Follow this and additional works at:

Transcription

1 The University of Toledo The University of Toledo Digital Repository Theses and Dissertations 2007 The role of L-type voltage-gated calcium channels in hippocampal CA1 neuron glutamate and GABA-A receptor-mediated synaptic plasticity following chronic benzodiazepine administration Kun Xiang Medical University of Ohio Follow this and additional works at: Recommended Citation Xiang, Kun, "The role of L-type voltage-gated calcium channels in hippocampal CA1 neuron glutamate and GABA-A receptormediated synaptic plasticity following chronic benzodiazepine administration" (2007). Theses and Dissertations This Dissertation is brought to you for free and open access by The University of Toledo Digital Repository. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of The University of Toledo Digital Repository. For more information, please see the repository's About page.

2 Health Science Campus FINAL APPROVAL OF DISSERTATION Doctor of Philosophy in Biomedical Sciences Examination Committee The role of L-type voltage-gated calcium channels in hippocampal CA1 neuron glutamate and GABAA receptor-mediated synaptic plasticity following chronic benzodiazepine administration Submitted by: Kun Xiang In partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biomedical Sciences Major Advisor: Elizabeth Tietz, Ph.D. Academic Advisory Committee: L. John Greenfield, M.D., Ph.D. Zi-Jian Xie, Ph.D. Howard C. Rosenberg, M.D., Ph.D. David Giovannucci, Ph.D. Senior Associate Dean College of Graduate Studies Michael S. Bisesi, Ph.D. Date of Defense: May 16, 2007

3 The role of L-type voltage-gated calcium channels in hippocampal CA1 neuron glutamate and GABA-A receptor-mediated synaptic plasticity following chronic benzodiazepine administration Kun Xiang Cellular and Molecular Neurobiology Program University of Toledo College of Medicine 2007

4 ACKNOWLEDGEMENTS I am deeply grateful to Dr. Elizabeth I. Tietz, my major advisor, for her commitment to these dissertation studies and for providing academic guidance, intellectual encouragement and patience for the completion of these dissertation studies. I would like to thank Drs. L.John Greenfield, Jr., Zi-Jian Xie, Howard C. Rosenberg and David Giovannucci for participating on my dissertation committee and for providing insightful comments about these dissertation studies. I would like specifically thank Dr. L.John Greenfield, Jr. for providing me the equipment and advice that were critical for my dissertation study. I would like specifically thank Dr. David Giovannucci for all academic and personal assistance over the past years. I would like to thank for Martha Heck for all secretarial and personal assistance during my graduate years. William C. Ferencak, III performed radioligand binding studies for the nimodipine brain concentration measurement. ii

5 Lastly, I would like to thank my wife, Janice M. Xiang for all the support, encouragement and love which made the completion of these studies possible. iii

6 TABLE OF CONTENTS INTRODUCTION..1 LITERATURE REVIEW.11 MANUSCRIPT MANUSCRIPT MANUSCRIPT SUMMARY/DISCUSSION.178 BIBLIOGRAPHY.194 APPENDICES..222 ABSTRACT..229 iv

7 INTRODUCTION BZs (benzodiazepines) are widely prescribed as anxiolytics, hypnotics and anticonvulsants. However clinicians require caution prescribing BZs for a prolonged period of time because of the development of functional tolerance, a reduction in response to the BZs actions during chronic administration (File 1985; Rosenberg, 1995); and dependence, manifest as the appearance of withdrawal symptoms including anxiety, insomnia, delirium and seizures after cessation of drug administration (Griffiths and Weerts, 1997; Schweizer and Rickles, 1998). To avoid withdrawal symptoms, some patients who are using BZs for medical purposes begin to abuse their medication (Griffiths and Weerts, 1997; Griffiths and Johnson, 2005). Thus, understanding the mechanism of BZ physical dependence may improve their clinical application as well as reduce the possibility of BZ abuse among patients. BZs are positive allosteric modulators of γ-amino butyric acid type A receptors (GABARs) and enhance GABA affinity for the receptor resulting in an increased frequency of chloride (Cl - ) channel opening and potentiation of inhibitory responses within the central nervous system (CNS) (Study and Barker, 1981; Chan and Farb, 1985). GABA is the major inhibitory neurotransmitter in the CNS and mediates fast inhibitory neurotransmission by activating the GABAR. Native GABARs are pentameric assemblies of subunit proteins (α 1-6, β 1-4, γ 1-3, ρ 1-3, σ, π, ε and θ) with an integral Cl - channel (Macdonald and Olsen, 1994; Costa et al., 2002). Studies in heterologous expression systems showed that a combination of α-, β-, γ-subunit variants are required for the expression of fully functional GABARs (Ducic et al., 1995). The presence of a γ 1

8 subunit is required for BZ actions and the particular α- and γ-subunit isoforms present largely determine the binding and functional properties associated with the benzodiazepine site (Ducic et al., 1995). Recombinant receptors containing α subunit variants (α1-α6) display different benzodiazepine pharmacologies, including anticonvulsant and hypnotic actions mediated by α1 subunit-containing GABARs and anxiolytic actions by α2 subunit-containing GABARs, in particular in cortex and hippocampus (Mohler et al., 2002). Since BZ exert their effects by acting at their binding site on the GABAR, development of tolerance to BZ actions has been extensively studied in GABAergic inhibitory systems. Changes in GABAR structure and function in different brain areas following chronic BZ administration have been well described including decreased BZ and GABA agonist binding affinity, decreased number of BZ binding sites, reduced allosteric coupling between GABA and BZ binding sites, modulation of GABAR subunit mrna and protein expression, changes in GABAR-mediated postsynaptic potential and currents coupled with a reduction of GABAR channel conductance, et al. (Heninger et al., 1990; Gallager et al., 1991; Xie and Tietz, 1992; Zhao et al., 1994; Barnes, 1996; Hutchinson et al., 1996; Impagnatiello et al., 1996; Pesold et al., 1997; Zeng and Tietz, 1999; Costa et al., 2002). Recently attention has been focused on the role that excitatory glutamergic receptors may play in mediating benzodiazepine dependence. Evidence that pharmacologic antagonism of α-amino-3-hydroxy-5-methylisoxasole-4-propionic acid receptors 2

9 (AMPARs) and N-methyl-D-aspartate receptors (NMDARs) differentially modify behavioral manifestations of benzodiazepine withdrawal (Steppuhn and Turski, 1993; Koff et al, 1997; Dunworth and Stephens, 1998; Van Sickle et al, 2004) and that AMPAR and NMDAR binding and subunit expression in rat hippocampus and frontal cortex are regulated following withdrawal from benzodiazepines (Allison et al, 1999; Izzo et al, 2001; Van Sickle and Tietz, 2002; Van Sickle et al, 2002) suggests that glutamatergic receptors may also play important roles in mediating BZ withdrawal symptoms following chronic BZ administration. Ionotropic glutamate receptors belong to excitatory ion channel families and mediate the majority of excitatory neurotransmissions in the mammalian central nervous system. Ionotropic glutamate receptors bind neurotransmitters released from pre-synaptic neurons, mainly glutamate, and activated channels allow cations such as Na + and Ca 2+ to pass through a channel in the center of the receptor complex. This flow of ions results in the plasma membrane depolarization and the generation of an electrical current that is propagated down the neuron processes (dendrites and axons) to the next neuron in line. AMPARs mediate fast synaptic transmission in the CNS and are composed of subunits GluR1-4, products from separate genes. They are ligand-gated and mediate the fast component of the excitatory postsynaptic potentials (EPSPs) (τ=7-12 ms) and have low Ca 2+ permeability (Hume et al., 1991). NMDA receptors are composed of assemblies of NR1 subunits and NR2 subunits, which can be one of four separate gene products (NR2A-D). They are ligand- and voltage-gated ion channels responsible for the slow component of the EPSC (τ= ms) and are characterized by slower kinetics, high 3

10 permeability to Ca 2+, voltage-dependence for removing Mg 2+ blockage and require glycine as a co-agonist (Mayer et al., 1984; Ascher and Nowak, 1998). NMDA receptor activation leads to a calcium influx into the post-synaptic cells, a signal thought to be crucial for the induction of NMDA-receptor dependent synaptic plasticity, including long term potentiation and long term depression. Although the exact neural mechanisms underlying the development of BZ tolerance and GABAR structural and functional changes remain incomplete, new evidence indicates a role of L-type voltage-gated calcium channel (VGCC) in mediating BZ tolerance and dependence associated with chronic BZ administration. BZs can directly inhibit VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Gershon, 1992; Reuveny E. et al, 1993; Ishizawa Y et al, 1997) and may modify L-type VGCC function following chronic FZP administration (Xiang et al., 2007). L-type VGCCs have been reported to regulate calcium signaling associated with synaptic plasticity related to dependence on several psychostimulants. (Kelley 2004; Rajadhyksha and Kosofsky, 2005) and also contribute to withdrawal symptoms associated with other GABA A receptor positive allosteric modulators, including ethanol and barbiturates (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999; Watson and Little, 2002). GABAinduced GABAR down-regulation has been suggested to be the product of a transcriptional repression of GABAR subunit genes (Lyons HR et al., 2000; Russek et al., 2000) that depends on activation of L-type VGCC (Lyons HR et al., 2001; Gravielle MC et al., 2005). Raising intracellular Ca 2+ and Ca 2+ /calmodulin-dependent protein kinase have been shown to regulate GABAR-mediated Cl - current (Wang et al., 1995; 4

11 Koninck and Mody, 1996; Aguayo et al., 1998; Churn and DeLorenzo et al., 1998; Alix et al., 2002). Recent studies suggest that administration of L-type VGCC antagonists can also interrupt withdrawal symptoms after ending benzodiazepine treatment (Gupta et al, 1996; Podhorna, 2002; Ganouni et al, 2004, Cui et al, 2006). Indeed, as reported in other forms of neuronal plasticity, acute systemic injection of an L-type VGCC antagonist, nimodipine averted hippocampal CA1 neuron AMPAR current enhancement and anxietylike behavior in benzodiazepine-withdrawn rats (Xiang and Tietz, 2007). Together these observations raise the possibility that chronic benzodiazepine administration might modulate L-type VGCC activity and downstream intracellular Ca 2+ -dependent signaling mechanisms, thus regulate both GABAergic and glutamatergic synaptic plasticity associated with benzodiazepine tolerance and dependence. At least nine types of VGCCs have been identified in neurons for different physiological functions, like synaptic plasticity, neuronal survival, synaptic transmission and neuro-muscular junction activity (Catterall, 2000). Compared to other types of VGCCs, the L-type is distinguished by high voltage of activation, large single-channel conductance, slow voltage-dependent inactivation, marked regulation by campdependent protein phosphorylation pathways, and specific inhibition by Ca 2+ antagonist drugs including dihydropyridines, phenylalkylamines, and benzothiazepines (Catterall, 2000; Lipscombe et al., 2004). The L-type VGCCs are multimeric complexes of a poreforming α1 subunit of ~ kda; a transmembrane, disulfide-linked complex of α2 5

12 and δ subunits; an intracellular β subunit; and in some cases a transmembrane γ subunit (Catterall WA, 2000). Ten α1 subunits, four α2δ complexes, four β subunits and two γ subunits are known (Catterall WA, 2000). The α1 subunit contains the Ca 2+ pore and binding sites for selective channel antagonists. It is composed of four homologous repeats (I-IV) each comprising six transmembrane segments (S1-S6) (Walter and Messing, 1999). Neuronal L-type channels contain α1 C or α1 D subunits, mainly localized to neuronal cell bodies and proximal dendrites, and are thought to regulate biochemical processes such as protein phosphorylation and gene expression in the cell soma (Johnston et al., 1992; Walter and Messing, 1999). The objective of these dissertation studies was to evaluate the role of L-type VGCC in the expression of BZ tolerance, as well as dependence manifested as withdrawal-anxiety, following chronic benzodiazepine treatment. Initially, whole-cell electrophysiological recordings of AMPAR- and NMDAR-mediated currents were used to examine the correlation between glutamate receptor-mediated function in hippocampal CA1 neurons, and BZ withdrawal-anxiety in rats withdrawn from 1-week flurazepam (FZP) administration, measured by an elevated plus-maze. Following pharmacological antagonism of AMPA, NMDA and VGCC function, a role for regulation of L-type VGCCs, but not NMDARs, in mediating enhanced AMPAR function and BZ withdrawal anxiety was identified. Further studies concentrated on the regulation of L-type VGCC function following chronic BZ administration using whole-cell voltage-clamp method. The temporal pattern of the regulation of L-type VGCC function was evaluated following 6

13 chronic BZ administration. The direct concentration- and use-dependent effect of the benzodiazepines to affect VGCC binding and function was also investigated. Based on the evidence of L-type VGCC function changes and its role in mediating AMPAR synaptic plasticity following chronic FZP administration, the relationship between L-type VGCC and GABAR was further explored. Pharmacological antagonism was again applied for investigating the role of L-type VGCC in mediating GABAR function using whole-cell electrophysiological recordings of GABAR-mediated currents. GABAR channel kinetics and single channel conductance was evaluated after L-type VGCC antagonist application both in vivo and in vitro. These studies would give the insight of an L-type VGGG-dependent Ca 2+ signaling mechanism in differentially mediating specific GABAR changes as previously mentioned. The first manuscript was focused on the functional regulation of glutamate receptors on hippocampal CA1 neurons associated with the role of L-type VGCCs after prolonged BZ administration. CA1 neuron hyperexcitability was evident from the significant increase in the frequency of extracellular, 4-aminopyridine (4-AP)-induced spike discharges in slices from 1-day FZP-withdrawn rats. A transient enhancement of AMPAR-mediated miniature excitatory postsynaptic currents (mepscs) and a reduction in NMDAR-mediated evoked (e)epscs were observed (Van Sickle 2004) in 2-day FZPwithdrawn rats. In this investigation the relationship between increased CA1 neuron AMPAR function and anxiety-like behavior in rats was assessed following pharmacological antagonism of glutamate receptor activity at the onset of withdrawal. 7

14 Next, systemic AMPA antagonist injections in other groups of rats was used to assess the role of AMPA plasticity in mediating subsequent effects on CA1 neuron NMDAR function. Further, by means of pharmacological antagonism of either NMDAR or L-type VGCC activity the possibility of a Ca 2+ -mediated mechanism underlying enhanced AMPAR activity during benzodiazepine withdrawal was explored. Finally, to serve as the foundation for more mechanistic studies, a model of the cellular mechanisms underlying BZ withdrawal anxiety was proposed based on the present findings, and on past evidence from our laboratory and others. The second manuscript concentrated on the study of L-type VGCC function following chronic BZ administration. It was postulated that chronic BZ administration modified L-type VGCC function and thus underlies AMPAR plasticity in CA1 pyramidal neurons in benzodiazepine-withdrawn rats. The properties of L-type VGCC-dependent Ca 2+ currents were investigated using whole-cell voltage-clamp methods in isolated CA1 pyramidal neurons. The temporal pattern of changes in CA1 neuron Ca 2+ current density was evaluated following acute desalkyl-fzp and chronic FZP administration respectively. In addition, it was examined whether nimodipine pre-injection would affect enhanced AMPAR-mediated mepsc amplitude in 2-day FZP withdrawn rats as reported previously (Xiang and Tietz, 2007). Finally, the direct concentration and use-dependent effect of the benzodiazepines to affect VGCC binding and function was evaluated. Collectively, the findings suggest that chronic FZP administration transiently enhances calcium entry through L-type VGCCs during FZP withdrawal with a time-course consistent with the possibility that a downstream VGCC-mediated intracellular calcium 8

15 signaling cascade may contribute to AMPAR plasticity and benzodiazepine withdrawal anxiety. Based on the evidence of L-type VGCC function modulation following chronic FZP administration, and its role in mediating AMPAR-mediated synaptic plasticity, it was postulated that L-type VGCC is involved in GABAR changes following chronic FZP administration. Thus the third manuscript investigated the role of L-type VGCC regulation in mediating GABAR functional plasticity. By means of pharmacological antagonism of VGCC activity, the possible role for a Ca 2+ -mediated mechanism underlying impaired GABAR activity following benzodiazepine administration was explored. Furthermore, nimodipine s effect on in vitro tolerance to zolpidem, a GABAR α1-subunit selective BZ receptor ligand, on GABA A receptor-mediated miniature inhibitory post-synaptic currents (mipscs) kinetics was also evaluated. GABAR unitary channel conductance was reduced by 1-week FZP treatment, which may explain the reduction of GABAR-mediated mipsc amplitude (Xu and Tietz, 1999). An L-type VGCC dependent signaling may regulate GABAR channel conductance by changing channel subunit composition or phosphorylation of specific site. Thus effects of in vivo administration of L-type VGCC antagonist nimodipine on GABAR single-channel conductance was further investigated by non-stationary variance analysis. Concentration response study between nimodipine and GABAR-mediated mipscs was also investigated to evaluate if there is a direct nimodipine and GABAR interaction. Evaluating the differential effects of VGCC antagonist might provide some insight into the role of L-type VGCC-mediated Ca 2+ influx in mediating GABAR function and 9

16 mechanisms underlying GABAR changes associated with BZ tolerance, and perhaps dependence. Regulation of L-type VGCC function in the hippocampal CA1 neurons following chronic FZP administration, and its role in mediating AMPAR and GABAR functional plasticity in the context of BZ dependence and tolerance, was examined in these studies using a variety of electrophysiology, neurochemistry and behavioral studies both in vivo and in vitro. These findings provided evidence that enhanced L-type VGCC function and Ca 2+ entry after chronic FZP administration contribute to the AMPAR plasticity and CA1 neuron hyperexcitability that correlates with anxiety-like behavior in rats, while NMDAR plays a compensatory role to blunt the effects of enhanced CA1 neuron AMPAR function. L-type VGCC also, in part, mediates GABAR functional modifications following chronic FZP administration, although other GABAergic mechanisms also appear to be involved. Taken together, localized enhanced Ca 2+ entry through L-type VGCC may represent a common neurophysiologic mechanism to a variety of drugs of abuse as for the expression of BZ dependence and tolerance. 10

