Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ketamine disrupts synaptic transmission by interacting at AMPA/Kainate receptor channels in neocortical… Leong, Darrell Andrew 2002

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

Item Metadata


831-ubc_2003-0104.pdf [ 2.81MB ]
JSON: 831-1.0090892.json
JSON-LD: 831-1.0090892-ld.json
RDF/XML (Pretty): 831-1.0090892-rdf.xml
RDF/JSON: 831-1.0090892-rdf.json
Turtle: 831-1.0090892-turtle.txt
N-Triples: 831-1.0090892-rdf-ntriples.txt
Original Record: 831-1.0090892-source.json
Full Text

Full Text

Ketamine Disrupts Synaptic Transmission by Interacting at AMPA/Kainate Receptor Channels in Neocortical Neurons by  DARRELL ANDREW LEONG  B.Sc, Simon Fraser University, 1997  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Pharmacology & Therapeutics)  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA December 2002 © Darrell Andrew Leong, 2002  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  Department of Pharmacology & Therapeutics The University of British Columbia Vancouver, Canada  Date  ^  *7*  11  Ketamine Disrupts Synaptic Transmission by Interacting at AMPA/Kainate Receptor Channels in Neocortical Neurons  ABSTRACT The dissociative anaesthetic, ketamine, is commonly known as a NMDA receptor channel blocker. However, additional mechanisms of action probably contribute to its various clinical effects. We asked if ketamine influences transmission via AMPA/kainate glutamate receptor channels. Using whole cell patch clamp techniques in brain slices we recorded excitatory postsynaptic potentials, evoked by electrical stimulation of axons in white matter,frompyramidal neurons in the auditory cortex of gerbils. The neurons were layer IU.-V large pyramidal neurons identified morphologically under an infrared differential interference contrast microscope, and electrophysiologically by virtue of their firing patterns. After blockade of the slow, NMDA receptor mediated EPSP component with APV, ketamine application produced a concentration-dependent reduction in amplitude of fast, CNQX-sensitive EPSPs. IC  50  values for ketamine were estimated to  range from 7 to 35 pM. The reduction of subthreshold EPSP amplitudes was maintained for a long time following cessation of the ketamine application, but the amplitude recovered when firing was induced in the postsynaptic neuron during the washout period. At concentrations of 25 to 125 uM that led to a reduction of the fast EPSP amplitude ketamine caused an increase in input resistance. At higher concentrations, which ranged from 50 to 250 pM, resistance fell again towards control levels.  Ketamine did not  increase input resistance after CNQX blockade of AMPA and kainate receptor channels. Thus, ketamine appears to reduce synaptic excitation and membrane conductance by an incomplete blockade of AMPA and/or kainate receptor channels that are constitutively active at a low level in the slice preparation of auditory cortex. After blockade of both the NMDA and AMPA/kainate receptor channels, the membrane responded to application of 100 uM ketamine with an increase in resistance.  Therefore, ketamine  appears to produce an increase in resting resistance by partial blockade of AMPA/kainate receptor channels (as reported here) as well as NMDA receptor channels (as reported previously). Furthermore, complete blockade of NMDA and AMPA/kainate receptor  Ill  channels with APV and CNQX revealed a third, yet unidentified ketamine effect associated with a resistance increase. Ketamine concentrations at synapses probably are within the range that attenuates glutamatergic synaptic transmission at AMPA/kainate and NMDA receptor channels during anaesthesia in human patients.  TABLE  OF  CONTENTS  Abstract  ii  Table of Contents  iv  List of Tables  vi  List of Figures  vii  Acknowledgements  1  2  viii  INTRODUCTION  1  1.1  Ketamine - The Beginnings  1  1.2  Ketamine - A Dissociative Anaesthetic  1  1.3  A Historical Perspective - Ketamine as a NMDA Receptor Blocker  2  1.4  Is Ketamine A Selective NMDA Receptor Blocker?  3  1.5  Ketamine Effects At Non-Glutamate Receptors  5  1.6  Ketamine - An AMPA/Kainate Receptor Channel Blocker?  6  1.7  Ketamine - A Non-Selective Glutamate Receptor Antagonist (Our Hypothesis)... 7  EXPERIMENTAL  PROCEDURES  10  2.1  Gerbil Cortical Slice Preparation  10  2.2  Whole Cell Recordings  11  2.3  Selection Criteria of Patched Neuron  12  2.4  Data Acquisition  12  2.5  Drugs  14  2.6  Data & Statistical Analysis  14  3  RESULTS 3.1  Electrophysiological Identification of Regular Spiking Pyramidal Neurons  3.1.1  4  Spike Adaptation of Pyramidal Neurons in Gerbil Neocortex  14 14 17  3.2  Pharmacologically Isolating AMPA/Kainate Receptor Channels  17  3.3  Ketamine Effects on Fast Excitatory Postsynaptic Potentials  18  3.4  Recovery of Ketamine-lnduced Effects  25  3.5  Lack of Presynaptic Ketamine Effects as Determined by Paired-Pulse Experiments  27  3.6  Ketamine Effects on Input Resistance (R)  29  3.7  Lack of Ketamine Effect on Holding Potential  31  3.8  Unmasking A Third Mechanism of Action by Ketamine  33  DISCUSSION 4.1  35  Ketamine Depressed AMPA/Kainate Receptor Channel-Mediated Fast Excitatory Postsynaptic Potentials (fast EPSPs)  36  4.2  Ketamine-lnduced Increases in Input Resistance, Rj  38  4.3  Recovery from Drug Effect and Effective Drug Concentrations  40  4.4  The Mongolian Gerbil in vitro Model  41  4.5  Functional Significance  42  5  CONCLUSIONS  42  6  REFERENCES  44  LIST  OF  TABLES  Table I. Ketamine Effect on Spike Firing Properties of 5 Pyramidal Neurons of Gerbil Neocortex 15 Table II. Ketamine-lnduced Reductions of Stimulus-Evoked fast EPSP Amplitudes  20  Table III. Ketamine-lnduced Increase in Input Resistance After Initial Application of Glutamate Receptor Blockers, APV and CNQX  23  Table IV. Input Resistance Changes in Response to Increasing Ketamine Concentrations  29  LIST  OF  FIGURES  Figure 1. Regular-Spiking Pyramidal Neurons of Gerbil Neocortex  16  Figure 2. EPSPs Evoked From Layer III Pyramidal Neuron by Identical Stimulation of Afferent Fibres in White Matter  19  Figure 3. The Reduction of APV-Resistant Fast EPSPs Depended on Ketamine Concentration  21  Figure 4. Variations in Fast EPSP Amplitude After NMDA Receptor Blockade with APV Depended on Ketamine Concentration  22  Figure 5. The Ketamine-Sensitive EPSP That Remained After NMDA Receptor Blockade Depended on AMPA/Kainate Receptor Channels  24  Figure 6. The Action Potential and EPSP Were Attenuated With Application of Ketamine and Recovered With Continuous Stimulation and Depolarization ...26 Figure 7. Ketamine Did Not Affect the Release of Transmitter From Presynaptic Terminals in Response to the Second Stimuli  28  Figure 8. Ketamine-lnduced Increase in Resistance Varied in Quantity and Quality but Depended on Ketamine Concentration  30  Figure 9. Increasing Concentrations of Ketamine Did Not Induce a Shift In Resting Membrane Potential  32  Figure 10. Ketamine-lnduced Increases in Input Resistance Observed After the Complete Block of NMDA and AMPA/Kainate Receptor Channels  34  ACKNOWLEDGEMENTS  Supervisors to this thesis: Dr. Dietrich Schwarz Dr. Ernest Puil Dr. Craig Ries  Department of Surgery, Division of Otolaryngology Department of Pharmacology & Therapeutics Department of Anesthesia  Technical Support: Mr. Bill Chan Mr. Stephan Ho Mr. Jooyong Chang  S.U.C.C.E.S.S. Canon, Inc. CulTech Entertainment Inc.  Encourgement: Langara Group Health Canada  2000 - 2003 Members of D.A.S. and C.F.I.A.  Most ofaCC  To  fi^tt^^ht^  L60NQfy*y«lf  » Than^s so much foryour Coving support and encouragement.  ,(  1  1  1.1  INTRODUCTION  Ketamine - The Beginnings In 1962, Calvin Stevens first synthesized ketamine at the Parke-Davis laboratories  in Michigan. Thereafter, ketamine was determined to be a shorter-acting and far less toxic anaesthetic than its cogener, phencyclidine (PCP). In 1970, the US Food & Drug Administration (FDA) approved its use in clinical medicine. Ketamine's hallucinogenic properties popularized its abuse in the psychedelic underground, especially after its clinical use as a battlefield anaesthetic during the Vietnam War (Jansen et al., 2000). Subsequent abuse of ketamine led to governmental regulatory control in the United States (1979). During the mid-1980's, basic research elucidated one mechanism of action of ketamine, a blockade of N-methyl-D-aspartate (NMDA) receptor channels (MacDonald et al., 1990; see below).  Ketamine is still favored for clinical use because of its  comparatively mild depressant effects on the respiratory system and of its stimulatory effects on the cardiovascular system (Mycek et al, 1997; details to follow). It continues to be sold on the black-market under a street-name, "Special K" or "K", to abusers (Jansen et al., 2000).  1.2  Ketamine - A Dissociative  Anaesthetic  The clinical end point of general anaesthesia is defined as "a loss of response to, and perception of, all external stimuli" (Evers et al., 2001). Ketamine's unique analgesic and anaesthetic effects challenge this textbook definition.  In contrast to most  anaesthetics, ketamine produces an intense analgesia at sub-anaesthetic and anaesthetic dosages.  Patients under ketamine anaesthesia may appear awake and responsive to  certain sensory stimuli but do not respond to noxious stimuli. They may respond to commands, but are amnesic. Their eyes may remain open, they move their limbs, and usually exhibit spontaneous respiration.  The clinical appearance of a patient  anaesthetized with ketamine resembles a cataleptic state. The unique property of intense analgesia dissociated from states of conscious awareness led to classification of ketamine and related chemical derivatives as "dissociative anaesthetics" (Corssen et al, 1966, Evers et al., 2001).  As observed with PCP, the onset of analgesia occurs prior to  anaesthesia reflecting dose-dependent dissociation of these states from environmental  2  awareness (Domino, 1964). This concentration-response pattern is of great theoretical and clinical interest.  1.3 A Historical Perspective - Ketamine as a NMDA Receptor  Blocker  Chemical and pharmacological properties of ketamine (Drug Code: CI-581) were reported after the time of its synthesis in the 1960's, and their study continued throughout 1970's and early 1980's.  The drug is an arylcycloalkylamine (Ci Hi ClNO) with a 3  6  molecular weight of 238 and a pKa of 7.5. Although ketamine is water soluble, it's lipid solubility is ten times that of the induction agent, thiopental (Evers et al., 2001). Structurally, the molecule, 2-(o-chlorophenyl)-2-(methylamino)cyclohexanone,  contains  a chiral center at the C-2 carbon of the cyclohexanone ring so that two enantiomers of ketamine molecule exist: S(+)ketamine and R(-)ketamine (reviewed in White et al., 1982).  Racemate preparations of ketamine contain equal concentrations of the two  enantiomers and are commercially available as, for examples, Ketaject (Bristol-Myers Squibb) and Ketalar (Parke-Davis). These enantiomers differ in potency, in responses of the electroencephalograph (EEG), and in their effects on catecholamine reuptake (White etal., 1980, 1982). Ketamine's lipophilic properties allow rapid entry through the blood-brain barrier into the brain tissue, followed by redistribution to other organs and tissues. Small amounts of ketamine are excreted unchanged in the urine. demethylated  ketamine, is the principal metabolite  Norketamine, or N-  of liver cytochrome P450  biotransformation and is excreted in urine. Norketamine is an active metabolite having one-third of the anaesthetic properties of ketamine (reviewed in Evers et al., 2001, White et al., 1982). Norketamine is hydroxylated to dehydronorketamine and conjugated to water-soluble compounds that are also excreted in the urine (Adams et al., 1981). Clinical effects of ketamine include mild depression of the respiratory system, and stimulation of the cardiovascular system (Mycek et al., 1997). Ketamine is used in children and young adults where immobilization is required during short procedures (e.g. facial lacerations). Other indications include surgery in children of less than 12 months of age with ariskof airway complications, and patients who are atriskof hypotension or bronchospasms (Reich et al., 1989; Myceck et al., 1997). However, ketamine is not  3  widely used because it increases cerebral blood flow and induces postoperative hallucinations. Scientific exploration of ketamine began in the 1970's and continued extensively in the 1980's.  