raven Press, New York. pp. 149-168. STORM-MATHI SEN, J . (1977a). Localization of transmitter candidates in the brain: the hippocampal formation as a model. Prog. -Neurobiol; 8: 119-181. ~ STORM-MATHI SEN, J . (1977b). Glutamic acid and excitatory nerve endings: reduction of glutamic acid uptake after axotomy. Brain Res. 120: 379-386. STORM-MATHISEN, J . (1978). Localization of transmitter amino acids: application in hippocampus and septum. In: F. Fonnum (Ed) Amino-Acids  as Chemical-Transmitters. Plenum Press, New York, pp 155-173. STORM-MATHISEN, J . and FONNUM, F. (1972). Localization of transmitter candi-dates in the hippocampal region. In: P. B. Bradley and R. W. Brimblecombe (Eds) Biochemical and Pharmacological Mechanisms - Underly- ing-Behaviour; Elsevier, Amsterdam, pp. 41-58. STORM-MATHISEN, J . and GULDBERG, H. C. (1974). 5-Hydroxytryptamine and noradrenaline in the hippocampal region: effect of transection of afferent pathways on endogenous levels , high a f f i n i t y uptake and some transmitter-related enzymes. J;-Neurochem; 22: 793-803. STORM-MATHISEN, J . and IVERSEN, L. L. (1979). Glutamic acid and exci -tatory nerve endings: selective uptake of [ 3H]glutamic acid revealed by light and electronmicroscopic autoradiography. Neuroscience 4: 1237-1253. ~ STRAUGHAN, D. W. (1975). Neurotransmitters and the hippocampus. In: R. L. Isaacson and K. H. Pribram (Eds) The Hippocampus. Plenum Press, New York. pp. 239-268. STUMPF, C. (1965). Drug action on the electr ical ac t ivi ty of the hippo-campus. Int; Rev; -Neurobiol; 8_: 77-138. TAKEUCHI, A and TAKEUCHI, N. (1962). Electrical changes in pre- and post-synaptic axons of the giant synapse of Loligo. J ; gen.- -Physiol; 45: 1181-1193. ~ ~ TAYLOR, C. P. and DUDEK, F. E. (1982). Synchronous neural afterdischarges in rat hippocampal slices without active chemical synapses. Science 218: 810-812. TEYLER, T. J . (1976). P las t i c i ty in the hippocampus: a model systems approach. In: A. Riesen and R. F. Thompson (Eds) Advances - - in  Psychobiology; Vol 3; Wiley, New York. pp. 301-326. - 175 -THIERRY, A. M . , STINUS, I . , BLANC, G . , and GLOWINSKI, J . (1973). Some evidence for the existence of dopaminergic neurons in the rat cortex. Brain Res. 50: 230-234. TURNER, R. W., BAIMBRIDGE, K. G. and MILLER, J . J . (1982). Calcium-induced long-term potentiation in the hippocampus. Neuroscience 7: 1411-1416. ~ TURNER, R. W., RICHARDSON, T. L. and MILLER, J . J . (1984). Ephaptic inter-actions contribute to paired pulse and frequency potentiation of hippo-campal f i e l d potentials. Exp;-Brain Res. 54: 567-570. UNGERSTEDT, U. (1971). Stereotaxic mapping of the monoamine pathways in the rat brain. Acta physio! . Scand; 83 (Suppl) 367: 1-48. VAN HARREVELD, A. and FIFKOVA, E. (1975). Swelling of dendritic spines in the fascia dentata. after stimulation of the perforant fibres as a mechanism of post-tetanic potentiation. Exp; Neurol. 