17 LITERATURE REVIEW Clinical Significance of Benzodiazepine Tolerance and Dependence The benzodiazepines are widely prescribed for the treatment of anxiety and insomnia and as adjunct treatment for seizure disorders. These agents are extremely safe, but tolerance, (i.e. a decrease in the ability of the drug to produce the same degree of pharmacological effect during prolonged exposure, or the need to increase drug intake to produce the same degree of effect), develops rapidly to their sedative activity, while their potent anticonvulsant effects are compromised within one or two months (Bateson 2002). Tolerance to BZ anxiolytic and hypnotic actions rarely develops (Mattson 1974; Davis and Gallager, 1988). Another problem of prolonged administration of benzodiazepine is dependence, sometimes referred to as physical dependence, wherein discontinuation of the drug following repeated exposure generates a characteristic withdrawal syndrome. Re-administration of the same drug relieves these syndromes. On abrupt withdrawal from benzodiazepine exposure, patients can experience a number of physical and psychological symptoms indicative of a dependent state and physical aspects of the withdrawal phenomena that can be reproduced in animals, including anxiety, insomnia, agitation, increased sensitivity to light and sound, muscle cramps, and even seizures and delirium after repeated high-dose use (Griffiths and Weerts, 1997). Some patients who are using benzodiazepines for medical purposes also begin to abuse their medication in order to avoid withdrawal symptoms. Thus a typical drug-dependence syndrome may underlie misuse of benzodiazepines among patients undergoing treatment, as well as contribute to their recreational abuse (Griffiths and Weerts, 1997). 11

18 Functional Anatomy of the Hippocampus The Hippocampus--An Important Locus for Benzodiazepine Actions. The ability of benzodiazepines to enhance neuronal inhibition in this seizure-prone structure (Semyanov, 2003) makes the hippocampal formation of significant importance for benzodiazepine actions (Haefely, 1985), in particular their anticonvulsant actions (Xie and Tietz, 1992). The sensitivity of hippocampal GABA A receptors to benzodiazepine modulation depends on receptors containing α1, α2, α3, or α5 subunits together with a γ subunit, all of which are expressed within hippocampus (Mohler et al., 2002). These facts support an important role for the hippocampus in benzodiazepine anticonvulsant actions through direct manipulation of GABA A receptor-mediated hippocampal inhibition. Moreover, transient glutamate plasticity in hippocampal pyramidal neurons contributes to benzodiazepine withdrawal anxiety in rats (Van Sickle et al., 2004; Xiang and Tietz, 2007), indicating that hippocampus is an important locus for the investigation of benzodiazepine actions following both acute and chronic administration. Since these entire dissertation studies were conducted in hippocampal CA1 pyramidal neurons, the next section will review the anatomy and physiology of hippocampus. Hippocampus Gross Anatomy The hippocampal formation is a crescent-shaped structure lying in each temporal lobe on the medial wall of the lateral ventricle. Its resemblance to a seahorse prompted the use of its Greek-derived name and it has also been referred to as Ammon s horn for its resemblance to a ram s horn (Amaral and Witter, 1995; Giap et al., 2000). The 12

19 hippocampal formation is composed of the hippocampus proper, dentate gyrus and subiculum, all of which are composed of three distinct layers. The hippocampus and subiculum each have a pyramidal layer, containing pyramidal cell bodies; a molecular layer, consisting of apical dendrites from pyramidal cells, along with axons originating from various neurons in other regions; and a polymorphic layer, containing various interneurons, including basket cells. The dentate gyrus has a granule cell layer, a molecular layer consisting of the apical dendrites of the granule cells, and a polymorphic layer. Pyramidal cells are efferent neurons, whereas the granule cells are interneurons. Basket cells are inhibitory neurons that synapse on pyramidal cells (Giap et al., 2000). The hippocampus proper was further divided into three sub-regions, including CA1, CA2 and CA3, based upon the morphology of the principal neurons present in each sub-region (Amaral and Witter, 1995). The CA1 region in the superior part of hippocampus consists of triangular shaped pyramidal neurons. The CA3 region is located at the curve of the hippocampus, adjacent to the dentate gyrus, and consists of larger pyramidal neurons. The small region between CA1 and CA3 is designated CA2 region (Knowles, 1992). Neuronal Connections within Hippocampus Early neuroanatomical and electrophysiological studies suggested that the hippocampus is organized in parallel lamellae, due to its neuronal connections and vascular supply (Lopes da Silva and Amolds, 1978; Amaral and Witter, 1989). While recent work has been more consistent with viewing the hippocampus as a series of threedimensional networks of connections versus isolated, independent processing units (Amaral and Witter, 1995; Moser and Moser, 1998), the functional trisynaptic circuit in 13

20 a transverse hippocampal slice continues to provide an excellent experimental model for the study of synaptic regulation. The trisynaptic pathway is composed of perforant path afferents, mossy fibers and Schaffer collateral pathway. Perforant path afferents are composed of axons of entorhinal cortex neurons, which make the first synaptic connection with dendritic spines of granule cells in the dentate gyrus. Mossy fibers are the axons of dentate granule neurons traveling through the hilus of the dentate fascia and form the second synapse with apical dendrites of pyramidal cells with CA3 region. Bifurcating axons of CA3 neurons leave the hippocampus via the fimbria and the other fibers synapsing on ipsilateral CA1 pyramidal neurons (Schaffer collateral pathway) or contralateral CA1 cells (commissural pathway) (Anderson et al., 1971; Lopes da Silva and Amolds, 1978; Amaral and Witter, 1995). Microanatomy of the Hippocampal CA1 Region The CA1 region of the hippocampus is composed of six layers that include the stratum moleculare, stratum lacunosum, stratum radiatum, stratum pyramidale, stratum oriens and the alveus. Pyramidal neurons are about 20-25µm in diameter and are tightly packed in the stratum pyramidale. The basal dendritic tree of pyramidal neuron projects toward the stratum oriens, while the apical dendrites project to the strata radiatum, lacunosum and moleculare (Amaral and Witter, 1995). Although there are a variety of non-pyramidal, GABAergic interneurons that are considered to mediate tonic, feedforward and feed-back inhibition of CA1 pyramidal neurons, the majority afferent input to the CA1 region is excitatory, glutamatergic Schaffer collateral axons from CA3 pyramidal neurons. The axons of CA1 pyramidal neurons form the alveus and make 14

21 glutamatergic projections to the contralateral hippocampal formation, subiculum, entorhinal cortex, septal nuclei, mammillary region of the posterior hypothalamus, nucleus accumbens, amygadala and other basal forebrain structures (Amaral and Witter, 1995). In addition, there are other significant inputs to CA1 region from other brain areas, including cholinergic and GABAergic projections from the medial septal nucleus, serotonergic projections from the raphé nuclei and noradrenergic projections from the locus coeruleus (Amaral and Witter, 1995). GABA A Receptors GABA A receptors are the main locus of benzodiazepine actions. γ-aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the CNS and GABA A receptors are the main locus of benzodiazepine action. Benzodiazepines have been shown to bind to a high-affinity BZ recognition site on the GABA A receptor (Mohler and Okada, 1977) and BZ agonists increase the frequency of Cl - channel opening in response to GABA without altering the mean duration of channel open time (Polc, 1988, Macdonald and Olsen, 1994). Blockade of the benzodiazepine site with a competitive antagonist, such as flumazenil, prevents all of the therapeutic actions of benzodiazepine agonists (Hunkeler et al., 1981) and benzodiazepine receptor inverse agonists. Evidence also supports the existence of a protein complex associated with postsynaptic GABA responses as a major site of action for the benzodiazepines and binding studies demonstrate a close association between GABA and benzodiazepine actions (Costa et al., 1978; Tallman et al., 1978; Schofield 1989). Thus, the GABA A 15

22 receptor is of fundamental importance for acute and chronic mechanisms of benzodiazepine actions. GABA A Receptor Family The GABA A receptor is a member of a superfamily of ligand-gated ion channels that also includes the nicotinic acetylcholine receptor, the 5-hydroxytryptamine type 3 receptor and the glycine receptor. They are heterooligomeric intrinsic membrane proteins, which are assembled with 5 subunits derived from 8 main subunit families (α 1-6, β 1-4, γ 1-3, ρ 1-3, σ, π, ε and θ) that are encoded by more than 18 different genes (Macdonald and Olsen, 1994; Rudolph et al., 2001). Most GABA A receptors, however, are comprised of α, β and γ subunits (Sieghart and Sperk, 2002). Studies of recombinant expressed GABA A receptors indicate that the γ 2 subunit is critical for BZ sensitivity ( Pritchett et al., 1989) and is required for BZ binding together with an α subunit (Olsen and Tobin, 1990). The most plentiful subunit combination (43% in rat brain) is α 1 β 2 γ 2 containing GABA A receptors, which is found in most brain areas, including hippocampal and cortical interneurons and cerebellar Purkinje cells (McKernan and Whiting, 1996). While the γ subunit is required for GABA A receptor sensitivity to benzodiazepine modulation, the α subunit defines the specific clinical actions of the benzodiazepines: α1 subunitcontaining GABA A receptors mediate anticonvulsant and hypnotic actions, α2 subunitcontaining GABA A receptors mediate anxiolytic actions, and α5 subunit-containing GABA A receptors mediate amnestic actions (Crestani et al., 2000; Low et al., 2000; Mohler et al., 2002). 16

23 GABA A Receptor Involvement in Benzodiazepine Tolerance and Dependence A variety of changes in the postsynaptic GABA/benzodiazepine/Cl - complex have been demonstrated following chronic benzodiazepine administration. Down-regulation in the number of GABA A receptors is a potential mechanism for the development of tolerance although studies aimed at testing this hypothesis by radioligand binding assay have met with mixed success (Rosenberg and Chiu, 1981; Tietz et al., 1986; Miller et al., 1988; Roca et al., 1990; Huang et al., 1995). Four-week chronic FZP administration also generated regional decreases in the number of benzodiazepine binding sites in cerebral cortex, hippocampus and medulla-pons, as well as substantia nigra pars reticulata, the superior colliculus, the lateral amygdala and lateral hypothalamus (Rosenberg and Chiu, 1981; Tietz et al., 1986). However, the Gallager group reported that [ 3 H]zolpidem and [ 3 H]flunitrazepam binding were unchanged after chronic diazepam treatment (Gallager et al., 1984b). Other studies failed to demonstrate changes in GABA A receptor number following chronic benzodiazepine administration (Gallager et al., 1984a; Miller et al., 1988; Roca et al., 1990; Brett and Pratt, 1995). There are a few other reports which demonstrated that benzodiazepine receptor downregulation, represented by BZ agonist and antagonist binding studies, does not correlate with benzodiazepine anticonvulsant tolerance, suggesting that additional mechanisms may also contribute to different benzodiazepine actions (Gallager et al., 1985; Rosenberg et al., 1985; Rosenberg et al., 1989; Rosenberg et al., 1995). There is increasing experimental evidence, however, that a number of molecular processes are invoked by long-term exposure of animals or cultured cells to benzodiazepine-site ligands. These include uncoupling of the allosteric linkage between 17

24 the GABA and benzodiazepine sites, a decreased ability of GABA to potentiate benzodiazepine binding (GABA-shift) (Gallager et al., 1984; Farb et al., 1984; Tietz et al., 1989; Hu and Ticku, 1994; Primus et al., 1996; Holt et al., 1999); changes in GABA A receptor protein internalization, degradation and synthesis (Angelotti and Macdonald, 1993; Connolly et al., 1999; Barnes, 2000; Meyer et al., 2000); and regulation of receptor gene expression (Kang and Miller, 1991; Primus and Gallager, 1992; Tietz et al., 1994; Zhao et al., 1994; Holt et al., 1996). Overall, it is still difficult to draw a general conclusion as to the effects of chronic BZ treatment on GABA A receptor subunit expression based on the results of studies to date. Thus it may be that changes at the level of the GABA A receptor itself may only be one component of the neuroadaptive mechanisms underlying benzodiazepine tolerance and dependence. Ionotropic Glutamate Receptors In recent years, the glutamate receptor family which mediates excitatory responses in brain has attracted more and more attention in relation to studies of benzodiazepine tolerance and dependence as well as a variety of drugs of abuse. Koob and Bloom suggest that chronic drug administration leads to the initiation of adaptive process, which counter the acute effects of the drug, and that these processes persist after the drug has been cleared from the brain, thereby leaving the opposing forces unopposed (Koob and Bloom, 1988). Considering the close neuroanatomical interrelationship between GABAergic and glutamatergic neurons in many brain regions and their interactions in 18

25 relation to mediating neuronal adaptive processes and synaptic plasticity, plus the fact that GABAergic system alone can not explain all neurophysiological changes underlying benzodiazepine tolerance and dependence, the observation that glutamate receptors are regulated following chronic benzodiazepine administration attracted more and more attentions recently. Glutamate receptors are also demonstrated in mediating a lot of other CNS depressants and stimulants dependence and addiction, including ethanol, barbiturate, cocaine, amphetamine, morphine (Molleman and Little, 1995; Jackson et al., 2000; Vekovischeva et al, 2001; Tzschentke and Schmidt, 2003; Van Sickle et al., 2004; Nestler, 2005; Sanchis-Segura et al, 2006; Zhong et al, 2006). The involvements and underlying mechanisms of glutamate receptors in benzodiazepine dependence and other drug dependence may share a lot of similarities. Thus the following sections will review the glutamate receptor family and their involvements in benzodiazepine tolerance and dependence. Glutamate Receptor Family Members The amino acid L-glutamate is one of the most important neurotransmitters in the mammalian CNS. In general, glutamate binds to several classes of receptors, which can be divided into two distinct receptor families based on their molecular structures, transduction mechanisms, and their pharmacological profiles. This includes the ligandgated ion channel glutamate receptors. At present, there are at least three distinct subtypes, i.e., NMDA receptor, AMPA receptor, and kainate receptors, based on their selective agonists and the G-protein coupled metabotropic receptors (mglurs) (for reviews, see Bigge, 1999). 19

26 AMPA Receptors and Subunit Composition The AMPA subtype of ionotropic glutamate receptor is generally a tetrameric or pentameric structure with different combinations of four subunits (GluR1-4). The precise subunit composition and stoichiometry of AMPA receptors in vivo differs throughout the central nervous system (Sommer and Seeberg, 1992; Wenthold et al., 1996). AMPA receptor subunits can form different combinations of a variety of homomeric or heteromeric receptor channels with distinct pharmacological and functional properties (Wenthold et al., 1992). Miscellaneous gene encoding mechanisms, including alternative splicing and post-transcriptional modifications, further contribute to AMPA receptor functional diversity. Flip and flop isomers were generated by alternative splicing which produces a particular mrna sequence in the GluR1-4 subunits (Sommer et al., 1990). A example of RNA editing of the AMPA receptor subunits is Q/R editing that takes place in the GluR2 subunit (Sommer et al., 1991) and results in Ca 2+ impermeability of AMPA receptor that are expressed in majority of native AMPA receptors in vivo (Boulter et al., 1990). Immunoprecipitation studies in hippocampal CA1 region indicate that the majority of complexes are composed of GluR1/GluR2 and GluR2/GluR3 with only a minor contribution from GluR1/GluR3, while ~8% of AMPARs are GluR1 homomers (Wenthold et al., 1996). NMDA Receptor and Subunit Composition NMDA receptors are composed of assemblies of two subunit subtypes, including NR1 and NR2A-D, where incorporation of both types of subunit are required for a functional receptor channel (Monyer et al., 1994). The NMDA receptor is widespread 20

27 throughout the central nervous system while displaying distinct expression patterns with different receptor subunit combinations. The pattern appears to be consistent with heteromeric receptors composed of a common subunit (NR1) with various NR2 subunit combinations (Blahos and Wenthold, 1996; Chazot and Stephenson, 1997; Luo et al., 1997). The NR2 subunits have been indicated by recombinant NMDA receptor studies to play a major role in determining receptor properties such as sensitivity to Mg 2+ block, offset kinetics of glutamate-induced currents and affinity for different modulators and antagonists (Monyer et al., 1994; Vicini et al., 1998). Immunoprecipitation studies indicate that the majority of NMDARs are composed of dual complex of NR1/NR2A (~40%) or NR1/NR2B (~40%), with a minor portion (~6%) from NR1/NR2A/NR2B complexes (Blahos and Wenthold, 1996; Chazot and Stephenson, 1997). Functional co-regulation of AMPA and NMDA Receptors Both AMPAR and NMDAR are tightly co-regulated and a constant ratio of NMDAR to AMPAR current is preserved following activity-dependent synaptic scaling (Watt AJ et al, 2000; Watt AJ et al, 2004; Perez-Otano and Ehlers, 2005). AMPA receptors are normally impermeable to Ca 2+, a feature controlled by incorporating of GluR2 subunit (Boulter et al., 1990; Sommer et al., 1991). The AMPA receptor has a rapid kinetics of activation ( ms) and deactivation (τ-deactivation: ms), while NMDA receptor displays much slower activation (10-50 ms) and deactivation (τ-deactivation: ms) kinetics (Dingledine et al., 1999). In addition to glutamate, the NMDA receptor requires a co-agonist, glycine, to bind on the NR1 subunit to allow the receptor to function. The NR2B subunit also possesses a binding site for polyamines, regulatory 21

28 molecules that modulate the function of the NMDA receptor (Monyer et al., 1994). At resting membrane potentials, NMDA receptors are inactive. This is due to a voltagedependent block of the channel pore by Mg 2+, preventing ion flow. Sustained activation of post-synaptic AMPA receptors by releasing glutamate from the pre-synaptic terminal, depolarizes the postsynaptic cell, removing the Mg 2+ block of channel pore and thus allowing the NMDA receptor to be activated. Unlike GluR2-containing AMPA receptors, NMDA receptors are permeable to calcium ions as well as other cations. Thus NMDA receptor activation leads to a calcium influx into the postsynaptic cell, a signal thought to be crucial for the induction of NMDA-receptor dependent synaptic plasticity, including long-term potentiation (LTP) and long-term depression (LTD), and excitotoxic process associated with a number of pathological or neurodegenerative states including epilepsy and Alzheimer s disease (Bliss and Collingridge, 1993; Lipton and Rosenberg, 1994; Whetsell, 1996). Electrophysiological Characteristics of AMPA and NMDA Receptors in Hippocampus Activation of synaptic glutamate receptors via the Schaffer collateral pathway in hippocampal slices has been used as a classic method to study glutamate receptor activity and excitatory post synaptic currents (EPSCs) in CA1 pyramidal neurons by whole-cell voltage clamp electrophysiology (Hestrin et al., 1990). There are two distinguishable components of an EPSC: a fast, AMPA receptor-mediated component generated by the flow of Na + and K +, and a slow, NMDA receptor-mediated component generated by the flow of Na +, K + and Ca 2+ (Collingridge et al., 1988). The fast, AMPA receptor-mediated EPSC is characterized as voltage-independent, fast rise (<5ms) and decay (<20ms) time, 22