During this "pioneering" research period, ketamine was observed to  depress the discharge activity of single neurons, whether spontaneous, evoked by afferent nerve stimulation (Conseiller et al., 1972), and /or induced by glutamate application (Sinclair et al., 1979). However, this observation of "neuronal depression", demonstrated in dorsal horn neurons, was not universally observed. Some neurons were unaffected (Kitahata et al., 1973), or even displayed increased excitability (Winters et al., 1972) after ketamine application.  The focus of research then turned to ketamine actions on  excitatory synapses. In the 1980's, the idea of a ketamine block of synaptic release of excitatory amino acid neurotransmitters in nervous tissue was proven to be incorrect (Minchin, 1981). Shortly thereafter, ketamine was shown to be a blocker of postsynaptic NMDA receptors as was a variety of other dissociative anaesthetics, including PCP (Anis et al., 1983; MacDonald et al., 1987, 1989). A research team at the University of Toronto, led by J.F. MacDonald, explored molecular mechanisms of ketamine block on the NMDA receptor channel (reviewed by MacDonald et al., 1990). Ketamine was determined to bind to a site within the lumen of activated NMDA receptor channels, resulting in a physical obstacle to ion currents through the channel pore. Activated "open" states of the channel provide access to the binding site for ketamine molecules (Beverley et al., 1997), which can become trapped there when the channel closes. Conversely, ketamine release from the binding site can occur during channel activation induced by agonist interaction with receptors. Ketamine blockade of NMDA receptor channels has been aptly described as voltage- (Honey et al, 1985) and use- (MacDonald et al, 1987) dependent. Subsequent recovery from steady state block at a given potential required repeated application of the agonist, L-asparate (Honey etal, 1985).  1.4  Is Ketamine A Selective NMDA Receptor  Blocker?  The advancements in molecular techniques during the 1980's and 1990's have lead to the discovery of a great diversity within the glutamate receptor channel family.  4  These receptors are classified into two main families, termed ionotropic glutamate receptors  (iGluR) and metabotropic glutamate receptors (mGluR) (reviewed by  Pellicciari et al., 2000; Mori et al., 1995; Hollman et al., 1994). Ionotropic glutamate receptors are subdivided into three subtypes: hydroxy-5-methyl-4-isoxazoleprionic  N M D A , kainate (KA), and a-amino-3-  acid (AMPA)  receptors  (Mori  et al., 1995).  Metabotropic glutamate receptors are classified into three groups (I, n, Ul) according to sequence homology, transduction mechanisms, and agonist pharmacology (Pellicciari et al., 2000). A literature search failed to produce reports on ketamine action at mGluR channels or ionotropic A M P A and kainate (AMPA/kainate) receptor channels.  The  search revealed that ketamine was widely considered a selective N M D A receptor channel blocker in clinical and experimental research. For example, the mechanism of ketamine action in the successful treatment of phantom limb pain was suggested to be via a selective N M D A receptor channel block (Stannard et al., 1993, Nikolajsen et al., 1997, 2001). We believe that ketamine selectivity for N M D A receptor channels should be reevaluated in the context of recent revelations of a great structural diversity within glutamate receptor family, which justify a new examination of possible effects on the AMPA/kainate receptor channels. Synaptic currents mediated by glutamatergic receptor channels are currently understood to have an important role in the transduction of pain in the central nervous system (Hewitt DJ, 2000) in learning and memory (Parada-Turska, 1990), and in certain pathophysiological conditions such as schizophrenia (Tsai et al., 2002) and epilepsy (Rogawski, 1992; Chapman A G , 2000). Defining the functional roles of, for example, AMPA/kainate  versus  NMDA  receptor  channels,  in various physiological and  pathophysiological conditions (as mentioned above) are the focus of current research (e.g.  Szekely et al., 2002).  Resurgence  in the popularity of ketamine,  as a  pharmacological tool, is underway, due to its versatility as a clinical and experimental drug, particularly because of its unique dissociative properties. It is of great theoretical interest and has potentially practical consequences to determine if ketamine acts on AMPA/kainate receptor subtypes of central glutamate receptors.  5  1.5 Ketamine Effects At Non-Glutamate  Receptors  Ketamine effects are not exclusive to N M D A receptor channels, but include antagonism of muscarinic and nicotinic cholinergic receptors, and enhancement of effects mediated by adrenergic receptors activation (reviewed in Reich et al., 1989, Kress et al., 1997). A recent study, for example, showed that acetylcholine-induced currents mediated by human nicotinic acetylcholine receptors, expressed in Xenopus oocytes, containing either (32 or P4 subunits, were inhibited in a concentration-dependent manner by ketamine, with ICso's of 9.5 to 29 u M and 50 to 92 u M , respectively (Yamakura et al., 2000; another example Irnaten et al., 2002). The same study showed a ketamine-induced concentration-dependent inhibition (IC50 = —910 uM) of currents mediated by activated 5-HT3  receptors. Ketamine stimulatory effects on adrenergic receptors are exemplified  by a study demonstrating increased norepinephrine-induced contractile responses on vascular smooth muscle (VSM) of rat aorta, in the absence and presence of endothelium (Akata et al., 2001).  Experimental studies, such as the above, provide a molecular  explanation for treatment of patients at risk of hypotension. Whole cell patch clamp experiments on human neuroblastoma SH-SY5Y cells expressing voltage-dependent K  +  channels demonstrated a ketamine inhibition of  potassium conductance in a concentration-dependent (IC50 = 300 uM) and reversible manner (Friederich et al., 2001). Inward sodium currents mediated by voltage-gated sodium channels, recombinantly expressed in Xenopus oocytes, exhibited tonic- (IC50 = 800 uM) and phasic- (IC50 = 2300 uM) concentration-dependent block by ketamine (Wagner et al., 2001). Ketamine inhibition of sodium currents was also demonstrated in both TTX-sensitive and TTX-resistant sodium channels expressed in dorsal root ganglion neurons. Half-maximal concentrations at resting membrane states were 146.7 u M and 866.2 (aM, respectively (Zhou et al., 2000). Ketamine's general block of Na -channels at +  high concentrations suggests an action of local anaesthesia, like that of lidocaine. Ketamine effects on opioid receptors have also been reported.  It is generally  believed that N M D A receptor channel antagonism accounts for most of ketamine's anaesthetic and part of its analgesic effects, whereas some of ketamine's analgesic effects may be mediated through an agonistic effect at opioid receptors within the CNS. For example, Smith et al. (1980) showed that racemic ketamine displaced tritiated naloxone  6 (an opioid receptor antagonist) from rat brain opioid receptors in vitro, and that S(+)ketamine was about twice as potent as R(-)ketamine. Finck et al. (1982) confirmed these results and also found similar stereospecific opioid receptor binding by S(+)ketamine in guinea pig ileum in vitro. At that time, ketamine's actions at the opioid receptors were controversial, particularly since a reversal of ketamine anaesthesia by naloxone could not be demonstrated. Recent reports, however, continue to support the involvement of opioid receptors in ketamine-induced antinociception (e.g. Sarton et al., 2001).  Antagonism of AMPA/kainate receptor channels, and not N M D A receptor  channels, located in the spinal cord has been suggested to be involved in enhancing the inhibition of the tail-flick response induced by stimulation of spinal p-, 8-, and K - opioid receptors (Suh et al., 2000). This particular finding is of particular interest to us because it supports our view that pain processing involves specific AMPA/kainate receptor channel subtypes. Ketamine concentrations required for most effects at non-glutamatergic channels tend to be high, >100 to 2300 uM (as cited above). However, concentrations of ketamine during anaesthesia were estimated to be in the order of tens of micromoles per litre in both cerebrospinal and in plasma fluids (Cohen et al., 1973). Therefore, we will study the concentration-dependent effects of ketamine on AMPA/kainate receptor channels and determine whether estimated  IC50  values are relevant for analgesic and anaesthetic  effects.  1.6 Ketamine - An AMPA/Kainate Receptor Channel Blocker? A possible role of non-NMDA receptor channels in ketamine anaesthesia and analgesia should be considered in view of the distribution of AMPA/kainate receptor subunits in the CNS (reviewed by Dingledine et al., 1999; Bleakman et al., 1998, Hollman et al., 1984).  Both A M P A and kainate receptor channels are pentameric  structures consisting of protein subunits, G l u R l , GluR2, GluR3 and GluR4, and K A 1 , K A 2 , GluR5, GluR6, and GluR7, respectively. The variations in subunit stoichiometry for AMPA/kainate receptor channels are not exactly known at most locations in the brain. Nevertheless, a variety of techniques including radioligand binding of [ H] A M P A , single 3  cell reverse-transciptase polymerase chain reaction, in situ hybridization histochemistry,  7  and immunocytochemistry using various AMPA receptor antibodies have been used to regionally localize high expressions of non-NMDA receptor protein subunits to thalamus, neocortex, hippocampus CAI, CA2 and CA3 pyramidal cell layers, piriform cortex, cerebellum and spinal cord. Concurrent studies of a similar nature have demonstrated that the NMDA receptor channels also consist of a variety of protein subunits (Mori et al., 1995; Hollmann et al., 1994). The NMDA receptor protein subunits, NMDAR1, NMDAR2A, NMDAR2B, are highly expressed in thalamus, neocortex, hippocampus CAI, CA2 and CA3 pyramidal layers, and are absent, have low expression, or have not yet been determined in piriform cortex, cerebellum and spinal cord. NMDAR2C proteins are absent in the brain regions mentioned above; and NMDAR2D regional expression and distribution have yet to be determined, (cf. Hollman et al., 1994). Variations of subunit compositions for glutamate receptor channels and differences in regional distributions suggest a diversity of possible effects, and a range of sensitivity to ketamine. For example, recent studies have implicated AMPA/kainate receptors containing GluR5 subunits in nociceptive processing in the spinal cord. Spinal AMPA/kainate receptors are also involved in the initiation of secondary tactile allodynia induced by focal thermal injury (Nozaki-Taguchi et al, 2002). Interestingly, the same study reports that spinal NMDA receptors play only a minimal role. Antagonists, LY294486 and LY382884, specific for GluR5 subunits have significantly reduced responses to noxious stimuli (Procter et al., 1998). Clinical trials of LY293558 provided evidence of reduction in clinical pain presumably by blocking AMPA/kainate receptors that contain GluR5 (Gilron et al., 2000). Since GluR5 is also expressed in neocortex (Hollman et al., 1998) a subunit-dependent role in pain signaling through AMPA/kainate receptor channels could well extend beyond the spinal cord toward higher centres. However, the functional roles of GluR5 containing AMPA/kainate receptor channels are not fully understood and they could be involved in the mediation of other central processes.  1.7 Ketamine - A (Our Hypothesis)  Non-Selective  Glutamate  Receptor  Antagonist  The history of ketamine covers about 40 years of research and development, which include some 25 years of clinical use as an anaesthetic. More work seems to be required to identify all the neuronal mechanisms responsible for the unique property of  8  intense analgesia contributing to dissociative anaesthesia, and to understand the interplay between these mechanisms. The claim that ketamine is a selective NMDA receptor channel blocker can be challenged as ketamine also acts on non-glutamate receptors and channels (e.g. Na , K channels, adrenergic, cholingeric receptors) at high concentrations +  +  of the drug (typically >100 uM). In comparison, IC50 values for complete block of NMDA receptor channels range from 1 to 100 uM depending on reported studies and, presumably, on the subunit constitution of the receptor channels.  