49: 736-749. VAN HARREVELD, A. and KHATTAB, F. I. (1967). Changes in cortical extra-cel lular space during spreading depression investigated with the electron microscope. J ; Neurophysiol; 30: 911-929. VORONIN, L. L. (1980). Cellular mechanisms of long-term posttetanic potentiation in the hippocampus. In: 0. Feher and F. Joo (Eds) Cellular- -Analogues - of--Conditioning and Neural P las t i c i ty ; Akademiai Kaido, Budapest, pp. 165-174. VORONIN, L. L. (1983). Long-term potentiation in the hippocampus. Neuroscience 10: 1051-1069. WALL, P. D. (1958). Exc i tabi l i ty changes in afferent f ibre terminations and their relation to slow potentials. J ; Physiol. (Lond;) 142: 1-21. WALL, P. D. and JOHNSON, A. R. (1958). Changes associated with post-tetanic potentiation of a monosynaptic reflex. J . Neurophysiol. 21: 148-158. WEIGHT, F. F. and ERULKAR, S. D. (1976). Modulation of synaptic transmitter release by repetitive postsynaptic action potentials. Science 193: 1023-1025. WHITE, W. F . , NADLER, J . V . , HAMBERGER, A . , COTMAN, C. W. and CUMMINS, J . T. (1977). Glutamate as a transmitter of the hippocampal perforant path. Nature (Lond.) 270: 356-357. WHITE, W. F . , NADLER, J . V. and COTMAN, C. W. (1979). The effect of acidic amino acid antagonists on synaptic transmission in the hippocampal formation i n - v i t r o . Brain Res. 164: 177-194. - 176 -WIERASZKO, A. (1983). Uptake of D-[ 3H]aspartic acid by hippocampal s l i ces : the influence of low and high frequency activation of nerve endings. Brain Res; 259: 324-326. WIGSTROM, H. and GUSTAFSSON, B. (1983). Facilitated induction of hippo-campal long-lasting potentiation during blockade of inhibi t ion . Nature  (Lond.) 301: 603-604. WIGSTROM, H. and GUSTAFSSON, B. (1984). A possible correlate of the post-synaptic condition for long-lasting potentiation in the guinea pig hippocampus in vi t ro . Neurosci; Lett; 44: 327-332. WIGSTROM, H . , GUSTAFSSON, B. and HUANG, Y-Y. (1985) A synaptic potential following single volleys in the hippocampal U\\ region possibly involved in the induction of long-lasting potentiation. Acta physiol ; Scand. 124: 475-478. WIGSTROM, H . , MCNAUGHTON, B. L. and BARNES, C. A. (1982). Long-term synaptic enhancement in hippocampus is not regulated by postsynaptic membrane potential . Brain Res. 233: 195-199. WIGSTROM, H. and SWANN, J . W. (1980). Strontium supports synaptic trans-mission and long-lasting potentiation in the hippocampus. Brain -Res;  194: 181-191. WIGSTROM, H . , SWANN J . W. and ANDERSEN, P. (1979). Calcium dependency of synaptic long-lasting potentiation in the hippocampal s l i c e . Acta  physiol. Scand-. 105: 126-128. YAMAMOTO, C. and CHUJO, T. (1978). Long-term potentiation in thin hippo-campal sections studied by intracellular and extracellular recordings. Exp; Neurol. 58: 242-250. YOUNG, A. B. , OSTER-GRANITE, M. L . , HERNDON, R. M. and SNYDER, S. H. (1974). Glutamic acid: selective depletion by viral induced granule ce l l loss in hamster cerebellum. Brain-Res; 73: 1-13. ZANOTTO, L. and HEINEMANN, U. (1983). Aspartate and glutamate induced reductions in extracellular free calcium and sodium concentration in area CA^ of ' i n v i t r o ' hippocampal slices of rats. Neurosci; -Lett; 35: 79-84. ZIEGLGANSBERGER, W. and PUIL, E. A. (1973). Intracellular investigations on the effect of microelectrophoretically applied glutamate antagonists upon spinal neurones of the cat. Naunyn-Schmiedeberg's-- Arch. Pharmacol. 277: R89. 

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097124/manifest

Comment

Related Items