29 a linear current-voltage relationship with a reversal potential about 0 mv, which can be abolished by the AMPA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) (Hestrin et al., 1990). In contrast, the slow, NMDA receptor-mediated EPSC is depolarization-dependent, and has a relatively slow rise time (8-20ms) and decay time ( ms), a Mg 2+ -sensitive current-voltage relationship which can be abolished by the NMDA receptor antagonist DL-2-amino-5-phosphonovalerate (APV) (Hestrin et al., 1990). Glutamate activation of AMPA receptors and NMDA receptors in CA1 pyramidal neurons via the Schaffer collateral pathway is a useful tool for investigating excitatory synaptic current and post-synaptic ion channel functions. Glutamate Receptor Regulation Associated with Benzodiazepine Dependence Activity-dependent Synaptic Plasticity Models Activity-dependent changes in the efficacy of synaptic communication are currently an area under intense investigation. Glutamate receptors play important roles in synaptic communication and plasticity and are involved in a variety of physiological and pathophysiological processes in the brain. The classic examples include LTP, behavioral sensitization, kindling events and glutamate excitotoxiticy. Since these examples may share similarities with activity-dependent synaptic plasticity in benzodiazepine tolerance and dependence, the following section will briefly review these models. 23

30 AMPA Receptor Plasticity in the LTP model LTP is considered to be one of the main cellular mechanisms underlying learning and memory and occurs in a variety of different brain regions, but the phenomenon is most easily demonstrated in hippocampus, in which it was first discovered by Bliss and Lomo (1973). It refers to along-lasting (at least hours) increase in synaptic transmission that is induced by repetitive, high-frequency activation of excitatory synapses (Bliss and Lomo, 1973). Application of a strong, high-frequency stimulus causes a sustained release of glutamate, which will activate post-synaptic AMPA and NMDA receptors at hippocampal synapses. The role of AMPA receptor during the induction phase is primarily to provide sufficient post-synaptic depolarization, which will remove the Mg 2+ block of the NMDA receptor and allow the influx of Ca 2+ through the opening of NMDA channels ( for reviews, see Bliss and Collingridge, 1993; Malenka, 1994). The subsequent elevation of intracellular Ca 2+ levels is the critical signal for the signaling pathways, including a number of kinases and phosphotases at different stages, which eventually produces synaptic plasticity and neuronal changes. Recent literature also suggests that regulation of L-type voltage-gated calcium channels (VGCCs) is another important source of Ca 2+ during LTP (Morgan and Teyler, 1999; Borroni et al, 2000). Enhanced postsynaptic sensitivity in LTP expression may be related to phosphorylation-mediated stimulation of AMPA receptor function, including by calcium/calmodulin-dependent protein kinase II (CAMKII), PKA and PKC (McGlade-McCulloh et al., 1993; Tan et al., 1994; Barria et al., 1997; Derkach et al., 1999; Malenka and Nicoll, 1999; Banke et al., 2000; Boehm et al., 2006). Activity-dependent recruitment of AMPA receptors is now well-described (O Brien et al., 1998; Turrigiano, 2000; Liao et al., 2001) and is strongly 24

31 implicated in various forms of activity-driven synaptic plasticity, including LTP at CA1 excitatory synapses, both in vivo and in vitro (Liao et al., 1995; Lledo et al., 1998; Nayak et al., 1998; Zamanillo et al., 1999; Malenka, 2003; Malinow, 2003) AMPA Receptor Plasticity in Other Activity-Driven Synaptic Changes AMPA receptors have also been demonstrated to play a role in the development of other events in the brain, including behavioral sensitization and kindling. Behavioral sensitization refers to the progressive augmentations of drug effects following repeated exposure for a period of time (for reviews, see Robinson and Bridge, 1993; Wolf, 1998). AMPA receptors have been implicated to play important roles in the neuronal changes after repeated exposure and withdrawal from psychostimulants, such as cocaine and amphetamine, and morphine, which may underlie the neurophysiological mechanism of behavioral sensitization (Churchill et al., 1999; Fitzgerald et al., 1996; Lu and Wolf, 1999). Kindling is the process of triggering major epileptic seizures by repeated subthreshold electrical or chemical stimulation of specific brain regions, a model that has been extensively studied, most often in rats, to investigate the synaptic mechanisms in the development and expression of limbic epilepsy (Loscher, 1998). Just as both AMPA and NMDA receptors play important roles in synaptic plasticity in LTP and sensitization development, they are similarly involved in the development of kindling events (Rogawski, 2001), although their specific roles in synaptic transmission remain to be defined in these models. 25

32 AMPA Receptor Involvement in Drug Dependence Numerous studies support an important role for glutamatergic mechanisms in tolerance and dependence phenomena associated with a variety of drugs of abuse including benzodiazepines (Jackson et al., 2000; Van Sickle et al., 2004). Increased glutamatergic strength in mesolimbic reward pathways has also been implicated in addictive behaviors (Tzschentke and Schmidt, 2003; Nestler, 2005) and evidence supports a role for enhanced hippocampal glutamatergic plasticity underlying dependence on ethanol (Molleman and Little, 1995; Sanchis-Segura et al, 2006) and morphine (Vekovischeva et al, 2001; Zhong et al, 2006). Likewise, there is growing evidence of neuroadaptive changes in the glutamatergic system associated with benzodiazepine withdrawal (Stephens, 1995; Allison and Pratt, 2003; Van Sickle et al, 2004). The observations of Steppuhn and Turski (1993) demonstrated that administration of GYKI during the silent phase, days 1-3 of withdrawal in which diazepam-dependent mice were symptom free, would prevent the subsequent development of withdrawal signs in the active phase, days 4-21 of withdrawal during which mice displayed classic benzodiazepine withdrawal signs. Studies have demonstrated changes in expression of particular AMPA receptor subunits and regionally specific alternations in AMPA receptor subunit mrnas following withdrawal from an oral (gavage) dosing regime with diazepam associated with withdrawal anxiety in rats (Izzo et al, 2001; Allison et al., 2002). Thus, AMPAR-mediated neuronal hyperexcitability is an important part of a functional anatomic circuit (Millan 2003) which may be a necessary component of the neurophysiological mechanisms underlying benzodiazepine dependence. 26

33 NMDA Receptor Involvement in Drug Dependence Extensive studies have demonstrated a role for NMDA receptors in tolerance and dependence to drugs of abuse (for reviews see Trujillo and Akil, 1995). With respect to benzodiazepine administration, there is a wide body of evidence (reviewed by Stephens, 1995) to suggest involvement of NMDA receptor-mediated processes in the development of kindling with FG7142, an inverse agonist at benzodiazepine receptors (Gilbert, 1988; Sato et al., 1988; Stephens and Turski, 1993). It has been shown that co-administration of the competitive NMDA antagonist CPP, but not the competitive AMPA receptor antagonist GYKI 52466, to mice undergoing chronic diazepam treatment prevents the development of sedative tolerance (Steppuhn and Turski, 1993), indicating that NMDA receptors have a role in benzodiazepine tolerance development. Administration of CPP, during the active phase but not silent phase, as Steppuhn and Turski (1993) described above, prevented the withdrawal symptoms. Other work showed that MK801, a noncompetitive NMDA antagonist, prevents development of tolerance to the sedative effects (File and Fernandes, 1994), but not the anxiolytic effects (Fernandes and File, 1999), of diazepam in rats, while tolerance to the anticonvulsant affects of lorazepam in mice were prevented by the NMDA receptor antagonist CPP (Koff et al., 1997). Taken together, although there are a number of pieces of evidence to suggest the involvement of NMDA receptors in expression of benzodiazepine tolerance and dependence following chronic benzodiazepine administration, the precise mechanism relating how NMDA receptor is involved remained to be clarified. 27

34 Voltage-gated Calcium Channel Regulation Mediates drug Tolerance and Dependence The precise cellular and molecular mechanisms remain unclear in benzodiazepine tolerance and dependence, especially at synapses, despite extensive studiesof GABAR and glutamate receptors. Recent years, activity-dependent synaptic plasticity has been well described and, as described earlier, a Ca 2+ -mediated signaling pathway has been suggested to play a critical role in mediating synaptic plasticities. One important Ca 2+ source is voltage-gated calcium channels and it has long been indicated to be involved in drug dependence, since VGCC antagonists could block withdrawal symptoms of a variety of CNS drugs. Thus the following sections will review VGCCs and their involvement in Ca 2+ -mediated signaling in shaping synaptic transmissions in drug tolerance and dependence. Voltage-gated Calcium Channels Classification of Voltage-Gated Calcium Channels Voltage-gated calcium channels (VGCCs) belong to voltage-gated ion channel families, composed of combination of transmembrane protein subunits that open in response to membrane depolarization and sub-threshold depolarization signals, subsequently allowing Ca 2+ ions to enter the cell from the extracellular space. Thus, Ca 2+ entering the cell through voltage-gated Ca 2+ channels mainly serves as the second messenger of electric signaling, initiating intracellular events such as contraction, secretion, synaptic transmission, and gene expression (for review, see Catterall 2000; 28

35 Felix 2005) in response to neuronal activation. VGCCs have been sub-divided into multiple types according to their physiological and pharmacological properties. At least nine types of VGCCs have been identified in neurons for different physiological functions, such as synaptic plasticity, neuronal survival, synaptic transmission and activity at the neuromuscular junction (Catteral 1998; Catteral 2000). They can be further divided into two sub-families: high voltage-activated (HVA) channels and low voltageactivated (LVA) channels. Perhaps the best-characterized HVA Ca 2+ channel is the L- type. It is characterized by a high voltage of activation, slow voltage-dependent inactivation, large single-channel conductance, and specific inhibition by Ca 2+ antagonist drugs including dihydropyridines, phenylalkylamines, and benzothiazepines (Catterall 2000; Lipscombe 2004). The major LVA Ca 2+ channel type is the T-type. In contrast to L-type currents, these T-type dependent Ca 2+ currents activate at much more negative membrane potentials, inactivate rapidly, deactivate slowly, have small single-channel conductance, and are insensitive to Ca 2+ antagonist drugs. They are defined as lowvoltage-activated Ca 2+ currents because of their more negative voltage dependence (Catterall 2000). Other types of VGCCs, including N-, P/Q- and R- types are characterized by their intermediate voltage dependence and inactivation rate: more negative and faster than L-type, but more positive and slower than T-type (Catterall 2000). They are prominent in neurons and can be distinguished pharmacologically. Most authors classify them as part of HVA family, although some others refer to the R- type as an intermediate type compared to HVA and LVA Ca 2+ channels (Birnbaumer 1994). 29

36 Expression of L-type VGCCs in Hippocampus Some literature suggests that the density of neuronal L-type Ca 2+ channels in the brain is significantly higher than that of vascular ones (Ricci et al., 2002). This might explain the more pronounced neuronal than vascular effects after pharmacological manipulation of cerebral Ca 2+ channels. A variety of VGCCs are heterogeneously expressed in different classes of neurons in the hippocampus. The relative abundance of each channel type is somewhat different between CA1 and CA3 neurons and very different between pyramidal neurons and granule cells (Johnson et al., 1992). While some L-type VGCCs are expressed in granule cells, they are expressed much more abundantly in both CA1 and CA3 neurons (Johnson et al., 1992). Using monoclonal antibodies from purified skeletal muscle Ca 2+ channels Ahlijanian and Westenbroek identified crossing-reacting L-type Ca 2+ channels in neurons, primarily localized in cell bodies and proximal dendrites of hippocampal pyramidal neurons, while in dentate gyrus, they are expressed more abundantly in the unmyelinated neuronal processes (axonal and dendritic) than the granule neuron cell bodies (Ahlijanian 1990; Westenbroek 1990). Consistent with this location, L-type VGCCs demonstrated a dominant role in Ca 2+ influx into the cell bodies in organotypic hippocampal slices (Elliott et al., 1995). These L-type VGCCs were not detected in high density in nerve terminals, arguing against their roles in neurotransmitter release (Catterall 1998). L-type VGCC Subunit Composition L-type VGCCs are multimeric complexes of a pore-forming α1 subunit of ~ kda; a transmembrane, disulfide-linked complex of α2 and δ subunits; an intracellular β 30

37 subunit; and in some cases a transmembrane γ subunit (Catterall 2000). Ten α1 subunits, four α2δ complexes, four β subunits and three γ subunits have been identified (Catterall 2000; Felix 2005). The α1 subunit contains the Ca 2+ pore and binding sites for selective channel antagonists. It is composed of four homologous domains (I-IV) each comprising six transmembrane segments (S1-S6) (Felix 2005). The β subunit is an entirely cytoplasmic protein and appears to function in a wide range of regulatory functions (Hanlon and Wallace, 2002). The α 2 δ subunit is the product of cleavage of a single gene product and consists of an extracellular α2 domain linked to the membrane by a disulfide bond transmembrane σ peptide. The α 2 δ subunit also appears to be responsible for important modulatory effects on the pore activity (Felix 1999; Klugbauer et al., 2003). The γ subunit, characterized by four transmembrane domains, was initially only detected in skeletal muscle Ca 2+ channels and has been recently associated with neuronal Ca 2+ channels (Kang and Campbell, 2003). One member of this γ subunit family, γ2, also known as stargazin, acts not only as a modulatory γ subunit for VGCCs, but also as a regulator of postsynaptic membrane targeting for AMPA-type glutamate receptors (Chen et al., 2000; Vandenberghe et al., 2005). Classification of VGCC Subunit Gene Expression Molecular and gene sequence studies have revealed a comprehensive sequence-based classification of VGCCs. Using functional and pharmacological criteria, three major groups of genes are defined that corresponds to the subtypes of VGCCs: Ca V 1 (L-type); Ca V 2 (P-, N-, and R-type), and Ca V 3 (T-type) (Felix 2005). Ca V 1 genes encode L-type VGCCs. Four Ca V 1 genes have been identified (Ca V ). The Ca V 1.1 gene, formerly 31

38 α 1S, is expressed in skeletal muscle (Catterall 2000). The Ca V 1.4 gene, formerly α 1F, is primarily in retina (Bech-Hansen et al., 1998). Only the neuronal Ca V 1 genes, Ca V 1.2 and Ca V 1.3, will be discussed here since they are the main focus of these dissertation studies. Characteristics of Ca V 1.2 α1 (α 1C )-containing L-VGCCs Ca V 1.2 α1 (α 1C ) and Ca V 1.3 α1 (α 1D ) are the two most widely expressed L-type channel subunits and together they are thought to underlie L-type currents in neurons (Hell et al., 1993). The Ca V 1.2 gene, formerly α1 C, is expressed in variety of cells including neurons (for reviews, see Catterall 2000). Functional studies of recombinant Ca V 1.2 α1 subunits have confirmed that they form high voltage-activated, long-lasting, slow inactivation, dihydropyridine-sensitive L-type currents, classical L-type currents as described before (Mori et al., 1993). In neurons Ca V 1.2 α1 channels are thought to couple membrane depolarization to regulation of gene expression and a variety Ca 2+ -mediated phosphorylation/dephosphorylation pathways (Dolmetsch et al., 2001; Weick et al., 2003). Interestingly, some studies reported that clusters of Ca V 1.2 subunit are located at postsynaptic membranes of asymmetric, glutamatergic synapses in hippocampal neurons. Activation of the NMDA receptors in vitro caused proteolytic processing of the COOH terminal domain of Ca V 1.2 subunit (Hell et al., 1996), which has been shown in Xenopus oocytes to increase Ca 2+ channel conductance (Wei et al., 1994). These results suggest local Ca 2+ signal could enhance L-type VGCC activity in postsynaptic membranes after activation of NMDA receptors during synaptic transmissions, and potential cross-talk between these two Ca 2+ -permeable ion channels (Catterall 1998). 32

39 Characteristics of Ca V 1.3 α1 (α 1D )-containing L-VGCC The Ca V 1.3 α1 gene, formerly α 1D, is expressed in many of the same cells that also express Ca V 1.2. In neurons, Ca V 1.2 and Ca V 1.3 are often found in the same general neuronal compartments, particularly cell bodies and proximal dendrites, although immunocytochemical analysis revealed a distinct difference in subcellular localization (Hell et al., 1993; Westenbroek et al., 1998). These studies are inconsistent, however, since some reported Ca V 1.2 extended far out the dendrites, including CA3 pyramidal neurons (Hell et al., 1993), whereas another study indicated that Ca V 1.3 channels extended into the distal dendrites of GABAergic neurons in mouse cerebral cortex (Timmermann et al., 2002). Despite their distribution characteristics, Ca V 1.2 α1 and Ca V 1.3 α1 L-type channels display distinct functional differences: Ca V 1.3 L-type channels activate at relatively hyperpolarized membrane potentials (about -55 mv), a voltage threshold that is approximately mv more hyperpolarized as compared with Ca V 1.2 L-type channels (Avery and Johnston, 1996; Xu and Lipscombe, 2001), and Ca V 1.3 L-type channels are only partially inhibited by dihydropyridines (Koschak et al., 2001; Xu and Lipscombe, 2001). These characteristics can have important physiological significance as Ca V 1.3 L-type channels will be activated in response to physiological stimuli that do not open Ca V 1.2 L-type channels. This broadens the functional importance of L-type calcium channels to include neuronal processes triggered by fast, sub-threshold depolarizations which may lead to distinct signaling pathways. Although still under investigation, Ca V 1.3 L-type calcium channels may contribute a significant fraction of Ca 2+ currents that were originally attributed to R-type current in many neurons (for 33