Therefore, in  comparison to investigated alternative channels, a claim of ketamine selectivity for NMDA receptor channels can be supported provided the drug concentration remains within limits of effective clinical concentrations found in cerebrospinal and plasma fluids, which have been reported to be in the lower micromolar range (Cohen et al., 1973). However, it is not clear what effects, if any, ketamine has on AMPA/kainate receptor channels. Ketamine action on AMPA/kainate receptor channels has aroused our interest, especially after enlightening reports of AMPA/kainate receptor channel involvement in nociception in spinal cord. Wide distribution of AMPA/kainate receptor channels in brain and, in particular, high expression levels in neocortex and thalamus suggest that AMPA/kainate receptor mediated transmissions is involved in central mediation of pain signals. Thalamic and cortical brain areas, expressing AMPA/kainate receptor channels, have been implicated in pain sensations (reviewed by Janig, 1987; Price, 2000). The spinothalamic and thalamocortical pathways are of particular interest since these neural mechanisms are described to involve ascending projections from the spinal cord dorsal horn directly to the ventroposterior lateral (VPL) and ventroposterial inferior (VPI) and intralaminar nuclei of the thalamus.  The thalamic nuclei, then, relay nociceptive  information to somatosensory cortices where it plays a role in generating conscious awareness and characteristic EEG rhythms (Albe-Fessard et al., 1985; Steriade et al., 1990). Neocortical AMPA/kainate receptor channels, therefore, play an important role in the processing, memory and perception of pain, similar to AMPA/kainate receptors of the spinal cord. We hypothesize that ketamine acts on AMPA/kainate receptor channels in addition to NMDA receptor channels.  The evidence of the role of AMPA/kainate  9  receptor channels i n spinal cord nociception extends our hypothesis to include ketamine effects on similar channels expressed in neocortex o f brain. W e entertain the idea that ketamine is non-selective and has different sensitivities to receptor subunits o f both N M D A and AMPA/kainate receptor channels.  If these effects on glutamate receptors  truly exist, perhaps this fact could lead to greater understanding o f "dissociative" anaesthetics and their molecular mechanisms. One viewpoint, included i n our hypothesis, is the possibility o f ketamine-induced reduction o f depolarizing fast synaptic transmissions mediated by AMPA/kainate receptor channels, which results in a concomitant reduction i n corresponding excitatory postsynaptic  potentials  (EPSPs).  AMPA/kainate receptor channels  predominantly  mediate the fast component o f an E P S P , evident during low frequency transmission. A t resting membrane potentials, N M D A receptor channels play a minor role because M g substantially  blocks  the  channel  pores,  and this block  may  be  hyperpolarization, including G A B A - m e d i a t e d synaptic inhibition.  intensified As  by  a result, a  ketamine block o f the amplifying component mediated by A M P A / k a i n a t e receptor channels could result in a concomitant reduction in the number o f N M D A channels reaching threshold to "open" channel-states following the release o f M g  receptor 2 +  block.  Thus, an AMPA/kainate receptor channel block may intensify the Mg -dependent block 2+  o f N M D A receptors. O f particular interest is the success in the treatment o f episodes o f phantom limb pain with ketamine, which may well be correlated to the plastic reorganization o f somatotopic representation i n cortex, following a limb amputation.  Another phantom  sensation, tinnitus, commonly accompanies the plastic change i n tonotopic representation in auditory cortex that occurs after band-limited hearing loss.  Thus, ketamine could  reduce both phantom sensations by similar mechanisms i n both somatosensory and auditory cortex. We, therefore, chose to look for changes in glutamatergic transmission through AMPA/kainate receptor channels in neurons o f the auditory cortex. Whole cell patch clamp techniques were employed to determine ketamine effects, mediated by pharmacologically isolated AMPA/kainate receptor channels, on membrane properties o f pyramidal neurons in tissue slices o f gerbil neocortex.  Channel activation  was achieved by release o f excitatory amino acids due to electrical stimulation o f cortical  10  afferents. Ketamine-induced changes of excitatory postsynaptic potentials (EPSPs) were observed at levels subthreshold to spike firing.  In addition, we measured ketamine  concentration-dependent effects on input resistances.  2  EXPERIMENTAL  PROCEDURES  Experimental use of animals was approved by the University of British Columbia Animal Care Committee. Tissue slices of neocortex were preparedfromMongolian gerbils, Meriones unguiculatus. Animals of both sexes between the ages of 16 to 24 days old were selected from our colony. The age of the animal was selected on the basis of developmental evidence indicating highly variable postnatal expression and distribution of glutamatergic receptors. At an age greater than P16 the gerbil auditory cortex seems to express reasonably mature voltage responses mediated by glutamatergic receptors. The maximal age, P24, provided the upper limit of the developmental period during which individual neurons could be readily patched without electrode block by myelin.  2.1  Gerbil Cortical Slice Preparation Gerbils were anaesthetized with halothane (MTC Pharmaceuticals, Ontario,  Canada) in a closed chamber for approximately 30 seconds until spontaneous limb movement was no longer observed, and spontaneous breathing was maintained. The gerbil was decapitated and the brain removed immediately and placed in ice-cold (~0°C) solution containing (in mM): sucrose, 125; KC1, 3.0; CaCl , 2.0; NaH P0 , 1.25; 2  2  4  NaHC0 , 25; MgCl , 2.0; D-Glucose, 25; buffered to pH 7.35 with continual aeration 3  2  with 95% 0 , 5% CO2 gas mixture. The isolated brain was left submerged in ice-cold 2  ACSF for approximately 1 minute to reduce 0 demand due to cellular metabolism and 2  increase subsequent neuronal survival. As guidelines for visual identification of pyramidal neurons of gerbil neocortex, we used the architectonic description of the gerbil auditory cortex (Budinger et al., 2000) and the general cyto-architecture of the cortex as described by pioneer scientists such as Brodmann (1909), Cajal (1911), and Lorente (1949) (described in Carpenter, 1976). We cut a tissue block from isolated brains using the middle cerebral artery and inferior  11  cerebral vein as markers to localize the auditory cortex.  This tissue was cut with a  PELCO 101 Vibratome Sectioning System (Series 1000) for preparation of coronal slices (300 pm). The tissue slices were immediately transferred to a nylon-mesh grid secured in a temperature controlled holding chamber containing artificial spinal fluid (ACSF) at 37°C, and aerated continually with 95% O2, 5% CO2 gas mixture. This ACSF is a solution containing (in mM): NaCl, 125; KC1, 3.0; CaCl , 2.0; NaH P0 , 1.25; NaHC0 , 25; 2  2  4  3  MgCl , 2.0; D-Glucose, 10; buffered to pH 7.35 with bubbling of 95% 0 , 5% C 0 . 2  2  Incubating tissue slices at 37 °C allowed metabolic recovery of the neurons.  2  The  temperature of the holding bath was allowed to slowly return to ambient room temperature (23 to 26 °C) where a balance between reduced oxygen demand and prolonged cell survival produced optimal conditions for patch recording. Slices were allowed to recover for one hour.  2.2  Whole Cell Recordings Intracellular recordings from pyramidal cells were made using electrodes drawn  from glass tubing with an outside diameter of 1.5 mm (World Precision Instruments, Inc., Sarasota, FL, USA) on a Narishige (type PP-83) two-stage vertical electrode puller. Patch electrodes were filled with a solution containing (in mM): KOH, 140; EGTA, 10; KC1, 5; NaCl, 4; MgCl , 3; HEPES free acid, 10; CaCl , 1; Na ATP, 3 added on the day 2  2  2  of the experiment; titrated to pH 7.3 with gluconic acid. Electrode tip resistances were 5 to 9 MQ. Current clamp recordings were obtained in the bridge mode of an Axoclamp 2B amplifier (Axon Instruments, Inc., Foster City, CA, USA). Individual slices were transferred from the holding container to the recording chamber equipped with in- and out- flow ports designed for laminar flow of the bathing solution. A flow rate of 3 mL/min was achieved with a peristaltic pump through the inflow port. The direction of flow and the volume of the recording bath in the chamber (approx. 1.5 mL) were maintained with vacuum suction through the outflow port. Each slice was oriented so that the cortical layers were orthogonal relative to the position of the recording electrode and held in place with a nylon-mesh spanned over and mounted on a platinum U-frame. This procedure allowed for visualization of large somata within the  12  mid-layers of the cortical cross-section.  Neurons were visualized with a Axioskop FS  (Carl Zeiss, Gottingen, Germany) equipped with differential interference contrast (DIC) optics (Stuart et al., 1993). Video images were obtained with infrared and/or white light illumination using a Hamamatsu C2400-07ER video camera system (Hamamatsu Photonics K.K., Hamamatsu, Japan) and displayed on a black and white Sony SSM-171 monitor (Sony Corporation, Tokyo, Japan). For recording, we selected large pear-shaped pyramidal neurons (Carpenter et al., 1976) estimated under the microscope to be located in cortical layers ni and V. Electrophysiological observations of spike frequency and firing patterns suggested that typical cell recordings were taken from "regular spiking" pyramidal cells (Beggs et al, 2000; Faulkner et al, 1999; Connors et al, 1990, 1982; McCormick et al., 1985).  2.3 Selection Criteria of Patched Neuron The criteria for accepting a patched neuron were:  1) the resting membrane  potential (V ) of < -50 mV (typically between -55 to -65 mV); 2) a patch seal of > 0.8 R  GQ (typically 1.0 to 1.5 GQ); 3) an action potential height of at least 50 mV (typically 80 to 90 mV from threshold at approximately -45 mV). From our experience, the use of these guidelines typically resulted in a selection of viable neurons suitable for recording periods upwards of 3 hours. This screening procedure also ensured consistent acquisition of data over the length of the recording period. Presumably, much of the consistency was due to the integrity of the gigaohm seal despite variable current inputs. And, finally, the criteria provide means to qualitatively monitor the state of health of the tissue slice prior to drug application, or the viability of a cell over the course of the recording period.  2.4 Data Acquisition Data acquisition, storage, and analysis were controlled using pClamp 8.1 software (Axon Instruments, Inc., Foster City, CA, USA) running on a PC. Using Axoclamp-2B (bandwidth = 30 kHz), membrane currents (xl; 100 mV/nA) and amplified membrane potentials (xlO) were low-pass filtered (model LPF-30A; World Precision Instruments, Sarasota, FL, USA) at a frequency of 5 kHz (-3 dB). Electrical activity was then amplified (x5) and digitized at 20.2 kHz (20,200 samples/sweep/signal) using the Digidata 1322A interface (Axon Instruments, Inc., Foster City, CA, USA).  