40 reviews, see Lipscombe et al., 2004), which were proposed to contribute to synaptic plasticity in dendritic spines of hippocampal pyramidal neurons (Yasuda et al., 2003). L-type VGCC Regulation Associated with Drug Tolerance and Dependence L-type VGCC Antagonists Prevent a Variety of Drug Dependence Syndromes A variety of studies have demonstrated the ability of L-type calcium channel blockers like nimodipine, nifedipine and verapamil to block a variety of benzodiazepine withdrawal signs including hyperkinesia, hyperthermia, hyperaggression and audiogenic seizures (Chugh et al, 1992; Gupta et al, 1996; Ganouni et al, 2004). Nitrendipine dosedependently decreased seizures precipitated by the benzodiazepine receptor partial agonist FG7142 (Dolin et al, 1990). The ability of L-type VGCC antagonist to prevent withdrawal signs also exists with other drugs of abuse. Multiple positive allosteric modulators of GABA A receptor, including ethanol and barbiturates, produce hyperexcitability and physical dependence after withdrawal from long-term administration (Whittington and Little, 1993; Rabbani and Little, 1999; Watson and Little, 2002). Nitrendipine, an L-type VGCC antagonist, can prevent the hyperexcitability produced by withdrawal from long-term ethanol administration, both in vivo (Watson and Little, 2002) and in isolated neuronal preparations (Whitting and Little, 1991; Whitting and Little, 1993). Alterations of both dihydropyridine-sensitive and insensitive neuronal calcium channels are also reported after chronic barbiturate treatment (Rabbani and Little, 1999). In morphine dependent rats, VGCCs are thought to be involved in intracellular calciummediated protein phosphorylation (Pourmotabbed et al., 1998). Thus, VGCCs contribute 34

41 to expression of dependence following withdrawal from multiple positive GABA A receptor allosteric modulators as well as other drugs of abuse. L-type VGGG-mediated Neuronal Plasticity in Drug Dependence Antagonist studies suggest a general role for VGCC-mediated Ca 2+ influx in CNS drug dependence. This is not surprising since CNS drug dependence is often associated with activity-driven neuronal plasticity (for reviews, see Nestler 2001; Kelly 2004; Nestler 2005). L-type VGCCs have been reported to play a role in various activitydependent models of neuronal plasticity (Morgan and Teyler, 1999; Borroni et al., 2000; Rajadhyksha and Kosofsky, 2005), a similar role played by NMDA receptor activation but with likely different signaling pathways. Activation of neuronal L-type VGCCs triggers a sustained influx of Ca 2+ upon depolarization which, via diverse soluble messengers and transcription factors, initiates long-term processes related to synaptic plasticity in the hippocampus and amygdala (Dolmetsch et al., 2001). Nicoll s group has shown that a rise in postsynaptic Ca 2+ concentration and calcium/calmodulin-dependent protein kinase II activity, through a voltage-gated calcium channel pathway, potentiated mepsc amplitude without channel kinetics changes (Wyllie et al., 1994; Wyllie and Nicoll, 1994). In fact, during ethanol withdrawal and following psychostimulant administration, activation of L-type VGCCs can result in the influx of intracellular calcium and downstream activation of Ca 2+ /CaM-activated kinase and phosphatase pathways, eventually involving CREB-induced gene expression, also important to neuronal and experience-dependent plasticity (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005). Thus, the accumulated evidence 35

42 supports the possibility that calcium influx through L-type VGCCs, may activate diverse intracellular messengers or transcription factors to increase glutamatergic strength during benzodiazepine withdrawal and contribute to benzodiazepine dependence. L-type VGCCs Mediate Downregulation of GABA A Receptors after Prolonged GABA Exposure, but not Uncoupling. Evidence of participation of L-type VGCC in mediating GABAergic system changes following chronic benzodiazepine administration is less abundant. As described in the previous section, multiple mechanisms involving GABAergic system regulation were suggested to explain benzodiazepine tolerance and dependence. GABA-induced GABA receptor downregulation has been suggested to be the product of a transcriptional repression of GABAR subunit genes (Lyons et al., 2000; Russek et al., 2000) that depends on activation of L-type VGCC (Lyons et al., 2001; Gravielle et al., 2005). Benzodiazepine chronic treatment also induces subunit-specific changes in the levels of region-specific GABA A receptor subunit mrnas and proteins as well as tolerance (Chen et al., 1999; Tietz et al., 1999a; Tietz et al., 1999b). However, decreases in the number of benzodiazepine receptors (down-regulation) were demonstrated only in some, but not all, of chronic benzodiazepine administration models (Rosenberg and Chiu, 1981; Tietz et al., 1986; Miller et al., 1988; Roca et al., 1990; Huang et al., 1996). The role of regulation of L-type VGCCs in mediating benzodiazepine tolerance remains to be identified. 36

43 Uncoupling, a reduction in the allosteric interactions between GABA and benzodiazepine binding sites, however, was observed to be independent of L-type VGCC activation (Lyons et al., 2001; Gravielle et al., 2005). Changes in GABA/benzodiazepine coupling are dependent on the selectivity and efficacy of the benzodiazepine subtypes and need not involve changes in the numbers of GABA or benzodiazepine binding sites (Primus et al, 1996). It has been suggested that changes in GABA A receptor subunit composition or post-translational modifications contribute to uncoupling (Roca et al., 1990; Lyons et al., 2001; Gravielle et al., 2005). Thus L-type VGCC may regulate GABAR channel number changes, likely through gene regulation, but not appears be involved in uncoupling, suggesting multiple mechanisms are underlying GABAR changes following chronic benzodiazepine administration. Ca 2+ -mediated phosphorylation/dephosphorylation contributes to regulation of GABA A receptor conductance Another potential mechanism associated with benzodiazepine tolerance in 1-week chronic flurazepam treated rats was a reduction in GABA A receptor unitary conductance, but not GABA A receptor number changes, which could the underlying mechanism in reduction of GABA A receptor-mediated mipsc amplitude, a measure of postsynaptic GABA A receptor-mediated function (Zeng and Tietz, 1999). Homeostasis of intracellular Ca 2+ is important to regulate GABA A receptor channel kinetics and Ca 2+ was demonstrated to have biphasic effects on synaptic GABA A receptor channels (Koninck and Mody, 1996; Aguayo et al., 1998). This process could be through Ca 2+ -mediated phosphorylation/dephosphorylation processes since both Ca 2+ /calmodulin-dependent 37

44 protein kinase and calcineurin have been demonstrated to directly regulate GABARmediated Cl - current (Wang et al., 1995; Churn and DeLorenzo et al., 1998; Lu et al., 2000; Alix et al., 2002; Wang et al., 2003). Thus it remains of significant interest to evaluate the role of L-type VGCC in mediating the effects of chronic benzodiazepine treatment on postsynaptic GABA A receptor conductance. GABAR-dependent Depolarization and Activation of L-type VGCCs The excitatory role of GABA A receptor related to elevated intracellular Cl - concentrations during the early development period and its dependence on intracellular Ca 2+ concentration through L-type VGCC has been well described (Ganguly et al., 2001; also see reviews Stein and Nicoll, 2003; Marty and Llano 2005). That is, electrochemical gradient of Cl -, which depends on the intra- and extracellular Cl - concentration, determines the excitatory versus inhibitory nature of GABA A receptor. However, newer studies have shown that a GABA A receptor-mediated depolarization can occur in mature neurons. In mature hippocampal pyramidal CA1 neurons, while single stimulation of GABAergic inputs is inhibitory, high-frequency trains can induce GABA to become depolarizing due to Cl - accumulation in CA1 neurons (Voipio and Kaila, 2000, Isomura et al., 2003). A few other studies have also shown an increased concentration of intracellular calcium through L-type VGCCs following prolonged GABA A receptor activation that leads to bicarbonate-driven Cl - entry and Cl - accumulation and eventually promotes cell depolarization (Reichling et al, 1994; Lyons et al., 2001; Chavas et al, 2004; Marty and Llano, 2005). A similar situation likely occurs with prolonged GABA A receptor activation during chronic exposure to benzodiazepines. As a consequence of 38

45 FZP administration numerous time-dependent changes occur at the GABA A receptor some of which influence CA1 neuron hyperexcitability (Van Sickle et al., 2004) including a bicarbonate-driven Cl - accumulation reflected in a shift in the Cl - reversal potential and the appearance of a bicuculline-sensitive depolarizing potential (Zeng et al, 1995; Zeng and Tietz, 1997; Zeng and Tietz, 2000). Moreover, spectral analysis of hippocampal electrical activity during withdrawal from 1-week FZP reveals increased power of a 7Hz (theta) peak (Poisbeau et al., 1997) that was suggested to be associated with GABAergic post-synaptic depolarization and a shift of reversal potential from Cl - toward HCO - 3 (Sun et al., 2001). These potential depolarizing driving forces following chronic FZP administration could promote the activation of L-type VGCCs and downstream Ca 2+- mediated signal transductions. Direct inhibition of L-type VGCC Current by Benzodiazepines. A few studies have suggested a direct inhibition of L-type VGCC current by benzodiazepines. Early radioligand binding studies demonstrated that benzodiazepines in micromolar concentrations significantly inhibit depolarization-sensitive Ca 2+ uptake in neurons and intact nerve-terminal preparations (Taft and DeLorenzo, 1984; Johansen et al., 1985). Electrophysiological studies in neuroblastoma cells indicated that chlordiazepoxide in micromolar concentrations reversibly blocked both L-type and T- type VGCCs in both closed and open configurations (Reuveny et al, 1993). A variety of benzodiazepines, including diazepam, midazolam, clonazepam and nitrazepam produced different degree of inhibitions on both L-type and T-type VGCCs in cultured neurons and neuroblastoma cells (Watabe et al., 1993; Ishizawa et al, 1997). These direct interactions 39

46 between benzodiazepines and VGCCs also suggest the potential for the direct functional modulation of L-type VGCC by chronic administration of benzodiazepines. 40

47 Manuscript I [Behavior Pharmacology, in press] Benzodiazepine-induced hippocampal CA1 neuron AMPA receptor plasticity linked to severity of withdrawal-anxiety: Differential role of voltage-gated calcium channels and NMDA receptors. Kun Xiang and Elizabeth I. Tietz*, Department of Physiology and Pharmacology and the Cellular and Molecular Neurobiology Program, University of Toledo College of Medicine, Health Science Campus, Toledo, OH Short Title: AMPA receptor plasticity and withdrawal-anxiety *Corresponding Author: Elizabeth I. Tietz, Ph.D. Department of Physiology and Pharmacology University of Toledo College of Medicine Health Science Campus (Formerly Medical University of Ohio) 3000 Arlington Ave., Mailstop 1008 Toledo, OH liz.tietz@utoledo.edu 41

48 ABSTRACT Withdrawal from one-week oral administration of the benzodiazepine, flurazepam (FZP) is associated with increased AMPA receptor (AMPAR) miniature excitatory postsynaptic currents (mepscs) but reduction of NMDA receptor (NMDAR)-evoked (e)epscs in hippocampal CA1 neurons. A positive correlation was observed between increased AMPAR-mediated mepsc amplitude and anxiety-like behavior in 1-day FZP-withdrawn rats. These effects were disrupted by systemic AMPAR antagonist administration (GYKI-52466, 0.5 mg/kg, i.p.) at withdrawal onset strengthening the hypothesis that CA1 neuron AMPAR-mediated hyperexcitability is a central component of a functional anatomic circuit associated with expression of withdrawal-anxiety. Abolition of AMPAR current upregulation in 2-day FZP withdrawn rats by GYKI injection also reversed the reduction in NMDAR-mediated eepsc amplitude in CA1 neurons from the same rats suggesting that downregulation of NMDAR function may serve a protective, negativefeedback role to prevent AMPAR-mediated neuronal over-excitation. NMDAR antagonist administration (MK-801, 0.25 mg/kg. i.p.) had no effect to modify increased glutamatergic strength or withdrawal-anxiety, while injection of an L-type VGCC antagonist, nimodipine (10 mg/kg, i.p.) averted AMPAR current enhancement and anxiety-like behavior suggesting that these manifestations may be initiated by a VGCCdependent signal transduction pathway. An evidence-based model of likely cellular mechanisms in the hippocampus contributing to benzodiazepine withdrawal-anxiety was proposed implicating regulation of multiple CA1 neuron ion channels. Key Words: drug abuse, dependence, glutamate, L-type VGCC 42

49 INTRODUCTION Benzodiazepines (BZs), a group of positive allosteric modulators of γ- aminobutyric acid type A receptors (GABARs), are widely prescribed for the treatment of anxiety and insomnia and are frequently used by polydrug abusers. After withdrawal from long-term use, symptoms such as anxiety, agitation, sleep disturbances and in some cases delirium or seizures may appear which can lead to physical dependence and benzodiazepine misuse or further abuse (Griffiths and Johnson, 2005). An extensive literature suggests that different neural mechanisms mediated by distinct neural pathways might underlie the different components of the BZ withdrawal syndrome (Allison et al, 1999; Podhorna 2002; Allison and Pratt, 2003; Millan 2003; Van Sickle et al, 2004; Ganouni et al, 2004). The hippocampus is as an important component in anxiety expression (Millan 2003) and a variety of CA1 pyramidal neuron membrane receptors are suggested to be involved in anxiogenic behavior observed in the elevated plus-maze test in rats (Izzo et al, 2001; Degroot et al, 2001; Lamprea et al, 2003; Van Sickle et al, 2004). Excitatory, glutamatergic neuronal circuits have been implicated in mediating withdrawal symptoms associated with a variety of drugs of abuse (Stephens 1995; Molleman and Little, 1995; Little 1999; Jackson et al, 2000; Jang et al, 2000; Vekovischeva et al, 2001; Rajadhyksha and Kosofsky, 2005; Zhong et al, 2006). The findings that pharmacologic antagonism of α-amino-3-hydroxy-5-methylisoxasole- 4-propionic acid receptors (AMPARs) and N-methyl-D-aspartate receptors (NMDARs) differentially modify behavioral manifestations of benzodiazepine withdrawal (Steppuhn and Turski, 1993; Koff et al, 1997; Dunworth and Stephens, 1998; Van Sickle et al, 43

50 2004) and that AMPAR and NMDAR binding and subunit expression in rat hippocampus and frontal cortex are regulated following withdrawal from benzodiazepines (Allison et al, 1999; Izzo et al, 2001; Van Sickle and Tietz, 2002; Van Sickle et al, 2002) suggest that glutamatergic receptors may also play important roles in mediating BZ withdrawal signs. Activity-dependent increases in AMPAR synaptic function are primarily the consequence of an intracellular calcium-dependent biochemical cascade and involve a variety of possible structural and or functional changes in synaptic AMPA receptors (Malenka 2003; Boehm and Malinow 2005). NMDAR-dependent calcium signaling has been shown to be a primary mediator of enhanced AMPAR function associated with both Hebbian and homeostatic plasticity phenomena (Malenka 2003; Perez-Otano and Ehlers, 2005). However, another important calcium source, L-type voltage-gated calcium channels (VGCCs) has also been implicated in activity-dependent synaptic plasticity (Morgan and Teyler, 1999; Rajadhyksha and Kosofsky, 2005). Moreover, L-type VGCCs have been shown to be involved in mediating withdrawal signs to a variety of CNS depressants (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999; Podhorna 2002; Ganouni et al, 2004). Whether NMDAR- or VGCC-mediated calcium signaling contributes to regulation of AMPAR synaptic plasticity during FZP withdrawal is unknown. Previous studies in our laboratory were focused on the functional regulation of glutamate receptors on hippocampal CA1 neurons after prolonged BZ administration. A transient enhancement of AMPAR-mediated miniature excitatory postsynaptic currents (mepscs) and a reduction in NMDAR-mediated evoked (e)epscs were observed (Van 44

51 Sickle 2004) in 2-day FZP-withdrawn rats. In the present investigation we first evaluated the relationship between increased CA1 neuron AMPAR function and anxiety-like behavior in rats by pharmacological antagonism of glutamate receptor activity at the onset of withdrawal. Next, systemic AMPA antagonist injections in other groups of rats was used to assess the role of AMPA plasticity in mediating subsequent effects on CA1 neuron NMDAR function. Further, by means of pharmacological antagonism of either NMDAR or VGCC activity the possible Ca 2+ -mediated mechanism underlying enhanced AMPAR activity during benzodiazepine withdrawal was explored. Finally, to serve as the foundation for more mechanistic studies, a model of the cellular mechanisms underlying BZ withdrawal anxiety was proposed based on the present findings, on past evidence from our laboratory and others. METHODS Experimental protocols involving the use of vertebrate animals were approved by the University of Toledo College of Medicine (formerly the Medical University of Ohio), Institutional Animal Care and Use Committee (IACUC) and conformed to National Institutes of Health ethical guidelines. Drug Treatments Chronic Flurazepam Administration. FZP treatment in rats was as previously described (Van Sickle 2004). In short, following a 2-4 day adaptation period when rats were offered only a 0.02% saccharin vehicle, male Sprague-Dawley rats (initial age PN22-25, Harlan, Indianapolis, IN) were offered FZP (flurazepam dihydrochloride, ph 5.8) for 1 week in 45

52 saccharin solution as their only source of drinking water. The concentration of FZP was adjusted daily according to each rat s body weight and fluid consumption (100 mg/kg X 3 days and 150 mg/kg X 4 days) appropriate to flurazepam s relative potency, oral bioavailability and biotransformation resulting in benzodiazepine brain levels equivalent to other common benzodiazepine chronic treatments (Gallager et al, 1985; Lau et al, 1987; Xie and Tietz, 1992). Only rats that consumed a criterion dose of an average >120 mg/kg/day were accepted for study. Saccharin water was again provided during the 2-day withdrawal period. Unlike in humans, residual FZP and metabolites rapidly decline over the first 24 hours after drug removal and are no longer detectable in hippocampus in 2- day FZP-withdrawn rats (Xie and Tietz, 1992). Pair-handled control rats receive saccharin water for the same length of time. Rats were euthanized for hippocampal slice preparation on PN Systemic Antagonist Injection. Three groups of control and FZP-treated rats were given a single intraperitoneal injection of one of three different ion channel antagonists, optimal doses determined in previous studies (Van Sickle et al, 2004). First, the selective, competitive AMPAR antagonist, GYKI (0.5 mg/kg, i.p.) or 1% Tween-20 vehicle (0.5 ml/kg) was injected immediately at the end of 1 week FZP treatment, 24 hours (1- day withdrawn) prior to behavioral testing and 24 or 48 hours (1- and 2-day withdrawn groups) prior to tissue preparation for electrophysiological recording. Among the range of doses tested, this dose was previously shown to reverse the increased AMPAR mepsc amplitude without anxiogenic effects in rats (Van Sickle et al, 2004). In the second group, the non-competitive NMDAR antagonist, MK-801 (0.25 mg/kg, i.p.) or saline vehicle (1ml/kg) were given at the end of 1 week FZP treatment and 24 hours prior to 46

53 both behavioral testing and hippocampal slice preparation. This dose was previously shown to reverse downregulation of NMDAR function in 2-day FZP withdrawn rats 1 day following systemic injection without adverse locomotor effects (Van Sickle et al, 2004). The third group was preinjected with the VGCC antagonist, nimodipine (10 mg/kg, i.p.) or the vehicle 0.5 % Tween 80 (2ml/kg) immediately after ending 1-week FZP treatment and 24 hours prior to behavioral testing and slice preparation. This dose of nimodipine has minimal effect on locomotion, does not produce ataxia, has no effect on seizure threshold, yet reverses behavioral signs of ethanol dependence (Watson and Little, 2002). Behavioral Testing Elevated Plus-maze test. The elevated plus-maze, which represents unconditioned responses to a potentially dangerous environment, is a widely used measure of anxiety and the anxiolytic effects of drugs (Rodgers and Dalvi, 1997). The plus-maze consisted of two opposite open arms, 50 X 10 cm, crossed with two closed arms of the same dimensions, with walls 40 cm high. The arms were connected with a central square, 10 X 10 cm. The plus-maze was elevated 50 cm above the floor in a dimly illuminated, 144 sq. ft. room. As described previously (Van Sickle et al, 2004), all rats were naϊve to the plusmaze, and were tested only once between 10:00 and 10:30 a.m. prior to being sacrificed for hippocampal slice preparation either the same or the following day as described below. The test was initiated by placing a rat in the central square facing an open arm. The number of arm entries and the time spent in open and closed arms were record for 5 min by an observer in the same room. An arm entry was scored when all four limbs were 47

54 on the arm. The maze was cleaned with 100% ethanol, then distilled water before and after each rat was tested. Electrophysiology Hippocampal slice preparation. Hippocampal slices (400 μm) were prepared from rats as previously described (Van Sickle et al, 2004). Briefly, transverse dorsal hippocampal slices were cut on a vibratome (Ted Pella, Inc.) in ice-cold, pre-gassed (95%O 2 /5% CO 2 ) artificial cerebrospinal fluid (ACSF) containing (in mm): NaCl, 120; KCl, 2.5; CaCl 2, 0.5; MgSO 4 7.0; NaH 2 PO 4 1.2; NaHCO 3, 2; D-glucose, 20; Ascorbate, 1.3, ph 7.4. Slices were maintained at room temperature (RT) for 15 min in gassed, low-calcium, highmagnesium ACSF, then transferred to normal ACSF containing (in mm): NaCl, 119; KCl, 2.5; CaCl 2, 1.8; MgSO 4 1.3; NaH 2 PO ; NaHCO 3, 26; D-glucose, 10; ph 7.4. Slices were maintained at room temperature for 1 hr in ACSF. During recording, slices were superfused at a rate of 2.5 ml/min with gassed ACSF at room temperature. AMPAR-mediated mepsc recording. AMPAR-mediated mepscs were isolated from CA1 neurons in ACSF plus 1 mm TTX, 50 mm picrotoxin and 25 mm CGP using whole-cell voltage-clamp techniques as previously described (Van Sickle and Tietz, 2002). Patch pipettes (3-6 MΩ) were filled with internal solution containing (in mm): Csmethanesulfonate, 132.5; CsCl, 17.5; HEPES, 10; EGTA, 0.2; NaCl, 8; Mg-ATP, 2; Na3- GTP, 0.3; QX-314, 2; ph 7.2 adjusted with CsOH. Resting membrane potential (RMP) was measured immediately upon cell break-in. Neurons were voltage-clamped (V H = -80 mv) in continuous mode (csevc) using an Axoclamp 2A amplifier (Axon Instr., Union City, CA). Current output was low-pass filtered (10 khz), DC-offset, amplified 10,000-48

55 fold and continuously monitored on-line (PClamp 8.0, Axon). The digitized signal (Digidata 1200A, Axon) was stored on disk for later off-line analysis. Cells in which the holding current changed by more than 20% or the seal degraded, were abandoned. mepsc activity was recorded 5 min and analyzed with MiniAnalysis software (Synaptosoft Inc., Leonia, NJ). Peak mepsc amplitude was measured from baseline. Decay kinetics and mepsc amplitude were estimated using a single exponential function: [y(t)=a*exp(-t/τ)]. Whole-cell data were compared by repeated measures ANOVA with post-hoc analysis by the method of Scheffé (p<0.05). AMPA-mediated mepscs were abolished in the presence of 10 µm DNQX (Van Sickle and Tietz, 2002). Stimulus-evoked, NMDAR-mediated EPSC recording. NMDAR-mediated, stimulusevoked (e)epscs were recorded from CA1 neurons in in vitro hippocampal slices by stimulation of the Schaffer collateral pathway in the presence of 10 mm DNQX, 10 mm glycine, 25 mm CGP and 50 mm picrotoxin using whole-cell voltage-clamp techniques as described previously (Van Sickle et al, 2002; Van Sickle et al, 2004). Patch pipettes (5-9 MΩ) were filled with internal solution containing (in mm): Csmethanesulfonate, 132.5; CsCl, 17.5; HEPES, 10; EGTA, 0.2; NaCl, 8; Mg-ATP, 2; Na 3 - GTP, 0.3; QX-314, 2; ph 7.2 adjusted with CsOH. After establishing input-output relationships, NMDAR-mediated EPSCs were elicited with a tungsten, bipolar stimulating electrode at a stimulus intensity half-maximal for the EPSC. Neurons were voltage-clamped from -80 to +40 mv in continuous mode (csevc) using an Axoclamp 2A amplifier (Axon Instruments Inc., Union City, CA) monitored and digitized as described above. Current-voltage (I-V) curves were compared by repeated measures ANOVA followed by post hoc analysis by the method of Scheffé (p<0.05). In 49

56 preliminary experiments NMDA-mediated currents were abolished in the presence of 50 µm APV (Van Sickle et al, 2002). Nimodipine concentration-response effects on mepscs To determine whether nimodipine had direct effects on AMPAR-mediated mepscs, nimodipine ( µm) was superfused onto hippocampal slices during mepsc recording. After recording baseline mepsc activity for 5 min, nimodipine or vehicle (0.0001% to 0.1% DMSO in water) was added to the superfusate in increasing concentrations for 10 min each. mepsc amplitude and kinetics were analyzed using the final 5 min recording period at each concentration. Drug solutions Drugs used for superfusion during whole-cell recording were dissolved at 100 times their final concentration and added to the superfusate with a syringe pump (Razel, World Precision Instruments, Inc., Sarasota, FL) at a rate of 25 to 75 μl/min to achieve their final concentrations. GYKI for systemic injection was suspended in 1% Tween-20 solution. MK-801 was dissolved in the saline vehicle. For in vivo injection, nimodipine was dissolved in 0.5 % Tween-80 solution and kept in a light-tight vial. For in vitro perfusion, nimodipine was dissolved in DMSO to make a 10 mm stock solution diluted to the final concentration as needed (from 0.1 to 100 µm). All other drugs were dissolved in dh 2 O. DNQX (6,7-dinitroquinoxaline-2,3-done), QX-314 (lidocaine N-ethyl bromide quaternary salt), MK-801, picrotoxin, FZP dihydrochloride, nimodipine and GYKI are all from Sigma-Aldrich Chemical Co. (St Louis, MO). Tetrodotoxin (TTX) was 50

57 obtained from Alamone Laboratories (Jerusalem, Israel). CGP was purchased from Tocris Bioscience (Ellisville, MO). RESULTS Positive correlation between increased AMPAR function and FZP-withdrawal anxiety reversed by systemic GYKI injection. Previously we reported that in 1-day FZP-withdrawn rats anxiety-like behavior is observed in the elevated plus-maze test concomitant with an increase in AMPARmediated mepsc amplitude in hippocampal CA1 neurons, effects prevented by systemic injection of the noncompetitive AMPAR antagonist GYKI (Van Sickle et al, 2004). However, evidence of a correlation between enhanced AMPAR function and BZ withdrawal anxiety was not sought. Therefore, we first examined whether there is a positive correlation between AMPAR-mediated mepsc amplitude and BZ withdrawal anxiety score and once established whether normalizing mepsc amplitude via prior systemic antagonist injection could disrupt the correlation. Control and FZP-withdrawn rats were injected with vehicle (1% TWEEN-20) or GYKI (0.5 mg/kg) immediately following removal of FZP from the drinking water. Rats were then tested 1 day later in the elevated plus-maze prior to hippocampal slice preparation for mepsc recording. Fig. 1A shows representative inward mepscs recorded in CA1 neurons (V h = -80 mv) from control or 1-day FZP-withdrawn rats 24 51

58 hrs after vehicle or GYKI injection. AMPA currents amplitude increased significantly (30%) in neurons from 1 day FZP-withdrawn rats as reported previously (15-30%, Van Sickle et al, 2004) and there was a significant positive correlation (R 2 = 0.66, *p < ) between AMPAR-mediated mepsc amplitude and the percent openarm time (Fig. 1B), indicating a strong correlation between increased CA1 neuron AMPAR function and FZP withdrawal-anxiety. A positive correlation was also observed when control data from all vehicle-injected rats (n=25 cells), used to evaluate AMPAR, NMDAR and VGCC antagonist effects, were pooled (R 2 = 0.25, p = 0.01, Fig. 1B, 3C and 4C). Prior systemic injection of GYKI-5246 abolished this correlation (Fig. 1C, R 2 = 0.03, p = 0.52). Together these findings further support a role for upregulation of CA1 neuron AMPAR function and increased hippocampal output activity in FZP withdrawalanxiety. Antagonism of AMPAR activation at withdrawal onset prevented both upregulation of AMPAR function and downregulation of NMDAR function in 2-day FZPwithdrawn rats We postulated that the increase in AMPAR-mediated current during FZP-withdrawal initiates downregulation of NMDAR-mediated channel activity to offset the AMPARmediated expression of withdrawal-anxiety, suggesting a compensatory role for downregulation of NMDAR-mediated function secondary to the AMPAR-mediated CA1 neuron hyperexcitability observed (Van Sickle et al, 2002; Van Sickle et al, 2004). To evaluate this hypothesis, additional groups of rats were injected with GYKI immediately after FZP withdrawal and tested 1 day later in the plus-maze. Hippocampal 52

59 slice preparation was performed the following day and both AMPAR-mediated mepscs and NMDAR-mediated eepscs were then evaluated in slices from the same, then 2-day FZP-withdrawn rats. As previously reported, in 1-day FZP-withdrawn rats injected with vehicle exhibited anxiety-like behavior (Van Sickle et al, 2004), i.e. there was a significant reduction in the percentage of time spent on open-arms (Fig 2A; FZP/VEH: 16.0 ± 1.6%, n=9) relative to controls (CON/VEH: 26.9 ± 1.3%, n=10, *p<0.05). Prior GYKI injection completely prevented the reduction in time spent on open-arms and thus the behavioral expression of anxiety (CON/GYKI: 29.6 ± 2.3%, n=9; FZP/GYKI: 26.9 ± 2.1%, n=15. p>0.05). AMPAR currents were elevated (~30%) in neurons from the same subset of rats in which whole-cell currents could also be recorded (Fig. 2B; CON/VEH: 8.7 ± 0.1 pa, n=5; FZP/VEH: 11.1 ± 0.5 pa, n=6.*p<0.05). This significant current elevation was also eliminated by prior GYKI injection (CON/GYKI: 9.7 ± 0.5 pa, n=6; FZP/GYKI: 9.1 ± 0.4 pa, n=6). There were no differences in resting membrane potential (CON/VEH: ± 1.0 mv; FZP/VEH: ± 1.9 mv; CON/GYKI: ± 1.5 mv; FZP/GYKI: ± 1.4 mv. p>0.05), mepsc decay (CON/VEH: 20.6 ± 2.1 ms; FZP/VEH: 20.8 ± 2.0 ms; CON/GYKI: 17.4 ± 2.7 ms; FZP/GYKI: 19.3 ± 3.0 ms. p>0.05) or in the frequency (CON/VEH: 0.26 ± 0.11 Hz; FZP/VEH: 0.31 ± 0.19 Hz; CON/GYKI: 0.27 ± 0.12 Hz; FZP/GYKI: 0.28 ± 0.06 Hz. p>0.05) of AMPAR-mediated events between control and FZP-withdrawn rats that received vehicle versus GYKI injection. The effect of GYKI to avert the increased AMPAR-medicated current in 1-day FZP-withdrawn rats also prevented AMPAR overactivity 2 days after FZP withdrawal. This observation suggests that enhanced AMPA function early during the FZP withdrawal phase may be 53

60 required to initiate sustained AMPAR overactivity as reported by Stepphun and Turski (1993). Since AMPAR functional changes appeared earlier following drug withdrawal than alterations in NMDAR function (Van Sickle et al, 2004), a critical question was whether the reduction in evoked NMDAR-mediated currents was secondary to increases in AMPAR-mediated currents. We thus evaluated NMDAR-mediated eepscs in hippocampal slices from the same group of 2-day FZP-withdrawn rats in which AMPAR mepscs were recorded after 0-day GYKI or vehicle injection and 1-day elevated plus-maze testing. In the presence of AMPA, GABA A and GABA B receptor antagonists, inward EPSCs were evoked in neurons voltage-clamped from -80 to +40 mv. Representative inward NMDAR eepsc traces at V H= -20 mv from control or FZPwithdrawn rats which received vehicle or GYKI injection immediately after drug withdrawal are shown in Fig 2C. As reported previously (Van Sickle et al, 2004), the NMDA-mediated eepsc amplitude in CA1 neurons from vehicle-injected 2 day FZPwithdrawn rats was significantly reduced in comparison to their matched vehicle-injected control group (Fig 2D. V H = -20mV: CON/VEH, ± 24.6 pa, n=7; FZP/VEH, ± 9.6 pa, n=8. * p<0.05; V H = +40mV: CON/VEH: ± 43.9 pa, n=7; FZP/VEH: 97.9 ±21.9 pa, n=8. *p<0.05). Prior systemic GYKI injection had no effect on NMDAR mediated currents in neurons from control rats (Fig 2D. V H = -20 mv: CON/GYKI: ± 20.5 pa, n=6; V H = +40 mv: CON/GYKI: ± 69.0 pa, n=6). However, in neurons from 2-day FZP-withdrawn rats injected with GYKI immediately after FZP withdrawal, unlike in those injected with vehicle, eepsc amplitudes were equivalent to those observed in neurons from control rats at both 54

61 negative and positive holding potentials (Fig 2D. V H = -20 mv: FZP/GYKI, ± 19.4 pa, n=7; at V H = +40 mv: FZP/GYKI, ±31.9, n=7. p>0.05). Thus GYKI injection immediately after FZP withdrawal not only reversed both anxiety in 1- day FZP withdrawn rats and AMPAR upregulation in (1-day and) 2-day FZP-withdrawn rats, but also reversed downregulation of NMDAR function 2-days after FZP withdrawal. This finding suggested that the reduction in NMDAR function might play a negative feedback role to prevent expression of CA1 neuron hyperexcitability and thus anxietylike behavior. MK-801 failed to prevent anxiety-like behavior and upregulation of AMPAR function. Using a similar pharmacological strategy we evaluated if, as with some forms activitydependent plasticity, antagonism of the NMDAR-dependent signaling pathway during FZP withdrawal is required for the enhancement of AMPAR mepscs observed. We injected MK-801 immediately after FZP withdrawal and rats were tested in the elevated plus-maze on day 1 of withdrawal followed 1 day later by hippocampal slice preparation. As shown in Fig. 3A and as observed following injection of the other antagonist vehicles (Fig. 1A and 2A), saline injection had no effect on the appearance of withdrawal anxiety measured as a decreased percentage of time spent on open-arms in 1-day FZP-withdrawn rats in comparison to control rats (CON/VEH: 35.4 ± 2.6%, n=5; FZP/VEH: 24.0 ± 1.7%, n=5. *p<0.05). Likewise, prior MK-801 injection had no effect to modify anxiety-like behavior in the plus-maze (CON/MK-801: 36.3 ± 2.9%, n=7; FZP/MK-801: 22.8 ± 2.7% 55

62 n=6, *p<0.05). As in other studies of AMPAR mepscs in CA1 neurons from the same rats (Fig 3B), there were no differences in resting membrane potential (CON/VEH: ± 0.7 mv, FZP/VEH: ± 0.8 mv; CON/MK-801: ± 1.3 mv, FZP/MK-801: 9.1 ± 0.4 mv. p>0.05), current decay (CON/VEH: 17.8 ± 3.5 ms; FZP/VEH: 17.8 ± 1.5 ms; CON/MK-801: 17.6 ± 3.1 ms; FZP/MK-801: 18.3 ± 2.9 ms. p>0.05) or in mepsc frequency (CON/VEH: 0.40 ±0.25 Hz; FZP/VEH: 0.38 ± 0.56 Hz; CON/MK-801: 0.28 ± 0.12 Hz; FZP/MK-801: 0.39 ± 0.28 Hz. p>0.05) between control and FZP-withdrawn rats. However, as in other FZP-withdrawn groups (Figs. 1 and 2) there was a significant increase in peak mepsc amplitude after FZP withdrawal in comparison to controls (CON/VEH: 8.7 ± 0.3 pa, n=5; FZP/VEH: 10.7 ± 0.5 pa, n=9. *p<0.05). Prior MK-801 injection also had no effect on control mepsc amplitude. Furthermore, AMPAR mepsc amplitude in CA1 neurons from FZP-withdrawn rats remained increased (CON/MK801: 8.5 ±0.5 pa, n=7; FZP/MK801: 11.1 ± 0.5 pa, n=9. *p<0.05). This finding suggests that in comparison to other forms of activity-dependent synaptic plasticity, NMDARdependent signaling may not be required for the increased AMPAR-mediated glutamatergic strength during FZP withdrawal. We also investigated the effect of MK-801 on the correlation between AMPAR mepsc amplitude and the percentage of time rats spent on open-arms of the plus maze. There was a positive correlation (R 2 =0.29, *p < 0.05) between AMPAR-mediated mepsc amplitude and percent time open-arm time (Fig. 3C) in vehicle-injected FZPwithdrawn neurons, indicating a strong correlation between AMPAR function and FZP withdrawal anxiety as shown in Fig. 1B. Prior systemic MK-801 injection also failed to modify this significant positive correlation (R 2 =0.74, *p<0.0001, Fig. 3D). These 56