13  Excitatory post-synaptic potentials (EPSPs) were evoked using bipolar glassinsulated tungsten (50 urn) electrodes placed in the white matter to stimulate projection and association fibers innervating cortical neurons. The stimulating electrode was placed in the white matter below the layers of cortex in an orientation orthogonal to the recording electrode, patched to a neuron. Electrical stimuli consisted of square wave pulses (100 p.s max. duration, 0.5 Hz) of fixed amplitudes (between 4 to 8 V) digitally timed using pClamp 8.1 software and controlled using NeuroLog System: pulse width and delay module & stimulus isolator WL800 (max. current output at 1 and/or 10 mA; Digitimer Ltd.). Experiments involving paired-pulse stimulation included an interpulse interval of 500 ms and delivered every 5 seconds (0.2 Hz) over a 30 second period. Input resistance ( R i ) was measured from induced changes in membrane potential (AV ) by injected hyperpolarizing current steps (AI), in accordance with Ohm's law. m  AI  R was calculated based on small deflections in V (e.g. 5 to 10 mV) to minimize a m  contribution of rectifying conductances (e.g. IH and k i r ) that are activated at greater hyperpolarized potentials. The current-voltage (1-V) relationships in the absence and presence of drugs were determined at the end of 500 and/or 1000 ms current steps (-0.01 to -0.02 nA) at steadystate membrane voltage responses.  Segments of steady-state membrane voltage  responses and of respective current inputs were selected for mean and standard deviation calculations. Steady-state responses were considered to be response segments with >50 ms of no significant (P > 0.05) changes in measured membrane potential.  These  calculations were routinely performed over the same selected response segments and over all treatment groups for an individual cell. Steady-state responses differed for each cell and thus required subjective selection in time and duration for the above-mentioned calculations.  14  2.5 Drugs Drugs were applied to the recording chamber by switching from the control perfusate (ACSF only) to ACSF containing a desired drug concentration: ketamine hydrochloride at 3 to 250 methylamino-cyclohexanone  (Ketalar; Pontypool, UK: hydrochloride;  phosphonovaleric acid, 100 uM (APV:  Parke  2-(o-chlorophenyl)-2-  Davis),  DL-2-amino-5-  Sigma, St. Louis, MO), and 6-cyano-7-nitro-  quinoxalinedione, 20 uM (CNQX: Sigma, St. Louis, MO) were dissolved in ACSF. The solutions were perfused at a rate of 3 mL/min using a peristaltic pump for 3 minutes.  2.6 Data & Statistical  Analysis  All responses were analyzed using ClampFit software (Axon Instruments, Inc., Foster City, CA, USA). The amplitude of evoked EPSPs was measured from prestimulation baseline to peak positive potentials.  All responses were normalized as a  percent of control, based on 6 to 21 measurements per treatment group (control versus drug). The statistical significance of data from two to eight groups was determined using student's t-tests and one-way analysis of variance (ANOVA) and post hoc tests.  3  RESULTS  3.1 Electrophysiological Neurons  Identification  of Regular  Spiking  Pyramidal  Neurons of the cortex (rat, mice) have been classified according to their morphological and physiological characteristics (White et al., 1989), and to their patterns of spike firing.  Distinct classes of spike firing are known:  regular-spiking (RS),  intrinsically bursting (IB), and fast-spiking (FS) (Connors et al, 1990, 1982; McCormick et al., 1985). More recently, late-spiking (LS) and single-spiking (SS) classes were identified in rat perirhinal cortex (McGann et al., 2001; Faulkner et al., 1999). Of particular interest to our study was the use of regular-spiking cortical neurons because of their strong thalamic inputs (Amitai Y., 2001). The same report states, "thalamocortical synapses have on average a higher release probability [of glutamate neurotransmitter] than intracortical synapses, and a much higher number of release sites per axon. As a  15  result, the transmission of each thalamocortical axon is significantly more reliable and efficient than most intracortical axons". Furthermore, both types of glutamate receptor channels have been reported to be predominant carriers of excitatory currents in both the thalamocortical and corticocortical pathways (Armstrong et al., 1993; Salt et al., 1995), of which AMPA/kainate receptor channels are of particular interest to us. Spike firing patterns identified approximately 40 patched neurons as regular spiking neurons (Figure 1), as described in the literature (Connors et al., 1990, 1982; White et al., 1989; McCormick et al., 1985). Regular spike firing patterns of neocortical neurons were used as functional criteria for a class of pyramidal neurons. We observed a significant decrease (P < 0.05; N = 5) in spike amplitude but no significant changes in spike frequency or kinetics after applications of ketamine (6 to 125 uM; P > 0.05; Table I) in the presence of APV (100 uM). Although spike firing patterns upon ketamine application were not the focus of this study, their similarity in the presence and absence of the drug is noteworthy since high concentrations ketamine has been suggested to act like a local anaesthetic.  Table I. Ketamine Effects on Spike Firing Properties of 5 Pyramidal Neurons of Gerbil Neocortex. CelM Control  Ketamne, 125pM  Cell 2 Control  Ketamne, 125 pM  Control  Ketamne, 125pM  Cell 5  Cell 4  Cell 3 Control  Ketamne, E 5 u M  Control  Ketamne, 125 pM  •59.7 ±0.3 -60.410.3 -59.8 ±0.6 -61.2 ±0.6 -59.6 ±0.2 -60.1 ±0.2 -59.0 ±0.2 -59.7 ±0.2 -59.7 ±1.0 -56.8 ±1.0 Spike Amplitude, mV'  71.3 ±0.2  48.7 ±0.7 59.1 ±5.0  48.3 ±5.0 87.2 ± 0.3  79.2 ±0.3 62.8 ±0.6  61.8 ±0.6  dV/dt Ratio  3.8 ±0.06 3.0 ±0.06 2.9 ±0.12 4.0 ±0.12 1.9 ± 0.04  2.0 ±0.04 8.8 ±0.14  7.9 ±0.14 4.6 ±0.07  5.0 ±0.07  1.8 ± 0.06 3.0 ±0.06 1.3 ±0.02  1.4 ±0.02 1.5 ±0.03  1.7 ±0.03 2.2 ± 0.04  2.4 ±0.04  s  Spike width base, ms  1.6 ±0.02  66.1 ±0.2 72.6 ±0.7 1.8 ±0.02  Spike width, Samplitude, ms 0.95 ±0.02 1.06 ±0.02 1.15± 0.04 1.97 ± 0.04 0.66 ±0.01 0.70 ±0.01 0.83 ±0.02 1.10 ±0.02 1.47 ±0.02 1.56 ±0.02 Threshold to spike firing, mV -40.2 ±0.14 -41.7 ±0.14 -40.0 ±0.40 -44.2 ±0.40 -23.4 ±0.01 -23.4 ±0.01 -26.3 ± 0.07 -25.1 ± 0.07 -18.4 ±0.23-16.5 ±0.23 ' P < 0.05 versus the value at 0.0 (M Ketamne (Control) concentrations by Bonf erroni post hoc test. 5 Ratio of Max. Rate of Rise to Max. Rate of Fall of Spi<e (dV/dt)  16  A C S F + APV, 100 uM  A C S F + APV, 100 uM + Ketamine, 125 uM  A l (nA) m  Figure 1. (A) Typical "regular-spiking" pattern of a pyramidal neuron of gerbil neocortex in the absence (red) and presence (black) of 125 uM ketamine. A depolarizing current step (0.1 nA, 1 second duration) induced a 50 to 200 Hz spike firing with weak spike adaptation (interspike intervals ranged from approximately 5 to 20 ms). Arrows indicate spike train afterhyperpolarization (AHP) with amplitude of 1.6 ± 0.6 mV and duration of -500 ms under control conditions; 1.3 ± 0.5 mV for -300 ms under ketamine application. (B) Overlay of spikes indicate significant decreases in spike amplitude (P < 0.01, N = 5) with 125 nM ketamine application. No significant differences (P > 0.01) in additional spike parameters (see Table I) were present. (C) Spike rate was not significantly (P > 0.05) affected by 125 \xM ketamine during injected current steps of various amplitudes.  17  3.1.1  Spike Adaptation of Pyramidal Neurons in Gerbil Neocortex Strong spike rate adaptation was observed during depolarizing current steps  between 0.01 to 0.04 nA, which were maintained for durations of either 500 or 1000 ms. Depolarizing current steps of magnitudes greater than 0.04 nA resulted in sustained firing with weaker adaptation, and produced spike discharge frequencies between 50 and 200 Hz. Spike trains evoked by injected current stimuli were often followed by prolonged afterhyperpolarizations (AHP's), afterdepolarizations (ADP's) or ADP's followed by AHP's that were several millivolts in amplitude and several hundreds milliseconds in duration.  The afterpotential ratio (ADP:AHP) was approximately 1:6 for neurons  observed in this study. Figure 1A exemplifies a typical cell with the characteristic AHP following spike train evoked by strong depolarization. Application of ketamine (125 uM) did not significantly affect spike frequency (Figure 1 A), spike duration (Figure IB), and firing rate per input current (Figure 1C), although the AHP following the train was eliminated after ketamine application in 3 neurons.  3.2 Pharmacologically  Isolating AMPA/Kainate Receptor Channels  The specific NMDA receptor channel blocker, 2-amino-5-phosphonovalerate (APV, 100 uM) was included in the artificial cerebrospinal fluid (ACSF) to pharmacologically isolate AMPA/kainate receptor channels preparations.  in the tissue slice  Typical concentration values used for the complete block of NMDA  receptor channel synaptic contributions have been reported to range from 10 to 50 uM in hippocampal tissue slices (MacDonald et al, 1987,1989, 1990). We decided to raise the APV concentration to 100 \xM in order to ensure that sufficient molecules were available for antagonism of all receptors in spite of variations in tissue slices (e.g. thickness), in number of NMDA receptor channels present in the slice, and in the differences in sex and age of animals.  At resting and hyperpolarized potentials, ketamine is expected to  reinforce the APV block since its binding within the channel pore does not compete with the APV binding site, which is located on the extracellular side of the NMDA channel. A complete block of NMDA receptor contributions is imperative to unmask a ketamine effect on AMPA/kainate receptor channels.  18  However, as previously mentioned, ketamine effects are diverse and may include action at the non-NMDA receptors.  Therefore, the specific AMPA/kainate receptor  channel blocker, CNQX (6-cyano-7-nitro-quinoxalinedione), was used in conjunction with APV in order to identify AMPA/kainate receptor channel involvement in membrane potential changes in response to ketamine. Effective CNQX concentrations for complete block of non-NMDA receptor channels were reported to range from 400 nM to 10 uM in non-neural expression systems and hippocampal slices, respectively (Stein et al., 1992; Blake et al., 1988). We report that the use of APV at 100 ^M, and of CNQX at 20 ^M, completely blocked the slow and fast components of stimulus-evoked excitatory postsynaptic potentials (EPSPs) mediated by NMDA receptor and AMPA/kainate receptor channels, respectively. In Figure 2, for example, EPSPs were electrically evoked from the same neuron in the absence of drugs, followed by co-application of ketamine (200 ^M) and APV (100 uM), and co-application of ketamine (200 uM) and CNQX (20 uM). Qualitatively, as indicated by the arrows, ketamine reduced both the slow EPSP and, to a much lesser degree, the fast EPSP with little or no effect on the action potential. Subsequent application of APV specifically abolished the NMDA receptor-mediated slow EPSP, which resulted in isolating the AMPA/kainate receptor-mediated fast EPSP. The isolated fast EPSP was completely abolished with CNQX. Therefore, we included APV (100 ^M) in all ACSF solutions to pharmacologically isolate AMPA/kainate receptor channel conductances, and determine whether ketamine induced changes in these synaptic currents.  3.3 Ketamine Effects on Fast Excitatory Postsynaptic  Potentials  Electrical stimulation techniques were used to release presynaptic excitatory neurotransmitters to activate glutamate-dependent AMPA/kainate receptor channel conductances.  The resulting synaptic transmission was assessed by recording fast  excitatory postsynaptic potentials (fast EPSPs) during blockade of the slow, NMDA receptor-dependent  EPSPs with APV (100  uM).  Fast EPSPs were detected  postsynaptically and maintained subthreshold to spike firing with appropriate adjustments  19  Figure 2. EPSPs evoked from layer III pyramidal neuron by identical stimulation of afferent fibres in white matter. The arrows point to the following recording conditions: 1) ACSF control: the EPSP without antagonists in the bath. 2) ACSF + ketamine (KET); reduction of the E P S P after application of 200 uM ketamine. 3) ACSF + KET + APV: further application of 100 uM APV produced a complete block of the slow EPSP phase. 4) ACSF + KET + APV + CNQX: application of 20 uM CNQX blocked the fast phase as well, eliminating the EPSP.  20  to stimulation strengths.  Acquired data was collected subthreshold to spike firing in  order to minimize the complications due to conductances from depolarizing and repolarizing currents generated during action potentials. This eliminated the need for further channel blockade with pharmacological agents, such as tetrodotoxin (TTX). Fast EPSP peak amplitudes were reduced with increasing concentrations of ketamine. Peak amplitudes were measured (N = 9-21) from pre-stimulus baseline to peak membrane potentials (Figure 3) and were normalized to control (Figure 4).  In the  presence of relatively low ketamine concentrations of 6 to 25 uM, small increases and decreases in normalized peak amplitudes of fast EPSPs were observed. For example, neurons 1 and 4 produced increases in fast EPSP peak amplitude whereas neurons 2 and 3 produced decreases in the EPSP response (Figure 4).  These changes were not  statistically significant (P > 0.05), except for cell 1, which had a statistical significant increase in peak amplitude (P < 0.05). Higher ketamine concentrations (> 25 uM, to a maximum concentration 100 uM) consistently produced concentration-dependent reductions in fast EPSP peak amplitude (Table U, Figure 4). Estimated IC50 values for decreases in fast EPSP peak amplitudes were evaluated at half peak values and determined to range from 7 to 35 uM.  Table II. Ketamine-lnduced Reductions of Stimulus-Evoked Fast EPSP Amplitudes. Each concentration was applied for 3 minutes.  [Ketamine], jiM  0.0 6.3 12.5 25.0 50.0 100.0  Cell 1  EPSP Peak Amplitudes, mV SE SE SE Cell 3 Cell 2  10.8 12.9 * 11.4 8.0 * 4.1 * 3.1 *  0.2 0.3 0.2 0.3 0.3 0.1  V = -60.2 ±0.3 h  * P< 0.001 versus the value at 0.0  8.5 7.9 7.8 7.9 7.0 * 6.1 *  0.2 0.2 0.2 0.2 0.2  V = -60.2 ±0.1 h  8.0 7.1 6.8* 6.2 *  0.4  5.8* 3.8 *  V = -62.1 ± 0.1 h  Cell 4  0.2 0.2 0.2 0.1 0.2 0.1  8.2 8.5 4.8 * 4.2 * 3.4 * 3.9 *  SE 0.4  0.2 0.2 0.3 0.2 0.1  V = -60.1 ±0.1 h  Ketamine (Control) concentrations by Bonferroni  post hoc  test.  21  Fast E P S P  Time, ms  Figure 3. The reduction of APV-resistant fast EPSPs depended on the ketamine concentration. The concentration was increased sequentially in a cumulative manner, as shown to the left of the curves. The black lines show individual EPSPs (N = 10), the red lines the corresponding averages. Pre-stimulus resting potential: -60 ± 0.2 mV.  22  4. Variations in fast EPSP amplitude, after NMDA receptor blockade with 100 uM APV, depended on ketamine concentration. Each curve characterizes a different neuron under similar stimulus conditions. Data points and error bars show, respectively, the mean fast EPSP peak amplitudes and standard deviations. Approximate IC 's were estimated at half-peak values to range from 7 to 35 uJVI ketamine. At 100 u,M ketamine, reduction of fast EPSP reached asymptotic minimum values between 25 to 65% of control. Figure  50  23 However, ketamine, at concentrations up to 100 uM, failed to completely abolish the stimulus-evoked fast EPSP (N = 4).  The concentration / peak amplitude curves  appeared to approach asymptotic minimal values between 25 to 65%. The remaining fast EPSP response that is unaffected by high concentrations of ketamine is referred to in this thesis as a ketamine-resistant fast EPSP.  CNQX (20 ^M) completely abolished the  ketamine-resistant fast EPSP responses (99 to 100%; P < 0.05; Figures 2 and 5A; Table UJ; N = 3). Conversely, an initial application of CNQX completely inhibited a stimulusevoked fast EPSP response (99 to 100%; P < 0.05; Figure 5B; Table UJ; N = 3) and precluded any demonstration of ketamine effects on fast EPSP amplitudes. These results support the hypothesis that ketamine causes a partial blockade of AMPA/kainate receptor channels.  Table III. Ketamine-lnduced Increase in Input Resistance After Initial Application of Glutamate Receptor Blockers, APV and CNQX. Input Resistance, R| (%R| )  Treatment  t  CelM  ACSF+APV, 100 ACSF+APV+CNQX, 20 u.M ACSF+APV+CNQX+KET, 100 T  Cell2'  Cell3  Cell4  525 ( 0 ) 442 ( 0 ) 812 ( 0 ) 409 ( 0 ) 665 (27 ) 490 (11 ) 847 ( 4 ) 497 (22 ) 713 (36) 445 (0.7) 851 ( 5 ) 522 (28)  %R| = p e r c e n t i n c r e a s e in R| n o r m a l i z e d w i t h r e s p e c t to A C S F + A P V C o n t r o l ( P < 0.05)  " P > 0.05, decrease in R|  Ketamine's lipophilic nature and binding to non-selective sites made in vitro washout of its effects on fast EPSP peak amplitudes very difficult, especially following cumulative applications of the anaesthetic. Despite this technical difficulty, each neuron was able to generate action potentials in the presence of various ketamine concentrations in response to depolarizing current pulses (not shown).  Furthermore, non-monotonic  increases in resting input resistance (Rj) with cumulative ketamine concentrations were observed, and the resting membrane potential ( V R ) remained stable (see below). Together, these observations indicated that the neurons were viable and of good health throughout the recording period. More importantly, they suggested that the reduction in fast EPSP peak amplitudes were due to a drug effect.  24  B ACSF+APV, IOOuM  ACSF+APV+CNQX, 20nM  A C S F + A P V + C N Q X + K E T , IOOuM  Figure 5. The ketamine-sensitive EPSP that remained after NMDA receptor blockade depended on AMPA/kainate receptors in 2 neurons. (A) After NMDA receptor blockade with APV, application of ketamine reduced the amplitude of the EPSP, which was completely blocked with 20 uM CNQX. (B) After APV blockade, the EPSP was completely blocked by CNQX; ketamine had no further effect. Black lines show individual EPSPs (N = 6) and red line represents the average curves.  25  3.4 Recovery of Ketamine-lnduced  Effects  The problem of recovery of the ketamine effect may have been enhanced with drastically reduced availability of water-soluble proteins under in vitro conditions, which facilitate the removal of lipophilic compounds in vivo. Furthermore, continuous application of fresh solutions containing ketamine limited biotransformation of the drug into its water-soluble metabolites. However, MacDonald et al. (1990) reported recovery upon both continuous depolarization and agonist-induced activation of the NMDA receptor channel.  Channel activation induced open states and permitted release of  ketamine from the lumen of the pore of the NMDA receptor channel. Based on these findings, we employed another experimental protocol, at a single ketamine concentration, in attempt to achieve partial recovery from ketamine effects on the stimulus-evoked EPSP. Our experimental protocol involved repeating a programmed cycle of depolarization followed by electrical stimulation of afferent fibres in white matter. Depolarized current-steps were of one-second duration and of a fixed magnitude. For each neuron, the magnitude of the current-step was determined by trial until action potentials were evoked between 50 to 100 Hz. Following depolarization at an interval of 750 ms, stimulus-evoked release of excitatory amino acids resulted in evoked EPSP responses. The cycle was programmed to repeat every 5 seconds in order to prevent receptor desensitization until recovery from ketamine effects was observed. Recovery from ketamine effects was demonstrated in all of two neurons. Figure 6, for example, graphically displays the stimuli-evoked EPSP that is significantly reduced in the presence of ketamine (200 uM). In this case, the action potential associated with the EPSP had been abolished. However, recovery of the action potential and partial recovery of the EPSP were observed after 46 minutes of continuous depolarization and stimulation.  2 6  ACSF Control  -60 mV J /  ACSF + Ketamine 200 u.M  ACSF + Depolarization + Stimulation Ji  20 mV 100 ms  Figure 6. The action potential and EPSP were attenuated with application of 200 u,M ketamine for 3 minutes and recovered after a 46 min period of stimulating afferent fibres in white matter and of depolarizing the cell membrane to evoke firing at ~10Hz (see text). The evoked firing was a necessary condition for the recovery.  27  3.5 Lack of Presynaptic Ketamine Effects as Determined by Paired-Pulse Experiments Ketamine reduction in fast EPSP peak amplitudes could be the result of ketamine inhibition of presynaptic release of neurotransmitters. To investigate this possibility, we established a paired-pulse protocol in which two electrical stimuli were delivered to the white matter at a 500 ms time interval. Stimulus strength was maintained subthreshold to spike firing and was identical for both pulses. In the presence of 100 uM APV, the second EPSP response was reduced in peak amplitude to a value ranging from 0.56 to 0.86 of the first EPSP (N = 3). However, there was no significant difference in the ratios of second to first EPSP amplitudes before (1.5 ± 0.1) and after (1.4 ± 0.1) applications of various ketamine concentrations (0-100 uM, N = 3). Figure 7 shows an example of typical fast EPSPs in response to pair-pulse stimulation.  28  A C S F Control  A C S F + 100 nM KET  5mV 100  ms  Figure 7. Ketamine did not affect the release of transmitter from presynaptic terminals in response to the second stimuli. Ketamine reduced amplitude of second response to a value of 0.73 ± 0.09 mV of the first fast EPSP. Time between stimuli = 500 ms. Ratio between the second to first peak amplitudes before and after 100 uM ketamine application was not statistically significant (-1.4 ± 0.1; P < 0.05). V = -70 mV. Black lines show individual EPSPs and red line represents the average curves. R  29  3.6 Ketamine Effects on Input Resistance (Rj) Ketamine typically induced an increase in input resistance in conjunction with a concentration dependent decrease in fast EPSP peak amplitude. A hyperpolarizing Al magnitude between 0.01 and 0.04 nA, and pulse duration of either 500 or 1000 ms, produced hyperpolarized A V between 5 to 10 mV from V R . Cumulative applications of m  ketamine concentrations produced non-monotonic responses in normalized Rj, as represented in five neurons (Table IV). Input resistance, R;, increased (0.4 to 21% cf control; P < 0.05; N = 5) with ketamine applications of 25 to 31 |jM, and consistently peaked (18 to 86% of control; P < 0.05; N = 5) between 25 to 125 uM, followed by a significant decrease (3 to 56% cf control; P < 0.05; N = 4) with concentrations upwards to 250 \iM ketamine. Typically, the deflection in the curve towards decreasing values of Ri, following the peak resistance, did not return to control values, but remained comparatively higher (P < 0.05; N = 4). To summarize the data, R; was normalized to control and represented as percent resistance ( % R ) (Figure 8). Estimated IC50 values for increases in resistance were evaluated at half peak values and determined to range from 11 to 60 p.M. These measurements show that ketamine can produce large changes in membrane conductances after elimination of NMDA receptor channel function with APV-blockade.  Table IV. Input Resistance Changes in Response to Increasing Ketamine Concentrations. Treatment ACSF+100 u.M APV Control ACSF+APV+KET 25 M ACSF+APV+KET 31 M ACSF+APV+KET 50 M ACSF+APV+KET 63 M ACSF+APV+KET 100 M ACSF+APV+KET 125 uM ACSF+APV+KET 250 M ACSF+APV+KET+20 M CNQX  Celll 305 ( 0 )  U  U  371 (21 )  U  U  440 (44 )  U  568 (86 ) 475 ( 5 6 ) '  T  * %Ri = percent increase in R| normalized with respect to Control Student's t-tests: ' P < 0.05 versus the value of Control. P < 0.05 versus the value at Peak R>. * P < 0.05 versus the value of CNQX in bath. r  (%Rj ) Cejl4 Cell5 647 ( 0 ) 400 ( 0 ) 761 (18) ' 405 ( 4 ) #  722 (12 )'  n  615 (42 )' 567 ( 3 7 ) '  n  U  U  Input Resistance, Ri CeJI2 CeN3 388 ( 0 ) 232 ( 0 ) 241 ( 4 ) 389 (0.4) 313 (35 ) ' * 450 (16 ) 239 ( 3 ) * 463 (19 ) 417 ( 7 )* 247 ( 6 ) ' f  862 (33 ) '  571 (35 )'  30  Figure 8. Ketamine-induced increase in resistance varied in quantity and quality but depended on ketamine concentrations. Curves were drawn by eye through data points for five neurons, representing the relative resting membrane resistance change during ketamine applications. In the neurons previously exposed to CNQX (20 u,M) the resistance did not change (dashed line, N = 4). Approximate IC 's were estimated at half-peak values to range from 11 to 60 u,M ketamine. 50  31  The non-monotonic dependence of Rj on ketamine concentration suggests that several mechanisms of action may interact in these concentration-response curves. We considered the possibility that constitutively active AMPA/kainate receptor channels may contribute to the resting conductance. Ketamine effect on those channels would then be reflected in ketamine-induced resistance changes.  CNQX, a specific AMPA/kainate  receptor channel blocker, was therefore used in assessing a possible contribution of AMPA/kainate receptor channels.  