63 findings further support the lack of involvement of NMDAR-mediated Ca 2+ influx function in initiating the enhancement of CA1 neuron AMPAR-mediated currents during FZP withdrawal and the associated behavioral expression of anxiety. Nimodipine reverses upregulation of AMPAR function and prevents anxiety-like behavior during FZP withdrawal L-type VGCCs have been shown to be involved in mediating withdrawal signs to a variety of selective and non-selective CNS depressants including morphine, ethanol, the barbiturates and the benzodiazepines (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999) as well as in several forms of activity dependent plasticity (Morgan and Teyler, 1999; Borroni et al, 2000; Schinnick-Gallagher et al, 2003). Therefore, we investigated the possible role of L-type VGCC activation to modulate the increased glutmatergic strength reflected in the enhanced AMPARmediated mepsc amplitude during FZP withdrawal. Thus, we systemically injected rats with nimodipine, a selective L-type VGCC antagonist, immediately after removal of FZP from the drinking water. Rats were then tested in elevated plus-maze 1 day later followed by hippocampal slice preparation and recording of AMPAR-mediated mepscs. Fig. 4A shows that vehicle injection again had no effect on anxiety-like behavior in this group of 1-day FZP-withdrawn rats compared to control rats (CON/VEH: 33.3 ± 2.9%, n=8; FZP/VEH: 18.9 ± 2.8% n = 8; CON/NIMO: 31.4 ± 2.6%, n=7; FZP/NIMO: 32.9 ± 2.8%, n=9. * p<0.05). Moreover, prior vehicle injection again had no effect on basal AMPAR mepsc amplitude in control neurons or their enhancement during FZP 57

64 withdrawal (CON/VEH: 8.0 ± 1.4 pa, n=8; FZP/VEH: 11.2 ±1.9 pa, n = 8. *p<0.05). However, following nimodipine injection at the onset of withdrawal the mean AMPAR mepsc amplitude returned to control levels in neurons from 2-day FZP-withdrawn rats (CON/NIMO: 31.4 ± 2.6%, n=7; FZP/NIMO: 32.9 ± 2.8%, n=9. p> 0.05). There were no differences in resting membrane potential (CON/VEH: ± 1.9 mv, FZP/VEH: ±1.7 mv; CON/NIMO: ± 1.2 mv, FZP/NIMO: ± 1.9 mv. p>0.05), current decay (CON/VEH: 17.6 ± 3.1 ms; FZP/VEH: 14.6 ± 1.2 ms; CON/NIMO: 15.1 ±0.8 ms; FZP/NIMO: 18.3 ± 2.8 ms. p>0.05) or in mepsc frequency (CON/VEH: 0.23 ± 0.03 Hz; FZP/VEH: 0.34 ± 0.06 Hz; CON/NIMO: 0.27 ± 0.06 Hz; FZP/NIMO: 0.27 ± 0.04 Hz. p>0.05) in neurons from control vs. FZP-withdrawn rats. The same electrophysiological and behavioral measures were also used to investigate the effect of systemic nimodipine or vehicle injection on the relationship between AMPAR mepsc amplitude and anxiety-like behavior. As in GYKI and MK-801 studies (Fig. 1 and Fig. 3), there was a positive correlation (R 2 =0.48, *P < 0.005) between AMPAR-mediated mepsc amplitude and percent open-arm time (Fig. 4C) in FZP-withdrawn rats, indicating an analogous, strong correlation between AMPAR function and FZP withdrawal-anxiety in these additional groups of vehicle-injected rats. Prior systemic nimodipine injection eliminated the significant positive relationship among the electrophysiological and behavioral measures across experimental groups (R 2 =0.20, p=0.08, Fig. 4D). Together these findings suggest that L-type VGCCdependent Ca 2+ signaling may be required for enhancement of AMPAR function and the associated anxiety during FZP withdrawal. 58

65 Nimodipine concentration-response To evaluate whether systemic nimodipine injection may have a direct effect to modify AMPA receptor activity, the concentration-response profile for the effect of nimodipine on AMPAR-mediated mepsc characteristics was evaluated. mepscs were recorded for 5 min without nimodipine during the baseline hippocampal slice-recording period. Nimodipine was then superfused in increasing concentrations ranging from 0.1 µm to 100 µm. Average AMPAR mepsc amplitude (V H = -80 mv) in CA1 neurons during vehicle (close circles) or nimodipine (open circle) superfusion was compared. As shown in Fig. 5, neither nimodipine (n=4 cells) or vehicle (n=3 cells) had any effect on mepsc amplitude (p>0.05) at any concentration evaluated as represented by the highest nimodipine concentration (100 µm) tested (VEH: 9.1 ± 0.3 pa; NIMO: 8.7 ± 0.4 pa, p>0.05). There were also no significant changes (p>0.05) in mepsc rise time (VEH: 3.8 ± 0.5 ms; NIMO: 3.6 ± 0.3 ms), decay (VEH: 23.4 ± 6.5 ms; NIMO: 20.8 ± 1.4 ms), or frequency (VEH: 0.30 ± 0.0 7Hz, NIMO: 0.34 ± 0.06 Hz) during 100 µm nimodipine superfusion. 59

66 DISCUSSION We previously reported that transient plasticity of hippocampal CA1 neuron glutamate receptors contributes to benzodiazepine withdrawal-anxiety (Van Sickle et al, 2004). The goal of the present studies was to provide further support for the relationship between hippocampal CA1 neuron AMPAR-mediated excitation and the possible neurophysiological mechanisms associated with the behavioral expression of withdrawalanxiety after chronic benzodiazepine treatment. In the present study, there was a positive correlation between upregulation of AMPAR-mediated function and anxiety-like behavior in 1-day FZP-withdrawn rats (Fig. 1A). Antagonism of AMPAR activation by GYKI immediately after cessation of FZP treatment prevented both the subsequent AMPAR functional upregulation in CA1 neurons and FZP-withdrawal anxiety in rats, as well as the correlation between them (Fig. 1B and Van Sickle et al, 2004). Further, prior GYKI injection prevented both AMPAR upregulation in 2- day FZP-withdrawn rats, as well as down-regulation of NMDAR function (Fig. 2). Using a similar strategy, we evaluated the actions of the NMDAR antagonist, MK-801 and the L-type VGCC antagonist, nimodipine, systemically injected at the onset of withdrawal to modify AMPAR plasticity. MK-801 failed to prevent enhancement of AMPAR currents and the correlated anxiety in 1-day FZP-withdrawn rats (Fig. 3) while nimodipine reversed both the enhancement of AMPAR function and the expression of anxiety, as well as their correlation (Fig. 4). Nimodipine had no direct effect on mepscs (Fig. 5), supporting the likelihood that its effect to reverse the withdrawal-associated increased in AMPAR function was through blockade of L-type VGCCs. 60

67 The neurophysiological mechanisms underlying the behavioral changes occurring during benzodiazepine withdrawal are presently not well defined. Numerous studies support an important role for non-nmdar glutamatergic mechanisms in drug dependence and addiction. Increased glutamatergic strength in mesolimbic reward pathways has been implicated in addictive behaviors (Tzschentke and Schmidt, 2003; Nestler, 2005) and evidence supports a role for enhanced hippocampal glutamatergic plasticity underlying dependence on ethanol (Molleman and Little, 1995; Sanchis-Segura et al, 2006) and morphine (Vekovischeva et al, 2001; Zhong et al, 2006). Likewise, there is growing evidence of neuroadaptive changes in the glutamatergic system associated with benzodiazepine withdrawal (Stephens, 1995; Izzo et al, 2001; Allison and Pratt, 2003; Van Sickle et al, 2004). We previously reported that during FZP withdrawal, anxiety-like behavior appears in rats concomitant with increased hippocampal CA1 neuron hyperexcitability reflected in an increased frequency of extracellular spiking during 4-aminopyridine superfusion, increased AMPAR-mediated mepsc amplitude, localized upregulation of AMPAR ligand binding and protein expression and insertion of GluR1-containing AMPARs (Van Sickle and Tietz, 2002; Van Sickle et al, 2004; Song and Tietz, 2006). This study further supports the strong link between CA1 neuron AMPAR plasticity in vitro and withdrawal-anxiety in vivo (Fig 1A), paralleling the rapid elimination of FZP and its metabolites in the rat hippocampus (Lau et al, 1987; Xie and Tietz, 1992). Systemic GYKI injection at the onset of this period abolished this correlation (Fig. 1B) by averting both the increased AMPAR mepsc amplitude in vitro and FZP-withdrawal anxiety in vivo, confirming a contributory role for hippocampal CA1 neuron AMPAR-mediated excitation in mediating withdrawal symptoms. Thus, 61

68 activation of CA1 neuron AMPARs was required for inducing withdrawal symptoms similar to the observations of Steppuhn and Turski (1993) in which administration of GYKI during the silent phase, days 1-3 of withdrawal in which diazepamdependent mice were symptom free, would prevent the subsequent development of withdrawal signs in the active phase, days 4-21 of withdrawal during which mice displayed classic benzodiazepine withdrawal signs. Consequently, this finding considerably strengthens the hypothesis that AMPAR-mediated hyperexcitability in hippocampal CA1 neurons is an important part of a functional anatomic circuit (Millan 2003) which contributes to the FZP-withdrawal anxiety in rats. Regulation of NMDAR receptors in a variety of brain areas including hippocampus has also been implicated in the development of tolerance to and dependence on a variety of drugs of abuse. However, while the synaptic strength of non-nmda receptors is typically increased in relation to dependence on CNS depressants (Molleman and Little, 1995; Jang et al, 2000; Vekovischeva et al, 2001; Zhong et al, 2006), the effects on NMDA receptors are less uniform (Whittington et al, 1995; Jang et al, 1998; Van Sickle et al, 2004). Based on the prior observation that AMPAR functional upregulation preceded NMDAR functional downregulation (Van Sickle et al, 2004), the observation that down-regulation of NMDARs in 2-day FZP-withdrawn rats was also averted by prior AMPAR antagonist injection suggested that downregulation of CA1 neuron NMDAR-mediated function may be secondary to enhanced AMPAR function (Fig. 2 C and D). Moreover, these findings suggest that NMDAR regulation may play a compensatory, feedback role to prevent CA1 neuron hyperexcitability, thus dampening hippocampal output activity and preventing benzodiazepine-induced withdrawal anxiety. 62

69 These studies support a neuroadaptive mechanism by which the imbalance in excitatory glutmatergic receptors during benzodiazepine withdrawal is initially AMPAR-dependent and is subsequently counterbalanced by down-regulation of CA1 neuron NMDAR function, represented by a reduction in both NMDAR-mediated eepsc amplitude (Fig 2, Van Sickle et al, 2004) and the levels of NR2B mrna and protein in the CA1 region (Van Sickle et al, 2002). Similarly, in previously fear-conditioned rats, a model of cueinduced anxiety in which increased glutamatergic strength is reported in lateral amygdala (LA) pyramidal neurons, downregulation of NMDA-mediated synaptic strength was indicated by a 3-4 fold shift in LA neuron NMDA potency concomitant with a decreased sensitivity of NMDA currents to the NR2B-subunit selective antagonist, ifenprodil. In fear-conditioned rats this effect was suggested to protect against NMDAR recruitment during induction and consolidation of fear memories (Zinebi et al, 2003). It will be of significant interest to determine if NMDAR subunit composition may be altered in FZPwithdrawn rats, specifically a dynamic shift in the NR2B to NR2A subunit ratio at CA1 synapses (Van Sickle et al 2002; Perez-Otano and Ehlers, 2005). While the role of NMDA receptors in expression of fear-induced anxiety is still controversial (Zinebi et al, 2003, Rodrigues et al, 2004), taken together with findings in FZP-withdrawn CA1 neurons these data suggest the possibility that downregulation of NMDARs, likely the regulation of NR2B-containing receptors, may be a common negative feedback mechanism to avert expression of anxiety-like behavior associated with increased glutamatergic strength in brain nuclei subserving anxiety. Neuronal plasticity is defined as a use-dependent increase in the efficiency of synaptic transmission. Examples include the phenomenon of LTP, fear-conditioning and 63

70 repeated drug administration (Malenka 2003; Rodrigues et al, 2004; Nestler 2005). Activity-dependent changes in synaptic function are primarily the consequence of an intracellular calcium-dependent biochemical cascade and involve changes in synaptic AMPAR number and/or function (Song and Huganir, 2002; Boehm and Malinow, 2005). Regulation of AMPAR recruitment by NMDAR-dependent calcium signaling has been well described in activity-dependent plasticity models (Shi et al, 1999; Malinow 2003). However, systemic administration of the NMDAR antagonist MK-801, previously reported to restore downregulation of NMDAR function in CA1 neurons from flurazepam-withdrawn rats (Van Sickle et al, 2004), failed to reverse the induction of CA1 neuron AMPAR upregulation and the concomitant anxiety in FZP-withdrawn rats (Fig. 3A and 3B) or the correlation between these measures (Fig. 3C and 3D). This suggests that an NMDAR-independent signaling pathway may underlie AMPAR upregulation associated with FZP-withdrawal anxiety. Voltage-gated calcium channels (VGCCs) have also been reported to play role in each of the above-mentioned activity-dependent models of synaptic plasticity (Morgan and Teyler, 1999; Borroni et al, 2000; Rajadhyksha and Kosofsky, 2005). Indeed, a role for VGCC-mediated Ca 2+ influx in dependence on other CNS depressants such as ethanol, barbiturates and morphine has been reported (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999). L-type calcium channel blockers like nimodipine, nifedipine and verapamil were reported to block a variety of lorazepam withdrawal signs including hyperkinesia, hyperthermia, hyperaggression and audiogenic seizures (Chugh et al, 1992; Gupta et al, 1996; Ganouni et al, 2004). Nitrendipine dosedependently decreased seizures precipitated by the benzodiazepine receptor partial 64

71 agonist FG7142 (Dolin et al, 1990). The finding that injection of the L-type VGCC antagonist, nimodipine reversed both the upregulation of CA1 neuron AMPAR function in 1-day FZP withdrawn rats as well as withdrawal-anxiety (Fig 4A and 4B) raises the possibility that calcium influx through L-type VGCCs, may activate diverse intracellular messengers or transcription factors to increase glutamatergic strength during benzodiazepine withdrawal. In fact, it has been well-described that in ethanol withdrawal and following psychostimulant administration, that activation of L-type VGCCs can result in the influx of intracellular calcium and downstream activation of Ca 2+ /CaMactivated kinase and phosphatase pathways, eventually involving CREB-induced gene expression, also important to neuronal and experience-dependent plasticity (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005). DHPs are highly selective calcium channel blockers. Nimodipine binds to rat, guinea pig and human brain membranes with high affinity (less than 1 nm) and concentration as low as nm were reported to block inward Ca 2+ currents (Scriabine et al, 1989). DHPs can block a variety of non-l-type calcium channels, including GABA A R, nicotinic acetylcholine receptors and 5-HT3A receptors (Hargreaves et al, 1996; Houlihan et al, 2000; Das et al, 2004) at IC 50 s in the micromolar to milimolar range. Considering the much higher affinity and efficacy of nimodipine on L-type VGCCs than other channels (nm vs. µm to mm), impairment of L-type VGCC mediatedsignaling pathways is the most reasonable explanation for nimodipine effects to avert the increase in CA1 neurons mepsc current amplitude and FZP-withdrawal anxiety. Indeed, Fig. 5 showed nimodipine in the µm concentration range had no direct effects on AMPAR-mediated mepscs supporting the latter explanation. Furthermore, it has been 65

72 suggested that in the brain the density of neuronal L-type Ca 2+ channels are significantly higher than that of vascular ones (Ricci et al, 2002). This might explain the more pronounced neuronal than vascular effects after pharmacological manipulation of cerebral Ca 2+ channels. Further studies are needed to determine the cellular mechanisms by which L-type VGCCs are activated during FZP withdrawal. Currently we favor two hypotheses. First, a few studies have shown an increased concentration of intracellular calcium through L- type VGCCs following prolonged GABA A receptor activation that leads to bicarbonatedriven Cl - entry and Cl - accumulation and eventually promotes cell depolarization (Reichling et al, 1994; Chavas et al, 2004; Marty A and Liano I, 2006). As a consequence of prolonged GABA A receptor activation during FZP administration numerous time-dependent changes occur at the GABA A receptor some of which influence CA1 neuron hyperexcitability (Van Sickle et al., 2004) including a bicarbonate-driven Cl - accumulation reflected in a shift in the Cl - reversal potential and the appearance of a bicuculline-sensitive depolarizing potential (Zeng et al, 1995; Zeng and Tietz, 1996; Zeng and Tietz, 2000). Secondly, since this flurazepam treatment and other common diazepam treatments (Gallager et al, 1985) result in brain levels of benzodiazepine metabolites in the low micromolar range (0.6 μm in diazepam equivalents), benzodiazepines might also directly modulate VGCCs activity, as shown by their ability to directly inhibit VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Gershon, 1992; Reuveny et al, 1993; Ishizawa Y et al, 1997). The modulation of L-type VGCC activity during benzodiazepine withdrawal is currently under investigation (Xiang and Tietz, 2006). 66

73 On the basis of evidence to date, a model of the proposed cellular mechanisms underlying benzodiazepine withdrawal hyperexcitability in hippocampal CA1 neurons is shown in Fig. 6, which provides a framework for the hypotheses that direct our ongoing mechanistic studies. L-type VGCCs could be activated by the GABA A receptor-mediated membrane depolarization (Zeng and Tietz, 1995) and/or channel activity enhanced by direct benzodiazepine effects during prolonged drug administration. Increased L-type VGCC-mediated Ca 2+ entry and activation of downstream Ca 2+ -mediated signaling cascades, including activation of well-described kinase and phosphatase pathways (Rajadhyksha and Kosofsky, 2005), could contribute to AMPAR plasticity, likely by increasing GluR1-containing AMPAR channel number, modifying AMPAR subunit composition (Van Sickle and Tietz, 2002; Song and Tietz, 2006) and/or subunit phosphorylation similar to other models of activity-driven synaptic plasticity (Malenka 2003; Boehm and Malinow, 2005). Enhanced AMPAR function could further promote CA1 neuron membrane depolarization and serve as a major driving force contributing to the observed CA1 neuron hyperexcitability and benzodiazepine withdrawal-anxiety (Van Sickle et al, 2004). In addition, enhanced AMPAR activation would lead to NMDAR activation and subsequent downregulation, possibly of NR2B-containing NMDARs (Van Sickle et al, 2002, Van Sickle et al, 2004), which likely plays a protective role to prevent AMPAR-driven over-excitation in CA1 neurons, as well as in anxiety circuits in other areas of the brain (Zinebi et al, 2003; Rodrigues et al, 2004). In summary, the present study provides evidence that enhanced CA1 neuron AMPAR activity contributes to BZ withdrawal anxiety, which leads to a localized downregulation of NMDAR activity. Further studies suggest that enhanced AMPA- 67

74 mediated glutamatergic strength may be through an L-type VGCC-, rather than NMDARdependent signaling pathway. Given that many drugs of abuse induce similar manifestations of drug withdrawal, L-type VGCC mediated calcium signaling and glutamate receptor plasticity may constitute a common neurophysiological mechanism for the expression of withdrawal-anxiety and drug dependence. 68

75 Acknowledgements: We are grateful to Krista Pettee, Margarete Otting and Eugene Orlowski for technical assistance. The National Institute of Drug Abuse Drug Supply Program supplied flurazepam. This work was supported by Department of Health and Human Services grants from the National Institute on Drug Abuse: R01-DA and R01-DA to EIT and a predoctoral fellowship from the University of Toledo, College of Medicine to KX. 69

76 References Allison C, Dunworth SJ, Stephens DN, Pratt JA (1999). Involvement of AMPA-receptors in benzodiazepine withdrawal and sensitization processes. Behav Pharmacol 10 (Suppl. 1): S2. Allison C, Pratt JA (2003). Neuroadaptic processes in GABAergic and glutamatergic systems in benzodiazepine dependence. Pharmacol Ther 98: Boehm J, Malinow R (2005). AMPAR receptor phosphorylation during synaptic plasticity. Biochem Soc Trans 33: Borroni AM, Fichtenholtz H, Woodside BL, Teyler TJ (2000). Role of voltage-dependent calcium channel long-term potentiation (LTP) and NMDA LTP in spatial memory. J Neurosci 20: Chavas J, Forero ME, Collin T, Liano I and Marty A (2004). Osmotic tension as a possible link between GABA A receptor activation and intracellular calcium elevation. Neuron 44: Chugh Y, Saha N, Sankaranarayanan A., Sharma PL (1992). Effect of peripheral administration of cinnarizine and verapamil on the abstinence syndrome in diazepamdependent rats. Psychopharmacology 106: Das P, Bell-Horner CL, Huang RQ, Raut A, Gonzales EB, Chen ZL, Covey DF and Dillon GH (2004). Inhibition of type A GABA receptors by L-type calcium channel blockers. Neuroscience 124: Degroot A, Kashluba S, Treit D (2001). Septal GABAergic and hippocampus cholinergic systems modulate anxiety in the plus-maze and shock-probe tests. Pharmacol Biochem Behav 69 :

77 Dolin SJ, Path TL, Rabbani M, Siarey RJ, Bowhay AR Little HJ (1990). Nitrendipine decreases benzodiazepine withdrawal seizures but not the development of benzodiazepine tolerance or withdrawal signs. Br J Pharmacol 101: Dunworth SJ, Stephens DN (1998). Sensitization to repeated withdrawal, in mice treated chronically with diazepam, is blocked by an NMDA receptor antagonist. Psychopharmacology 136: Gallager DW, Malcolm AB, Anderson SA, Gonsalves SF (1985). Continuous release of diazepam: electrophysiological, biochemical and behavioral consequences. Brain Res 342: Ganouni SE, Hanoun N, Boni C, Tazi A, Hmon F (2004). Prevention of diazepam withdrawal syndrome by nifedipine behavioral and neurochemical studies. Pharmacology, Biochemistry and Behavior 79: Gershon E (1992). Effect of benzodiazepine ligands on Ca 2+ channel currents in Xenopus oocytes injected with rat heart RNA. J Basic Clin Physiol Pharmacol 3: Groth RD, Dunbar RL, Mermelstein PG (2003). Calcineurin regulation of neuronal plasticity. Biochem Biophys Res Commun 311: Griffiths RR and Johnson MW (2005). Relative abuse liability of hypnotic drugs: a conceptual framework and algorithm for differentiating among compounds. J Clin Psych 66 (Suppl 9): Gupta MB, Nath C, Patnaik GK, Saxena RC (1996). Effect of calcium channel blockers on withdrawal syndrome of lorazepam in rats. Indian J Med Res 103:

78 Hargreaves AC, Gunthorpe MJ, Taylor CW, Lummis SCR (1996). Direct inhibition of 5- hydroxytryptamine 3 receptors by antagonists of L-type Ca 2+ channels. Mol Pharmacol 50: Houlihan LM, Slater EY, Beadle DJ, Lukas RJ, Bermudez I (2000). Effects of diltiazem on human nicotinic acetylcholine and GABA(A) receptors. Neuropharmacology 39: Ishizawa Y, Furuya K, Yamagishi S and Dohi S (1997). Non-GABAergic effects of midazolam, diazepam and flumazenil on voltage-dependent ion currents in NG cells. NeuroReport 8: Izzo E, Auta J, Impagnatiello F, Pesold C, Guidotti A, Costa E (2001). Glutamic acid decarboxylase and glutamate receptor changes during tolerance and dependence to benzodiazepines. Proc Natl Acad Sci USA 98: Jackson A, Meas AN, Stephens DN (2000). Behavioral effects of α-amino-3-hydroxy-5- methyl-4-isoxazolepropionate-receptor antagonists and their relevance to substance abuse. Pharmacol Ther 88: Jang Choon-Gon, Oh S, Ho IK (1998). Changes in NMDAR2 subunit mrna levels during pentobarbital tolerance/withdrawal in the rat brain: an in situ hybridization study. Neurochem Res 23: Jang Choon-Gon; Rockhold RW; Ho IK (2000). An autoradiographic study of [ 3 H]AMPA receptor binding and in situ hybridization of AMPA sensitive glutamate receptor A (GluR-A) subunits following morphine withdrawal in the rat brain. Brain Res Bull 52:

79 Koff JM, Pritchard GA, Greenblatt DJ, Miller LG (1997). The NMDA receptor competitive antagonist CPP modulates benzodiazepine tolerance and discontinuation. Pharmacololgy 55: Lamprea MR, Cardenas FP, Silveira R, Walsh TJ, Morato S (2003). Effects of septal cholinergic lesion on rat exploratory behavior in an open-field. Braz J Med Biol Res 36: Lau CE, Falk JL, Dolan S, Tang M (1987). Simultaneous determination of lorazepam and five metabolites in serum by high-performance liquid chromatography and its application to pharmacokinetic studies in rats. J Chromatogr 423: Little HJ (1999). The contribution of electrophysiology to knowledge of the acute and chronic effects of ethanol. Pharmacol Ther 84: Malenka RC (2003). Synaptic plasticity and AMPA receptor trafficking. Ann N Y Acad Sci 1003:1-11. Malinow R (2003). AMPA receptor trafficking and long-term potentiation. Philos Trans R Soc Lond B Biol Sci 358 : Marty A. and Llano I (2006). Excitatory effects of GABA in established brain networks. Trends Neurosci 28: Millan MJ (2003). The neurobiology and control of anxious states. Prog Neurobiol 70: Molleman A, Little HJ (1995). Increases in non-n-methyl-d-aspartate glutamatergic transmission, but no change in gamma-aminobutyric acid B transmission, in CA1 neurons during withdrawal from in vivo chronic ethanol treatment. J Pharmacol Exp Ther 274:

80 Morgan SL, Teyler TJ (1999).VDCCs and NMDARs underlie two forms of LTP in CA1 hippocampus in vivo. J Neurophysiol 82: Nestler EJ (2005). Is there a common molecular pathway for addiction? Nat Neurosci 8: Perez-Otano I, Ehlers MD (2005). Homeostatic plasticity and NMDA receptor trafficking. Trends in Neuroscience 28: Podhorna J (2002). The experimental pharmacotherapy of benzodiazepine withdrawal. Current Pharmaceut Design 8: Pourmotabbed A, Motamedi F, Fathollahi Y, Mansouri FA, Semnanian S (1998). Involvement of NMDA receptors and voltage-dependent calcium channels on augmentation of long-term potentiation in hippocampal CA1 area of morphine dependent rats. Brain Res 804: Rabbani M, Little HJ (1999). Increase in neuronal Ca 2+ flux after withdrawal from chronic barbiturate treatment. E J Pharmacol 364: Rajadhyksha AM and Kosofsky BE (2005) Psychostimulants, L-type calcium channels, kinases, and phosphatases. The Neuroscientist 11: Reichling DB, Kyrosis A, Wang J and MacDermott AB (1994). Mechanisms of GABA and glycine depolarization-induced calcium transients in rat dorsal horn neurons. J Physiol 476: Reuveny E, Twombly DA and Narahashi T (1993). Chlordiazepoxide block of two types of calcium channels in neuroblastoma cells. J Pharmacol Exp Ther 264:

81 Ricci A, Sabbatini M, Tomassoni D, Mignini F, Petrelli C, Amenta F (2002). Neuronal populations of rat cerebral cortex and hippocampus expressed a higher density of L-type Ca2+ channel than corresponding cerebral vessels. Clin Exper Hyper 24: Rodgers RJ, Dalvi A (1997). Anxiety, defense and the elevated plus-maze. Neurosci Biobehav Rev. 21: Rodrigues SM, Schafe GE and LeDoux JE ( 2004). Molecular Mechanisms underlying emotional learning and memory in the lateral amygadala. Neuron 44: Sanchis-Segura C, Borchardt T, Vengetina V, Zghoul T, Bachteler D, Gass P, Sprengel R, Spanagel R (2006). Involvement of the AMPA receptor-c subunit in alcohol-seeking behavior and relapse. J Neurosci 26: Scriabine A, Schuurman T, and Traber J (1989) Pharmacological basis for the use of nimodipine in central nervous system disorders. FASEB J 3: Shi SH, Hayashi Y, Petralia RS, Zaman SH, Wenthold RJ, Svoboda K, Malinow R (1999). Rapid spine delivery and redistribution of AMPA receptors after synaptic NMDA receptor activation. Science 284: Shinnick-Gallagher P, McKernan MG, Xie J and Zinebi F (2003). L-type voltage-gated calcium channels are involved in the in vivo and in vitro expression of fear conditioning. Ann. NY Acad. Sci 985: Song I, Huganir RL (2002). Regulation of AMPA receptors during synaptic plasticity. Trends Neurosci 25: Song J, Tietz, EI (2006). Benzodiazepine-withdrawal involves LTP-like insertion of GluR1- containing AMPA receptors. Abstract Viewer/Itinerary Planner. Washington, DC: Soc for Neurosci., Online. 75

82 Steppuhn KG, Turski L (1993). Diazepam dependence prevented by glutamate antagonists. Proc. Natl Sci USA 90: Stephens DN (1995). A glutamatergic hypothesis of drug dependence. Behav Pharmacol 6: Taft WC, DeLorenzo RJ (1984). Micromolar-affinity benzodiazepine receptors regulate voltage-sensitive calcium channels in nerve terminal preparations. Proc Natl Acad Sci 81: Tzschentke and Schmidt (2003). Glutamatergic mechanisms in addiction. Mol Psychiatry. 8: Van Sickle B, Tietz EI (2002). Selective enhancement of AMPA receptor-mediated function in hippocampal CA1 neurons from chronic benzodiazepine-treated rats. Neuropharmacology 43: Van Sickle B, Cox A, Schak K, John GL, Tietz EI (2002). Chronic benzodiazepine administration alters hippocampal CA1 neuron excitability: NMDA receptor function and expression. Neuropharmacology 43: Van Sickle B, Xiang K Tietz EI (2004). Transient plasticity of hippocampus CA1 neuron glutamate receptors contributes to benzodiazepine withdrawal anxiety. Neuropsychopharmacology 29: Vekovischeva OY, Zamanillo D, Echenko O, Seppala T, Uusi-Oukari M, Honkanen A, et al. (2001). Morphine-induced dependence and sensitization are altered in mice deficient in AMPA-type glutamate receptor-a subunits. J of Neurosci 21:

83 Whittington MA, Little HJ (1993). Changes in voltage-operated calcium channels modify ethanol withdrawal hyperexcitability in mouse hippocampal slices. Exp Physiol 78: Watson WP, Little HJ (2002). Selectivity of the protective effects of dihydropyridine calcium channel antagonist against the ethanol withdrawal syndrome. Brain Res 930: Xia Z, Storm DR. (2005). The role of calmodulin as a signal integrator for synaptic plasticity. Nat Rev Neurosci. 6: Xiang K, Giovannucci DR, Greenfield Jr, LJ, Tietz EI. (2006) Enhanced calcium entry through L-type voltage-gated calcium channels contributes to AMPA receptor plasticity in hippocampal CA1 neurons after chronic benzodiazepine treatment. Abstract Viewer/Itinerary Planner. Washington, DC: Soc for Neurosci., Online. Xie XH, Tietz EI (1992). Reduction in potency of selective gamma-aminobutyric acid A agonists and diazepam in CA1 region of in vitro hippocampal slices from chronic flurazepam-treated rats. J Pharmacol Exp Ther 262: Zeng X, Xie XH and Tietz EI (1995). Reduction of GABA-mediated inhibitory post-synaptic potentials in hippocampus CA1 pyramidal neurons following oral flurazepam administration. Neuroscience 66: Zeng X and Tietz EI (1997). Depression of early and late monosynaptic inhibitory postsynaptic potentials in hippocampal CA1 neurons following prolonged benzodiazepine administration: role of reduction in Cl - driving force. Synapse 25: Zeng X and Tietz EI (2000). Role of bicarbonate ion in mediating decreased synaptic conductance in benzodiazepine tolerant hippocampal CA1 pyramidal neurons. Brain Res 868:

84 Zhong W, Dong Z, Tian M, Cao J, Xu T, Xu L, Luo J (2006). Opiate withdrawal induces dynamic expressions of AMPA receptors and its regulatory molecule CaMKII alpha in hippocampal synapses.2006, Life Sci 79: Zinebi F, Xie JG, Liu J, Russell RT, Gallagher JP, McKernan MG, and Shinnick-Gallagher P. (2003). NMDA currents and receptor protein are downregulated in the amygdala during maintenance of fear memory. J Neurosci 23:

85 A. CON/VEH 1 Day FZP/VEH CON/GYKI FZP/GYKI 20 pa 150 ms B. mepsc Amplitude (-pa) CON-VEH FZP-VEH % Open Arm Time C. mepsc Amplitude (-pa) CON-GYKI FZP-GYKI % Open Arm Time 79

86 Fig. 1 Effects of systemic GYKI injection on AMPAR-mediated mepscs in hippocampal CA1 neurons and anxiety-like behavior during flurazepam (FZP) withdrawal. A single injection of the non-competitive AMPAR antagonist, GYKI (GYKI, 0.5 mg/kg, i.p.) or 1% TWEEN-20 (0.5 ml/kg, i.p.) vehicle was given to rats immediately after stopping oral administration of FZP. Rats were tested 1 day later in the elevated plus-maze, followed 30 minutes later by hippocampal slice preparation for electrophysiological recording. A. Representative traces of mepscs from individual CA1 neurons from control (CON) or 1 day FZP-withdrawn rat groups. B. A significant positive correlation (R 2 =0.6591, P < ) was found between CA1 neuron AMPARmediated mepsc amplitude and the percent time spent on open arms of the elevated plus-maze, a measure of anxiety-like behavior, in individual rats that received vehicle injection. C. There was no correlation (R 2 = , p=0.5205) between AMPARmediate mepsc amplitude in CA1 neurons and percent open arm time from rats that received prior systemic GYKI injection. 80

87 A Day B Day % Open Arm Time * 10 0 CON FZP CON FZP Vehicle GYKI mepsc Amplitude (-pa) * CON FZP CON FZP Vehicle GYKI C. 50 D. 2 Day ms CON-VEH 400 CON-GYKI 300 * -50 FZP-VEH 200 FZP-GYKI mv eepsc Amplitude (-pa) FZP-VEH CON-GYKI FZP-GYKI CON-VEH * pa 81

88 Fig. 2 GYKI injection at the onset of withdrawal averts the appearance of anxiety-like behavior in 1 day FZP-withdrawn rats as well as the increased AMPAR-mediated mepsc amplitude and reduced NMDAR-mediated eepsc amplitude in the same, 2-day withdrawn rats. Rats were given single injection of GYKI (0.5 mg/kg, i.p.) or vehicle (0.5 ml/kg, i.p.) immediately after removal of FZP from the drinking water. Rats were tested 1 day after injection in the elevated plus-maze, followed by hippocampal slice preparation for electrophysiological recording 2 days after injection. A. Open arm time on an elevated plus-maze test, expressed as a percent of total time, 1 day following vehicle or GYKI injection. 1 day FZP-withdrawn rats (white bars, n=9) injected with vehicle showed evidence of anxiety-like behavior, measured as a significant reduction (p<0.05) in the percentage of open arm time compared to control rats (black bar, n=10). GYKI injection blocked the appearance of anxiety, in 1 day FZPwithdrawn rats (white bar, n=15) compared to control rats (black bar, n=9). B. Two days after FZP withdrawal, the rats tested in the plus maze the previous day were sacrificed and AMPAR mepscs (V H = -80 mv) were recorded for 5 min in hippocampal CA1 neurons from the both vehicle-injected and GYKI-injected control rats (black bars, n=6) or 2-day FZP-withdrawn rats (white bars, n=6). Prior GYKI injection (control rats, black bar, n=6; FZP rats, white bar, n=6) blocked AMPAR-mediated current. C. Representative whole-cell NMDAR current (V H = -20 mv) traces evoked by electrical stimulation ( ma) of the Schaffer collateral pathway. D. Averaged current-voltage (I-V) curves of peak NMDAR eepsc amplitude (V H =-80 to +40 mv) generated in CA1 neurons from vehicle-injected FZP-withdrawn rats (closed circles, n=8) in comparison to 82

89 neurons from vehicle-injected control rats (open circles, n=7). Prior GYKI injection reversed the reduction in NMDAR eepsc amplitude in neurons from FZPwithdrawn rats (open squares, n=7), and was without effect on neurons from control rats (closed squares, n=6, *p<0.05). 83

90 A. 40 B. 15 % Open Arm Time 30 * * CON FZP CON FZP Vehicle MK-801 mepsc Amplitude (-pa) * * CON FZP CON FZP Vehicle MK Day C. mepsc Amplitude (-pa) CON-VEH FZP-VEH D. mepsc Amplitude (-pa) CON-MK-801 FZP-MK % Open Arm Time % Open Arm Time