The subsequent application of CNQX (20 p.M)  resulted in further increase in R; to values reaching a maximum of 21% (P < 0.05; N = 2; Table UJ). This demonstrated that, even in the presence of ketamine, a remarkable proportion of the membrane conductance was due to constitutively active CNQXsensitive ion channels. We asked if an initial blockade of those channels could present the ketamine-induced resistance increase in the absence of functional NMDA receptor channels. CNQX (20 p.M) was, therefore, included in the initial bath prior to ketamine (100 uM). In 4 neurons, CNQX application increased R by 4 to 27% (P < 0.05) (Table HI).  However, CNQX significantly abolished (96%, P < 0.05, N = 4) increases in Rj  induced by ketamine (100 uM) (red-dashed line in Figure 8; Table U). This suggested that a partial block of constitutively active AMPA/kainate receptor channels may contribute to ketamine-induced increases in resting input resistance.  3.7 Lack of Ketamine Effect on Holding Potential A constitutive activity of ketamine-sensitive AMPA/kainate receptor channels would be expected to influence the resting membrane potential and the current needed to maintain a hyperpolarization. Therefore, we registered the current required to maintain the membrane potential at -60 mV before and after application of ketamine, in the presence of APV. We used APV, at 100 (J.M, in all experiments to assess AMPA/kainate receptor channel mediated fast synaptic transmissions and observed their effects on the resting membrane potential. Resting membrane potentials were determined from a steady-state membrane potential reading taken at a time of 5 minutes or more after patch formation; that is, after the membrane potential had stabilized. An offset potential measured upon withdrawal and rupture of the seal at the completion of the recording was subtracted to  32  0.10  <  l i o  a i_ o Q .  >>  o 0.05 o a 3  a c  0.00 ACSF + 100 IJM A F V  ACSF+AFV + KET 6  ACSF+AFV + KET 13 \iM  ACSF+APV+ KET 25 uM  ACSF+AFV+ KET 50 \iM  ACSF+AFV+ KET 100  Figure 9. Increasing concentrations of ketamine did not induce a shift in resting membrane potential as derived from the input hyperpolarizing current (nA) required to maintain a V of -60 mV. The dashed line indicated a trend towards membrane hyperpolarization with increasing ketamine concentrations (P > 0.05). h  33  derive the resting membrane potential. Typical resting membrane potential ( V R ) readings were observed between -55 to -65 mV. Measurements of changes in V R were derived from DC current inputs required to maintain a steady V usually between -60 to -80 mV. m  Ketamine applications at low concentrations of 6 to 25 uM, produced no statistically significant shift in V (P > 0.05, N = 20, A V at 25 uM ketamine; 0.00 to 0.01 mV); R  R  however, a trend was observed towards hyperpolarizing potentials (Figure 9). Higher concentrations of ketamine (>50 to 100 pJVI) produced a trend in V towards depolarizing R  potentials, although the observed potential differences were not statistically significant (P > 0.05, n = 20, A V at 100 uM ketamine; 0.01 to 0.09 mV). R  3.8  Unmasking A Third Mechanism of Action by Ketamine The current/voltage plot of Figure 10A shows an increase in membrane resistance  (slope), most evident at strongly hyperpolarized potentials, produced by application of 100 uM ketamine. Since APV (100 p.M) had been added to the bath (ACSF), an NMDA receptor-blockade could not account for this resistance increase. Subsequent application of CNQX (20 pM) led to a small further slope increase, indicating that ketamine had not completely blocked AMPA/kainate receptor channels. When NMDA and AMPA/kainate glutamate receptor channels were both blocked, with addition of APV (100 uM) and CNQX (20 uM) to the ACSF, application of ketamine (100 pM) still produced an increase of voltage responses, particularly responses to strong hyperpolarizing current pulses (Figure 10B). This shows that, at least in the hyperpolarized range, there exists at least one additional conductance that is sensitive to ketamine. Thus, in addition to its blockade of NMDA receptor channels, and its reduction of glutamatergic synaptic transmission through AMPA/kainate receptor channels, ketamine reduced membrane conductance by mechanisms that have yet to be identified.  34  -30 - ACSF+AFV ACSF+AFV+KET100 (jM ACSF+AFV+KET+CNQX 20 pM  -40 -50  -0.12  0.08  B  -30 -•—ACSF+AFV - A — ACSF+AFV+CNQX, 20 (jM •  -0.12  ACSF+AFV+CNQX+KET, 100  •40 -50  0.08  Figure 10. Ketamine-induced increases in R| are prominent at greater hyperpolarized potentials when compared to control as represented in 2 different cells. (A) Ketamineresistant AMPA/kainate receptor channels is completely blocked by subsequent application of CNQX (20 uM) resulting in a 3 to 2 1 % (N = 2; Table IV) increase in Ri. (B) An additional 1 to 9 % (N = 3; Table III) increase in voltage amplitude with ketamine (100 pM) was unmasked after complete block of AMPA/kainate receptor channels with CNQX (20 pM). Ri changes are greater at hyperpolarized potentials.  35  4  DISCUSSION Clinically relevant concentrations required for ketamine anaesthesia were reported to  be tens of micromolar in cerebrospinal and plasma fluids, respectively (Cohen et al., 1973). At these concentrations, ketamine inhibits glutamate NMDA receptor-mediated synaptic responses, to which its molecular mechanism was attributed (MacDonald et al., 1990). The relative degree of ketamine-induced depression observed for NMDA and AMPA/kainate receptor-mediated responses is probably variable among neurons. Although, ketamine effects on different receptor channel families have been reported (as mentioned in the introduction), the effective concentrations required were comparatively high (>150 to 2300 pM), and may not have any bearing in a clinical setting. Mounting evidence  of AMPA/kainate receptor channel involvement in  nociception and its presence and wide distribution in central neural mechanisms of pain highlights the need to investigate the role of non-NMDA receptor channels in central pain perception. Ketamine's unique property of intense analgesia during anaesthesia is well documented, which then raises the question as to whether its analgesic effects could be attributed to actions at central AMPA/kainate receptor channels.  To the best of my  understanding, reports showing a direct effect of ketamine on central AMPA/kainate receptor channels have not yet been published. However, reports of ketamine-induced reduction of spinal cord nociception and successful treatment of patients experiencing phantom limb pain sensations have suggested to us that ketamine may act on central AMPA/kainate receptor channels.  Despite this possibility, reported literature has  attributed much of ketamine's actions to a use- and voltage- dependent blockade of the NMDA receptor channel. Considering the limited understanding of ketamine analgesia and anaesthesia, we endeavored to provide evidence in support of our hypothesis that ketamine affects synaptic transmission at the AMPA/kainate receptor channels, in addition to known effects at the NMDA receptor channels. The major findings of this current study are that ketamine depressed the amplitude of fast EPSPs mediated by AMPA/kainate receptor channels, indicating that ketamine effects on glutamate-dependent excitatory synaptic transmission are not restricted to NMDA receptor channels. In addition, ketamine concurrently induced non-monotonic  36  increases in resting input resistance that were due, in part, to a blockade of AMPA/kainate receptor channels.  4.1 Ketamine Depressed AMPA/Kainate Receptor Channel-Mediated Excitatory Postsynaptic Potentials (fast EPSPs)  Fast  Excitatory synaptic transmission mediated by glutamate receptors is known to comprise slow and fast components of EPSPs mediated by NMDA and AMPA/kainate receptor channels, respectively.  Previous electrophysiological studies reported a  ketamine block of the NMDA receptor mediated slow component of the EPSP (MacDonald et al., 1990). The current study investigated the role of AMPA/kainate receptors in ketamine actions on fast synaptic transmission of gerbil neocortex, including the auditory cortex.  Electrical stimulation was employed to release presynaptic  excitatory amino acids, including glutamate, and produce corresponding excitatory postsynaptic potentials (EPSPs). DL-2-amino-5-phosphonovaleric acid, APV, at 100 pM was effective for a complete block of the NMDA receptor-mediated slow component of the EPSP. The pharmacologically isolated AMPA/kainate receptor-mediated fast EPSPs were reduced by a ketamine blockade of the AMPA/kainate receptor channels. In our case, the stimulus-evoked fast EPSPs were typically strong and lasted several hundred milliseconds. This indicated the possibility of recruitment of strong synaptic inputs from multiple afferent fibers. Other possibilities for this observation could be a high density of the activated channels present in gerbil neocortex, or to the capacity for greater conductance per channel or longer channel open times. Low concentrations of ketamine (6 to 25 uM) produced variable changes in fast EPSP peak amplitudes (Figure 4). Our results exemplified four neurons having increases (N = 2) and decreases (N = 2) in evoked fast EPSP peak amplitudes with low ketamine concentrations. One of these neurons produced a statistical significant increase in fast EPSP peak amplitude. A definitive explanation for this observation is not available on the basis of our results. However, we can speculate that, at the time of drug application, small increases in cellular resistance outweighed the effects on synaptic efficacy resulting in a corresponding increase in membrane potential, in accordance with Ohm's Law. The gigaohm seals between the recording electrode and the somata were stable over this  37  recording period as established by our criteria and experience.  Therefore, we cannot  dismiss these observations as mere artifacts. Perhaps, the variability in EPSP amplitudes was due to a drug effect at the non-glutamate receptors, as previously mentioned. For example, ketamine has been reported to stimulate adrenergic receptors in cultured vascular smooth muscle cells and result in calcium-dependent phosphorylation and channel activation. Perhaps, ketamine activation of central adrenergic receptors could have induced cascade mechanisms that resulted in changed conductance through various ionotropic channels (Akata et al., 2001).  Such effect could either shunt or amplify  synaptic transmission mediated by the AMPA/kainate receptor channels.  However,  considerably higher ketamine concentrations were required for action at non-glutamate receptors.  The possibility of a concurrent ketamine action at the G A B A receptor A  channels may induce small hyperpolarizing conductances contributing to an inhibitory postsynaptic potential (IPSP) that may negate increases in EPSPs and result in a ketamine-induced decrease in evoked fast EPSPs (cf Little & Atkinson, 1984; Gage & Robertson, 1985, Wakasuki et al., 1999).  Alternatively, ketamine's block of  AMPA/kainate receptor channels might require activation to channel open-states in order for the molecule to gain access to the channel pore, similar to its action at the NMDA receptor channel. Despite variations in fast EPSP peak amplitudes in response to low ketamine concentrations, cumulative applications greater than 25 uM ketamine consistently decreased fast EPSP peak amplitudes in a concentration-dependent manner to asymptotic minimal values, ranging between 25 to 65% of the controls. The remaining ketamineresistant fast EPSP responses were completely abolished by the specific AMPA/kainate receptor channel blocker, CNQX. Conversely, an initial application of CNQX prevented any observations of effects with subsequent ketamine applications. The concentrationdependent decreases in CNQX-sensitive evoked fast EPSP were observed in neurons that were healthy and viable as determined by our stringent criteria (see Methods). In addition, these same neurons generated action potentials upon artificial depolarization of the membrane after each cumulative application of ketamine (not shown). The spikes were generally unaffected, with the exception of spike amplitude, after application of up to 125 uM ketamine (Figure 1, Table I). Furthermore, ketamine consistently induced an  38  increase in input resistance (discussed later), in contrast to a decrease typically observed in dying neurons. These considerations, therefore, led to the conclusion that ketamine partially attenuated synaptic currents mediated by active AMPA/kainate receptor channels. An incomplete block of CNQX-sensitive fast EPSP responses by ketamine suggested the existence of ketamine -sensitive and -insensitive channels, which is quite possible considering the numerous permutations available by virtue of the great diversity in AMPA/kainate receptor channel subtypes and the differential channel specificity governed of enantiomers of ketamine, S(+) and R(-) (White et al., 1980, 1982). Three possible mechanisms for ketamine reduction in AMPA/kainate receptor channel mediated synaptic transmission are 1) a reduction of presynaptic release of excitatory amino acid neurotransmitters, or 2) a decrease in postsynaptic conductance by means of channel blockade or 3) an artifactual decrease in resistance that shunts the conductances at the AMPA/kainate receptor channels (e.g. holes in cell membrane). Ketamine inhibition of neurotransmitter release at the presynaptic terminal was not supported with paired-pulse experiments. The results of these experiments consistently showed a reduction in the second fast EPSP peak amplitude following 500 ms delay after the first stimulation due to an expected partial exhaustion of neurotransmitter stores. However, the ratio between the first and second fast EPSP peak amplitude did not significantly change after ketamine application. These results did not point to a ketamine inhibition of neurotransmitter release at the presynaptic sites. Thus, we assumed that the ketamine reduction in AMPA/kainate receptor-mediated synaptic transmission occurred at the postsynaptic sites. Furthermore, intermittent monitoring of changes in resting input resistance after each applied concentration showed that ketamine consistently increased input resistance (see below) and therefore ruled out the possibility of a leaking patch where the integrity of the gigaohm seal has been compromised or involvement of physiological leak conductances.  4.2 Ketamine-lnduced  Increases in Input Resistance, Rj  Ketamine induced increases in resting input resistance in pyramidal neurons of the gerbil neocortex in the absence and presence of APV. We eliminated increases in resting resistance due to ketamine blockade of NMDA receptor' channels by including  39  APV (100 uM) in the bath applications. APV-insensitive increases in Rj suggested that the integrity of electrode seal was maintained, and that the cell's health was not a factor in the observed decrease in fast EPSP peak amplitudes, as mentioned above. Therefore, these increases in resting R; suggested a concentration-dependent ketamine effect. The ketamine-induced concentration-dependent (and APV-insensitive) changes in Ri in pyramidal neurons of gerbil neocortex may be summarized as follows.  Low  concentrations (25 to 31 pM) consistently induced increases in Rj to peak values of 18 to 86% of control that peaked between 25 to 125 uM. Subsequent decrease in Rj occurred at concentrations greater than 150 uM (Figure 7) that did not return to the control Rj values. Overall, as demonstrated in five cells, ketamine induced variable non-monotonic changes in Rj. This observation suggested the presence of more than one mechanism for the resistance change. Possible ketamine actions on non-glutamate receptors might contribute to this complex concentration/resistance function. Ketamine concentrations effective at nonglutamate channels were considerably higher (150 to 2300 uM) than the concentration range used to induce peak resistance changes (25 to 125 uM). The non-NMDA receptor channel antagonist, CNQX, was applied in the initial bath applications to eliminate AMPA/kainate receptor channel contributions to the changes in Rj. This paradigm nearly abolished (-96% block, Figure 3: red dashed-line; N = 4) ketamine-induced increases in Ri.  These results imply that ketamine acted at constitutively active AMPA/kainate  receptor channels, in a concentration-dependent manner. At concentrations between 6 and 31 uM of ketamine, both the decrease in fast EPSP peak amplitude and the increase in input resistance strongly suggested that one of ketamine's actions is a blockade of the AMPA/kainate receptor channel.  However, an explanation for the decrease in Rj  following the respective peak value at relatively higher ketamine concentrations requires further experimentation. A ketamine-induced activation of a K conductance could not have played a +  major role in the return to lower resistance values at higher ketamine concentrations, because of insignificant changes in resting membrane potentials (Figure 8). However, ketamine may have stimulated chloride currents mediated by the  GABAA  receptor  channels, in accordance with findings reported in the literature (Little & Atkinson, 1984;  40  Gage & Robertson, 1985; Wakasugi et al., 1999). The return in resistance towards control values without change in resting potential is consistent with the small driving force between the resting membrane potentials (approximately -60 mV) and chloride equilibrium potentials (Eci = -53 mV), as calculated from the Nernst Equation. We further postulated that ketamine-induced effects on resistance might not only depend on concentration, but also on time. For example, early applications of low concentrations of ketamine may have saturated available binding sites (and/or pores) on AMPA/kainate receptor channels during the cumulative concentration experiments.  As a result,  saturating concentrations of ketamine may not further raise input resistance but permit time-dependent activation of non-glutamatergic conductances, such as chloride currents, and thus reduce resistance. Ketamine-induced increases to peak resting resistances highly varied from neuron to neuron. These variations may be due to the number and presence of constitutively active AMPA/kainate receptor channels governing, in part, the level of excitability for each neuron. Ketamine blockade of active AMPA/kainate receptor channels in "excited" cells would have greater responses in resistance changes. Furthermore, each neuron may differentially express subunit proteins stoichiometrically forming various combinations of AMPA/kainate receptor channels, where each heterochimeric channel may have a unique sensitivity to the anaesthetic.  4.3 Recovery from Drug Effect and Effective Drug  Concentrations  The availability of ketamine at various binding sites in cortical tissue could follow a complicated time course, which may account for the difficulty in our experiments to eliminate ketamine actions on glutamate receptors during long wash periods. It is not impossible that time-dependent, accumulative binding contributes to the dramatic difference in ketamine concentrations required for non-glutamate (250 to 2300 pM) and glutamate receptor blockade (1 to 100 pM). The resistance decrease above 150 pM of ketamine in our cumulative concentration/resistance investigation could be a result of complicated time courses of ketamine-binding to its various binding sites.  Our  observations suggest that ketamine might also be trapped at the AMPA/kainate receptor channel, similar to NMDA receptor channel blockade, and requiring strong activation of  41  the neuron for its release. The plasma and brain plasma effective concentrations required to produce anaesthesia range in the tens of micromoles per litre for humans (Cohen et al., 1973), and are comparable to our finding, which is estimated to be 7 to 60 pM. The approximate 10 fold increase from low to high values of the concentration range could, in part, be a consequence of tissue binding.  4.4  The Mongolian Gerbil in vitro Model The functional organization of gerbil auditory cortex has been described in detail  to be comprised of physiologically identified fields, including primary (Al), anterior (AAF), dorsal (D), ventral (V), dorsoposterior (DP), and ventroposterior (VP) (Budinger et al., 2000). Briefly, these fields have extensive corticocortical connections as well as projections to and from the medial geniculate body, subgeniculate thalamic nuclei (including VPL and VPI nuclei), inferior colliculus, amygdaloid nuclei, caudate putamen, globus pallidus and pontine nuclei (Budinger et al., 2000).  The auditory cortex is,  therefore, widely connected with brain regions that likely contribute to pathological auditory sensations, including the phantom sensation, tinnitus.  Pathological pain  sensations, including phantom pain, depend probably on similarly widespread connections of the somatosensory cortex.  We presumed that the spinothalamic and  thalamocortical pathways in gerbils are involved in nociception, the perception of pain, as in the rat and human.  AMPA/kainate receptor channels certainly mediate sound  sensations just as these channels are mediating central pain perception.  There are  similarities between the pathophysiological conditions of tinnitus and phantom limb pain, as outlined by Simpson et al. (1999). Tinnitus is defined as "the perception of a sound that results from activity within the nervous system without any corresponding mechanical, vibratory activity within the cochlea" (Jastreboff, 1996). A similar definition can be applied to phantom limb pain, a pain that is centrally perceived despite the absence of the limb in which the pain is subjectively located (e.g. in amputated limbs). In both conditions, central perception of pain or sound could be caused by a plastic synaptic reorganization of sensory regions, including the cortex. Although we do not address questions directly related to tinnitus and phantom limb pain, we share the viewpoint of Simpson et al. (1999). We presumed that AMPA/kainate receptor channels present in the  42  auditory cortex mediate glutamate-activated synaptic transmission involved in central pain perception and thereby justify the use of an in vitro model using tissue slices of gerbil neocortex, including auditory fields. As a result, we included auditory cortex in tissue preparations throughout this experimental study. Perhaps, in the future, the use of a tone-induced tinnitic model (Ahissar et al., 1992; Vaadia et al, 1992) may help understand tinnitus / phantom limb pain, and conceptually extend the principles learned to that of central pain perception.  4.5 Functional  Significance  The ketamine-induced decrease in fast EPSP peak amplitude and concurrent increase in resistance by partial block of AMPA/kainate receptor channels support our hypothesis that the reduction of fast synaptic currents attenuates depolarizing shifts that contribute to voltage-dependent release of M g  2+  block of NMDA receptor channels. This  mechanism of action by ketamine may contribute to the rapid onset of analgesia. A reduction of fast synaptic currents by ketamine throughout the CNS can result in a lowered excitability of higher centers involved in the processing, perception, and memory of pain. However, ketamine can globally affect the brain and is not specific to neural mechanisms of pain. Therefore, a partial blockade of AMPA/kainate receptor channels may functionally be significant in reducing glutamate-dependent actions in pain perception, and yet provide adequate inward conductances sufficiently contributing to global excitation of other neural mechanisms, for example, those involved in learning and memory, and states of conscious awareness. As a result, the dissociative properties of ketamine may be a result of a fine balance of excitability between different neural mechanisms involved with sensory perception. This fine balance may involve effective concentrations of ketamine, ketamine activation of time-dependent mechanisms, and expression of ketamine sensitive- and resistant- AMPA/kainate receptor channels.  5  CONCLUSIONS Our hypothesis is that ketamine acts, not only via a selective block of NMDA  receptors, but also by a reduction of fast inward currents mediated by AMPA/kainate  43  receptor channels.  The functional consequence of a ketamine block of fast  AMPA/kainate receptor mediated EPSPs may serve to reduce the amplification of depolarizing shifts towards the threshold for NMDA receptor channel open-states. Our observations suggest that this phenomenon occurs in pyramidal neurons of gerbil neocortex, including the auditory cortex. Ketamine, therefore, causes a non-selective block of the two ionotropic glutamate receptor channel-types in a concentrationdependent manner. The results of this study indicate that ketamine partially blocks synaptic transmission mediated by AMPA/kainate receptor channels of gerbil neocortex in a concentration-dependent manner. IC50 values fall within the micromolar range, similar to concentrations during human anaesthesia, as reported by Cohen et al. (1973). However, ketamine analgesia and anaesthesia cannot, at present, be fully explained by a block of glutamate receptors, but may involve further mechanisms of action via non-glutamate receptors (e.g. GABAergic, cholinergic receptors).  44  6  REFERENCES  Adams JD, Baillie TA, Trevor AJ, Castagnoli N. Studies on the biotransformation of ketamine - identification of metabolites produced in vitro from rat liver microsomal preparations. Biomed. Mass Spec. 1981;8:527-538. Akata T, Izumi K, Nakashima M. Mechanisms of direct inhibitory action of ketamine on vascular smooth muscle in mesenteric resistance arteries. Anesthesiology 2001; 95:452462. Albe-Fessard D, Berkley KJ, Kruger L, Ralston HJI, Willis WD Jr. Diencephalic mechanisms of pain sensation. Brain Res. Rev. 1985;9:217-296. Ambalavanar R, Purcell L, Miranda M, Evans F, Ludlow CL. Selective suppression of late laryngeal adductor responses by N-methyl- D-aspartate receptor blockade in the cat J. Neurophysiol. 2002; 87: 1252-62. Amitai Y. Thalamacortical synaptic connections: efficacy, modulation, inhibition and plasticity. Reviews in the Neurosciences 2001;12:159-173. Anis NA, Berry SC, Burton NR., Lodge D. The dissociative anaesthetics, ketamine and phencyclidine, selectively reduce excitation of central mammalian neurons by N-methylD-aspartate. Br. J. Pharmacol. 1983; 79:565. Beggs JM, Moyer JR, McGann JP, Brown TH. Delayed synaptic integration in layer n/in pyramidal neurons of rat perirhinal cortex. J. Neurophysiol. 2000; 83:3294-3298. Beverley AO, Pennefather PS, MacDonald JF. Multiple mechanisms of ketamine blockade of N-methyl-D-aspartate receptors. Anesthesiology 1997; 86: 903-917. Blake JF, Brown MW, Collingridge GL. CNQX blocks acidic amino acid induced depolarizations and synaptic components mediated by non-NMDA receptors in rat hippocampal slices. Neuroscience Letters 1988;89:182-186. Bleakman D, Lodge D. Neuropharmacology of AMPA and kainate receptors. Neuropharmacology 1998; 37: 1187-1204. Budinger E, Heil P, Scheich H. Functional organization of auditory cortex in the Mongolian gerbil (Meriones unguiculatus). DJ. Anatomical subdivisions and corticocortical connections. Eur. J. Neurosci. 2000;12:2425-2451. Budinger E, Heil P, Scheich H. Functional organization of auditory cortex in the Mongolian gerbil (Meriones unguiculatus). IV. Connections with anatomically characterized subcortical structures. Eur. J. Neurosci. 2000; 12: 2452-2474. Chapman AG. Glutamate and Epilepsy. Journal of Nutrition 2000; 130: 1043S-1045S.  45  Cohen, M.L.; Chan, S.L.; Way, W.L.; Trevor, A.J. 1973 Distribution in the brain and metabolism of ketamine in the rat after intravenous administration. Aesthesiology 39: 370-376 Connors BW, Gutnick MJ. Intrinsic firing patterns of diverse neocortical neurons. Trends Neurosci. 1990;13:99-104. Connors BW, Gutnick MJ, Prince DA. Electrophysiological properties of neocortical neurons in vitro. J. Neurophysiol. 1982; 48: 1302-1320. Conseiller C, Benoist JM, Hammann KF, MacMaillard K, Besson JM. Effects of ketamine (CI-581) on cell responses to cutaneous stimulations in lamina rv and V in the cat's dorsal horn. Eur. J. Pharmac. 1972; 18:346-352. Corssen G, Domino EF. Dissociative anesthesia: further pharmacologic studies and first clinical experience with the phencyclidine derivative CI-581. Anesth. Analg. 1966; 45: 29-40. Domino EF. Neurobiology of phencyclidine (Sernyl), a drug with unusual spectrum of pharmacological activity. Int. Rev. Neurobiol. 1964; 6: 303-347. Evers AS, Crowder CM. "General Anesthetics" In Goodman & Gilman's: The Pharmacological Basis of Therapeutics, eds. Hardman JG, Limbird LE and Gilman AG. 2001; pp. 337-365. International: The McGraw-Hill Companies. Faulkner B, Brown TH. Morphology and physiology of neurons in the rat perirhinallateral amygdala area. J. Comp. Neurol. 1999; 411: 613-642. Finck AD, Ngai SH. A possible mechanism of ketamine-induced Anesthesiology 1982; 56: 291-297.  analgesia.  Friederich P, Benzenberg D, Urban BW. Ketamine and propofol differentially inhibit human neuronal K+ channels. Eur. J. Anaesthesiol. 2001; 18: 177-183. Gage PW and Robertson B. Prolongation of inhibitory postsynaptic currents by pentobarbitone, halothane and ketamine in CAI Pyramidal cells in rat hippocampus. Br. J. Pharmacol. 1985; 85: 675-81. Gilron I, Max MB, Lee G, Booher SL, Sang CN, Chappell AS, Dionne RA. Effects of the 2-amino-3-hydroxy-5-methyl-4-isoxazole-proprionic acid / kainate antagonist LY293558 on spontaneous and evoked postoperative pain. Clin. Pharmacol. Ther. 2000 68: 320-327. Hewitt DJ. The use of NMDA-receptor antagonists in the treatment of chronic pain. The Clinical Journal of Pain 2000; 16: S73-S79.  46  Hollman M, Heinemann S. Cloned glutamate receptors. Annu. Rev. Neurosci. 1994; 17: 31-108. Honey CR, Miljkovic Z, and MacDonald JF. Ketamine and phencyclidine cause a voltage-dependent block of responses to L-aspartic acid. Neuroscience Letters 1985; 61: 135-139. Irnaten M, Wang J, Venkatesan P, Evans C, Chang KSK, Andresen MC, Mendelowitz D. Ketamine inhibits presynaptic and postsynaptic nicotinic excitation of identified cardiac parasympathetic neurons in nucleus ambiguus. Anesthesiology 2002; 96: 667-674. Janig W. Neuronal mechanisms of pain with special emphasis on visceral and deep somatic pain. Acta Neurochir. Suppl. (Wien). 1987;38:16-32. Jansen KLR, Psych MRC. A review of the nonmedical use of ketamine: Use, users and consequences. Journal of Psychoactive Drugs 2000; 32(4): 419-433. Jastreboff PJ. In Mechanisms of Tinnitus, eds. Vernon JA and Moller AR. Allyn and Baker  1996; pp.74.  Kitahata LM, Taub A, Kosaka Y. Lamina-specific suppression of dorsal horn unit activity by ketamine hydrochloride. Anesthesiology 1973;38:4-11. Kress HG. S8-19.  Mechanisms of actions of ketamine. Der Anaesthesist 1997; 46 (Suppl 1):  Little HJ, Atkinson HD. Ketamine potentiates the responses of the rat superior ganglion to GABA. Eur. J. Pharmacol. 1984; 98: 53-59. MacDonald JF, Miljkovic Z, Pennefather P. Use-dependent block of excitatory amino acid currents in cultured neurons by ketamine. Journal of Neurophysiology 1987; 58(2): 251-266. MacDonald JF, Mody I, Salter MW. Regulation of N-methyl-D-aspartate receptors revealed by intracellular dialysis of murine neurons in culture. J. Physiol. 1989; 414: 17. MacDonald JF, Nowak LM. Mechanisms of blockade of excitatory amino acid receptor channels. Trends Pharmacol. 1990;11:167-172. McCormick DA, Connors BW, Lighthall JW, Prince DA. Comparative electrophysiology of pyramidal and sparsely spiney stellate neurons of the neocortex. J. Neurophysiol. 1985; 54: 782-806.  47  McGann JP, Moyer Jr JR, Brown TH. Predominance of late-spiking neurons in layer VI of rat perirhinal cortex. J. Neurosci. 2001; 21(14): 49699-4976. Minchin MC. The effect of anaesthetics on the uptake and release of gammaaminobutyrate and D-aspartate in rat brain slices. Br. J. Pharmacol. 1981; 73: 681-689. Mori H, Mishina M. Review: Neurotransmitter Receptors VUI. Structure and function of the NMDA receptor channel. Neuropharmacology 1995;34:1219-1237. Mycek MJ (ed). In Lippincott's Illustrated Reviews Pharmacology 2 Mycek MJ, Harvey RA, and Champe PC. 1997; p. 116 Lippincott-Raven.  nd  Edition, eds.  Nikolajsen L, Hansen PO, Jensen TS. Oral ketamine therapy in the treatment of postamputation stump pain. Acta Anaesthesiol. Scand. 1997; 41: 427-9. Nikolajsen L, Jensen TS. Phantom limb pain. Br. J. Anaesth. 2001 87: 107-16. Nozaki-Taguchi N, Yaksh TL. Pharmacology of spinal glutamatergic receptors in postthermal injury-evoked tactile allodynia and thermal hyperalgesia. Anesthesiology 2002; 96: 617-626. Oshima E and Richards CD. An in vitro investigation of the action of ketamine on excitatory transmission in the hippocampus of the guinea pig. Eur. J. Pharmacol. 1988 148: 23-33. Parada-Turska J, Turski W. Excitatory amino acid antagonists and memory: effect of drugs acting at N-methyl-D-aspartate receptors in learning and memory tasks. Neuropharmacology 1990; 29: 1111-1116. Pellicciari R, Costantino G, Macchiarulo A. Metabotropic glutamate receptors: structural view point. Pharmaceutica Acta Helvetiae 2000; 74: 231-237.  a  Procter MJ, Houghton AK, Faber ESL, Chizh BA, Ornstein PL, Lodge D, Headley PM. Actions of kainate and AMPA selective glutamate receptor ligands on nociceptive processing in the spinal cord. Neuropharmacology 1998; 37: 1287-1297. Price D. Psychological and neural mechanisms of affective dimension of pain. Science 2000; 288: 1769-1772. Reich DL and Silvay G. Ketamine: an update on the first twenty-five years of clinical experience. Can. J. Anaesth. 1989;36:186-197. Rogawski MA. The NMDA receptor, NMDA antagonists and epilepsy therapy: a status report. Drugs 1992; 44: 279-292.  48  Schneiderman JH, MacDonald JF. Excitatory amino acid blockers differentially affect bursting of in vitro hippocampal neurons in two pharmacological models of epilepsy. Neuroscience 1989; 31: 593-603. Smith DJ, Pekoe GM, Martin LL, Coalgate B. The interaction of ketamine with the opiate receptor. Life Sci. 1980; 26: 789-795. Sarton E, Teppema LJ, Olievier C, Nieuwenhuijs D, Matthes HWD, Kieffer BL, Dahan A. The involvement of the mu-opioid receptor in ketamine-induced respiratory depression and antinociception. Anesth. Analg. 2001;93:1495-1500. Simpson JJ, Davies WE. Recent advances in the pharmacological treatment of tinnitus. TIPS 1999; 20: 12-18. Sinclair JG, Tien AF. Neuronal responses to ketamine administered microiontophoretically or intraperitoneally in the rat. Gen. Pharmac. 1979; 10: 51-55. Stannard CF, Porter GE. Ketamine hydrochloride in the treatment of phantom limb pain. Pain 1993;54:27-30. Stein E, Cok JA, Seeberg PH, Verdoorn TA. Complex pharmacological properties of recombinant AMPA receptors subtypes. Mol. Pharmacol. 1992; 42: 864-871. Steriade M, Jones EG, Llinas RR. Thalamic Oscillations and Signaling. 1990. pp. 141237. Wiley, New York. Stuart GJ, Dodt HU, Sakmann B. Patch-clamp recordings from the soma and dendrites, of neurons in brain slices using infrared videomicroscopy. Pflugers Arch. 1993; 423: 511-518. Suh H, Song D, Huh S, Kim Y. Differential potentiative effects of glutamate receptor antagonists in the production of antinociception induced by opioids administered intrathecally in the mouse. Brain Research Bulletin 2000; 52: 143-150. Szekely JI, Torok K, Mate G. The role of ionotropic glutamate receptors in nociception with special regard to the AMPA binding sites. Current Pharmaceutical Design 2002; 8: 887-912. Tsai G, Coyle JT. Glutamatergic mechanisms in schizophrenia. Ann. Rev. Pharmacol. Toxicol. 2002; 42: 165-79. Wagner LE, Gingrich KJ, Kulli JC, Yan J. Ketamine blockade of voltage-gated sodium channels. Anesthesiology 2001; 95: 1406-1413.  49  Wakasugi M, Hirota K, Roth SH, Ito Y. The effects of general anesthetics on excitatory and inhibitory synaptic transmission in area CAI of the rat hippocampus in vitro. Anesth. Analg. 1999; 88: 676-680. Watkins JC, Evans RH. Excitatory amino acid transmitters. Annu. Rev. Pharmacol. Toxicol. 1981; 21: 165-204. White EL, Keller A. Cortical circuits : synaptic organization of the cerebral cortexstructure, function, and theory. 1989. pp. 20-54. Birkhauser. White PF, Way WL, Trevor AJ. Ketamine - its pharmacology and therapeutic uses. Anesthesiology 1982;56:119-136. White PF, Ham J, Way WL, Trevor AJ. Pharmacology of ketamine isomers in surgical patients. Anesthesiology 1980;52:231-9. Winters WD, Ferrar-Allado T, Guzman-Flores C, Alcaraz M . The cataleptic state induced by ketamine: a review of the neuropharmacology of anesthesia. Neuropharmac. 1972; 22:303-315. Yamamura T, Harada K, Okamura A, Is the site of action of ketamine anesthesia the N-methyl-D-aspartate receptor? Anesthesiology 1990 72:704-710. Zhou Z, Zhao Z. Ketamine blockage of both tetrodotoxin (TTX)-sensitive and TTXresistant sodium channels of rat dorsal root ganglion neurons. Brain Research Bulletin 2000; 52: 427-433.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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


Related Items