91 Fig. 3 Effects of systemic MK-801 injection on AMPAR-mediated mepscs in hippocampal CA1 neurons and anxiety-like behavior during FZP withdrawal. A single injection of the noncompetitive NMDAR antagonist, MK-801 (0.25 mg/kg, i.p.) or saline (1 ml/kg, i.p.) was given to rats immediately after FZP-withdrawal. Rats were tested 1 day after injection in the elevated plus-maze, followed by hippocampal slice preparation for electrophysiological recording. A. Average AMPAR mepsc amplitude (V H = -80 mv) in CA1 neurons from control (black bars) or 1-day FZP-withdrawn (white bars) rats. Neither vehicle (CON, n=5; FZP rat, n=9) nor MK-801 injection (CON, n=7; FZP, n=9) had an effect on up-regulation of AMPAR mepsc amplitude. B. Open arm time on an elevated plus-maze, expressed as a percent of total time, following vehicle or MK-801 injection. 1-day FZP-withdrawn rats (white bars, n=5), but not control rats (black bars, n=5), injected with vehicle showed anxiety-like behavior. MK-801 injection had no effect on the appearance of anxiety, measured as a significant reduction (* p< 0.05) in percent open arm time, in 1-day FZP-withdrawn rats (white bar, n=6) compare to control rats (black bar, n=7). C. Relationship between AMPAR-mediate mepsc amplitude in CA1 neurons and percent open arm time in individual rat that received vehicle injection. A significant positive correlation (R 2 =0.2934, p < 0.05) was found between mepsc amplitude and percent open arm time. D. Prior systemic MK-801 injection at the onset of FZP-withdrawal failed to reverse the significant positive correlation (R 2 =0.7397, p<0.0001) between CA1 neuron AMPAR-mediate mepsc amplitude and percent open arm time measured in the elevated plus-maze test. 85

92 A. 40 B. % Open Arm Time * 10 0 CON FZP CON FZP Vehicle Nimodipine mepsc Amplitude (-pa) * CON FZP CON FZP Vehicle Nimodipine 1 Day C. 15 D. CON-VEH FZP-VEH mepsc Amplitude (-pa) 10 mepsc Amplitude (-pa) CON-NIM FZP-NIM % Open Arm Time % Open Arm Time

93 Fig. 4 Effects of systemic nimodipine injection on AMPAR-mediated mepscs in hippocampal CA1 neurons and anxiety-like behavior during FZP withdrawal. Rats injected with the L- type voltage gated calcium receptor antagonist, nimodipine (10 mg/kg, i.p.) immediately after FZP withdrawal were compared to those after vehicle injection (0.5% Tween-80, 2 ml/kg). The elevated plus-maze test was performed 1 day after nimodipine injection, followed 30 minutes later by hippocampal slice preparation and electrophysiological recording following the same protocol. A. Open arm time in the elevated plus-maze test in 1-day FZP-withdrawn rats, 1 day following vehicle or nimodipine injection. FZPwithdrawn rats (white bars, n=9) injected with vehicle showed evidence of anxiety-like behavior, measured as a significant reduction (p<0.05) in the percentage of open arm time compared to control rats (black bar, n=10). There were no differences in percent open arm time between control (n=8) and FZP-withdrawn rats (n=9) after prior nimodipine injection. B. Prior nimodipine injection averted the upregulation of AMPA receptor function 1 day after FZP withdrawal. The average amplitudes of AMPA receptor mediated mepscs (V H =-80mV) in CA1 neurons of control rats (black bar) and FZPwithdrawn rats (clear bar) were compared. The significant increase of mepsc amplitude (*p<0.05) in CA1 neurons from FZP withdrawn (n=8 cells) compared control rats (n=8 cells) was replicated following vehicle injection. However, after prior nimodipine injection there was no difference in mepsc amplitude in CA1 neurons from control rats (n=7 cells) vs. FZP-withdrawn rats (n=9 cells). C. Relationship between AMPARmediate mepsc amplitude in CA1 neurons and percent open arm time in individual rat that received vehicle injection. A significant positive correlation (R 2 =0.4807, *P < 0.005) 87

94 was found between mepsc amplitude and percent open arm time. D. Prior systemic nimodipine injection reversed the positive correlation between AMPAR-mediated mepsc amplitude in CA1 neurons and open arm time percentage in the elevated plusmaze test (R 2 =0.1985, p=0.0837). 88

95 mepsc Amplitude (-pa) 12 VEH NIMO Baseline [nimodipine] (μm) 89

96 Fig 5. Nimodipine has no direct effects on AMPAR-mediated mepscs. Mean concentrationresponse profile for the effect of nimodipine on CA1 neuron AMPAR-mediated mepscs. Average AMPAR mepsc amplitude (V H = -80 mv) during vehicle (close circles) or nimodipine (open circles) superfusion of hippocampal slices. Baseline mepsc amplitude was recorded for 5 minutes without nimodipine. In comparison to vehicle superfusion (n=3), nimodipine (n=4) at concentrations ranging from 0.1 µm to 100 µm had no effect on mepsc amplitude (p>0.05). There were no changes of mepsc frequency, decay, or rise time during nimodipine perfusion. 90

97 91

98 Fig. 6 Proposed cellular mechanisms in hippocampal CA1 neurons associated with FZPwithdrawal anxiety. Chronic FZP treatment results in prolonged GABA A receptor activation and bicarbonate-driven Cl - entry and accumulation that lead to GABA A receptor-mediated depolarization. This depolarization may activate L-type VGCCs and increase Ca 2+ entry. A downstream Ca 2+ -mediated signaling cascade could facilitate an increase in AMPAR-mediated glutamatergic strength via an increase in AMPAR channel number related to an increase in exocytosis or gene expression, or through an alteration in AMPAR subunit composition and/or phosphorylation state, similar to other activitydriven synaptic plasticity phenomena. Enhanced AMPAR function would further promote CA1 neuron membrane depolarization and serve as the major driving force contributing to CA1 neuron hyperexcitability, thus modulation of hippocampal output and the expression of anxiety-like behavior. Enhanced AMPAR-mediated function will also leads to NMDAR activation. The subsequent downregulation of NMDAR function, may serve as a negative feedback mechanism to regulate expression of benzodiazepine withdrawal-anxiety in response to the increased AMPAR-mediated glutamatergic strength. 92

99 Manuscript II Enhanced calcium entry through L-type voltage-gated calcium channels contributes to AMPA receptor plasticity in hippocampal CA1 neurons associated with chronic benzodiazepine administration. Kun Xiang 1,4, David R. Giovannucci 2,4, L. John Greenfield, Jr. 3,4, and Elizabeth I. Tietz 1,4 Department of Physiology and Pharmacology 1, Dept. of Neurosciences 2, Dept. of Neurology 3, and the Cellular and Molecular Neurobiology Program 4, College of Medicine, University of Toledo Health Science Campus, Toledo, OH,

100 ABSTRACT Benzodiazepines (BZs) are commonly prescribed hypnotic and anxiolytic drugs. However, tolerance and dependence limit their usefulness during prolonged administration. In the current study, the role of voltage-gated calcium channels (VGCCs) was investigated since growing evidence for their involvement in drug dependence. Highvoltage activated (HVA) calcium currents were recorded by whole cell voltage clamp and the voltage-current relationship was compared in dissociated hippocampal CA1 neurons immediately after 1-week flurazepam (FZP) treatment. Increased calcium entry through L-type VGCC following FZP treatment was indicated in the voltage-current relationship by the increased calcium current amplitude and density. A significant negative shift of the half-maximal potential of activation (Boltzmann curve) was demonstrated from FZPtreated rats ( ±3.286 mv) compared to control rats ( ±2.513 mv), while the slope of activation remained unaffected. Steady-state inactivation of the HVA current remained unchanged. Previously we reported upregulation of CA1 neuron AMPA receptor (AMPAR) function, represented by significantly increased amplitude of AMPAR-mediated miniature excitatory postsynaptic current (mepsc), 2 days after 1- week flurazepam (FZP) treatment. Corresponding investigation of the temporal pattern of L-type VGCC functional upregulation over this time period was performed. Increased calcium current amplitude and significant negative shift of the half-maximal potential of activation was also observed ( ±2.425 mv from FZP treatment; ±1.779 mv from control group) but not the steady-state inactivation. Since this enhanced calcium entry precedes and is sustained during the period of AMPAR plasticity, we hypothesized that it contributes to the AMPAR plasticity. Prior systemic injection of nimodipine, an L- 94

101 type VGCC antagonist (10 mg/kg), reversed the increased amplitude of AMPARmediated mepscs in 2-day FZP-withdrawn rats, suggesting an altered calcium conductance through L-type VGCC may contribute to AMPAR plasticity. FZP and diazepam have direct concentration- (as low as 1µM) and use-dependent inhibitions of L- type VGCC-dependent Ca 2+ current, pointing to a potential mechanism of modulation of L-type VGCC function following chronic BZ administration. Taken together, these studies indicate chronic FZP treatment modifies the function of L-type VGCC in hippocampal CA1 neurons. An enhanced calcium entry through L-type VGCC contributes to AMPAR function unregulation and may serve as the potential underlying mechanism for BZ tolerance and dependence. 95

102 INTRODUCTION Benzodiazepines, a group of positive allosteric modulators of γ-amino butyric acid type A receptors (GABARs), are widely prescribed for the treatment of anxiety disorders and insomnia. However, widespread legitimate prescription of benzodiazepines increases the proportion of patients who develop dependence. Reducing the dose or abrupt discontinuation of treatment may result in withdrawal symptoms, such as anxiety, agitation, sleep disturbances and seizures or delirium after high-dose use (for reviews see Bateson, 2002; Griffiths and Johnson, 2005). Thus, understanding the mechanisms underlying physical dependence may improve benzodiazepine s clinical usefulness, as well as reduce benzodiazepine misuse. We previously demonstrated that a transient enhancement of α-amino-3-hydroxy- 5-methyl-4-isozaxolepropionic acid receptor (AMPAR) function in rat CA1 pyramidal neurons during benzodiazepine withdrawal is tightly coupled to increased open arm time in an elevated plus maze (Van Sickle et al, 2004, Xiang and Tietz, 2007), which suggested that hippocampal AMPAR plasticity may contribute to benzodiazepine withdrawal anxiety. Recent studies suggest that administration of L-type voltage-gated calcium channel (VGCC) antagonists can interrupt withdrawal symptoms after ending benzodiazepine treatment (Gupta et al, 1996; Podhorna, 2002; EI Ganouni et al, 2004, Cui et al, 2006). Indeed, as reported in other forms of neuronal plasticity, acute systemic injection of an L-type voltage-gated calcium channel (VGCC) antagonist, nimodipine, averted hippocampal CA1 neuron AMPAR current enhancement and anxiety-like behavior in benzodiazepine-withdrawn rats (Xiang and Tietz, 2007), pointing to a role of L-type VGCCs in mediating benzodiazepine dependence. 96

103 L-type VGCCs have a central role in neuronal function converting electrical activity into neurochemical events, and have been implicated in mediating activitydependent synaptic plasticity (Berridge, 1998; Morgan and Teyler, 1999; Rajadhyksha and Kosofsky, 2005). Moreover, L-type VGCCs have been reported to regulate calcium signaling associated with synaptic plasticity related to dependence on several psychostimulants. (Kelley 2004; Rajadhyksha and Kosofsky, 2005) and also contribute to withdrawal symptoms associated with other GABA A receptor positive allosteric modulators, including ethanol and barbiturates (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999; Watson and Little, 2002). Moreover, studies have also shown that benzodiazepines can directly inhibit VGCCmediated Ca 2+ flux (Taft and DeLorenzo, 1984; Gershon, 1992; Reuveny et al, 1993; Ishizawa Y et al, 1997). Together these observations raise the possibility that chronic benzodiazepine administration might modulate L-type VGCC activity and downstream intracellular Ca 2+ -dependent signaling mechanisms, and thus glutamatergic synaptic plasticity associated with benzodiazepine dependence. We postulated that chronic BZ administration modifies L-type VGCC function underlying AMPAR plasticity in CA1 pyramidal neurons in benzodiazepines-withdrawn rats. The properties of L-type VGCC-dependent Ca 2+ currents were investigated using whole-cell voltage-clamp methods in isolated CA1 pyramidal neurons. The temporal pattern of changes in CA1 neuron Ca 2+ current density was evaluated following acute and chronic administration of the benzodiazepine, flurazepam. In addition, we examined if nimodipine preinjection would affect enhanced AMPAR-mediated mepsc amplitude in 2-day FZP withdrawn as reported previously (Xiang and Tietz, 2007). Finally, the direct 97

104 concentration- and use-dependent effect of the benzodiazepines to affect VGCC binding and function was evaluated. Collectively, the findings suggest that chronic FZP administration transiently enhances calcium entry through L-type VGCCs during FZP withdrawal with a time-course consistent with the possibility that a downstream VGCCmediated intracellular calcium signaling cascade may contribute to AMPAR plasticity and benzodiazepine withdrawal anxiety. METHODS MATERIALS AND METHODS Experimental protocols involving the use of vertebrate animals were approved by the University of Toledo, College of Medicine (formerly the Medical University of Ohio) Institutional Animal Care and Use Committee (IACUC) and conformed to National Institutes of Health guidelines. In Vivo Drug Treatments Chronic Flurazepam Administration. Oral FZP treatment of rats was as previously described (Van Sickle et al, 2004). In short, male Sprague-Dawley rats (initial age PN22-25, Harlan, Indianapolis, IN) were offered only saccharin solution for a 2 to 4 day adaptation period. FZP (flurazepam dihydrochloride ph 5.8) was then offered in 0.02% saccharin solution as their only source of drinking water for 1 week. The concentration of FZP was adjusted daily according to each rat s body weight and fluid consumption (100 mg/kg X 3 days and 150 mg/kg X 4 days). Only rats that consumed a criterion dose of an average 120 mg/kg/day were accepted for study. At the end of drug administration, saccharin water was again provided for 0, 2 or 4 days prior to hippocampal slice 98

105 preparation on PN For 3-day FZP treatment, rats were given oral FZP (100mg/kg X 3 days) in saccharin solution. Pair-handled control rats receive saccharin water for the same length of time. Unlike in humans, brain levels of residual FZP and metabolites, which are equivalent to that of other common chronic benzodiazepine treatments (Gallager et al, 1985), rapidly decline over the first 24 hours after drug removal and are no longer detectable in hippocampus in 2-day FZP-withdrawn rats (Xie and Tietz, 1992). Acute desalkyl-fzp administration. To determine whether changes in hippocampal slices are specific to chronic BZ treatment, an acute dose of the primary FZP active metabolite, desalkyl-fzp, were given to another group of rats. Desalkyl-FZP (2.5 mg/kg, p.o.) was administered by gavage in an emulsion of peanut oil, water and acacia (4:2:1) as previously described (Van Sickle and Tietz, 2002). Rats were euthanized 30 min later for hippocampal slice preparation. Control rats received an equivalent volume of emulsion vehicle. The single dose of desalkyl-fzp was previously demonstrated by radioreceptor assay to result in levels of benzodiazepine activity (0.57 µm diazepam (DZP) equivalents) similar to that found in hippocampi of rats after 1-week FZP treatment without a concomitant effect on GABAR-mediated inhibition (Xie and Tietz, 1992). Systemic Antagonist Injection. Control and FZP-treated rats were given a single intraperitoneal injection of nimodipine (10 mg/kg, i.p.), a L-type VGCC antagonist, or 2 ml/kg of vehicle, (0.5 % Tween 80) 1 day after ending 1-week FZP treatment and 24 hours prior to hippocampal slice preparation. 99

106 Electrophysiology Acutely isolated hippocampal CA1 pyramidal neurons: CA1 pyramidal neurons were acutely isolated using modifications of the methods described previously (Van Sickle et al, 2002). Briefly, hippocampal slices, 400 μm were prepared on a vibratome in ice-cold, pre-gassed (95% O 2 /5% CO 2 ) buffer containing (in mm): NaCl, 120; KCl, 2.5; CaCl 2, 0.1; MgCl 2 4; PIPES 20; D-glucose, 25; ph 7.4. Slices were maintained at room temperature for two hours then incubated for 40 min at 37ºC in the following buffer (mm): protease XIV (Sigma-Aldrich) 1.4 units/ml, NaCl, 120; KCl, 2.5; CaCl 2 1.5; MgCl 2 4; PIPES 20; D-glucose, 25; ph 7.4. Slices were washed for 5 min with 20 ml of a 1 mg/ml bovine serum albumin (BSA) solution. The hippocampal CA1 region was microdissected on ice, notched and triturated using a 25 gauge, then 30 gauge bore Pasteur pipette in 100 µl of ice cold PIPES. Fifty mls of the cell suspension was plated on each of two poly-lysine-coated (2mg/ml, poly-d-lysine; 2 mg/ml, poly-dl-lysine) culture dishes for minutes prior to recording. Hippocampal slice preparation. For mepsc recording, transverse dorsal hippocampal slices (400 μm) were prepared on a vibratome (Ted Pella, Inc., Redding, CA) as previously described (Xiang and Tietz, 2007) in ice-cold, pre-gassed (95%O 2 /5% CO 2 ) artificial cerebrospinal fluid (ACSF) containing (in mm): NaCl, 120; KCl, 2.5; CaCl 2, 0.5; MgSO 4 7.0; NaH 2 PO 4 1.2; NaHCO 3, 2; D-glucose, 20; Ascorbate, 1.3, ph 7.4. Slices were maintained at room temperature (RT) for 15 min in gassed, low-calcium, highmagnesium ACSF, then transferred to normal ACSF containing (in mm): NaCl, 119; KCl, 2.5; CaCl 2, 1.8; MgSO 4 1.3; NaH 2 PO ; NaHCO 3, 26; D-glucose, 10; ph

107 Slices were maintained at room temperature for 1 hr in ACSF. During recording, slices were perfused at 2.5 ml/min with gassed ACSF at room temperature. Hippocampal neuron cell cultures. To establish whether the reported effects of benzodiazepines on VGCC function in hippocampal neurons the concentration-dependent effects of FZP and DZP on Ca 2+ currents was evaluated in a limited number of hippocampal culture cells. Briefly, Hippocampal tissue was dissected from E-18 Sprague-Dawley rats, collected in 1X HBSS (hanks balanced salt solution) (Gibco, Cat # 14180), and incubated for 20 min in 0.25% trypsin-edta (Cellgro, Cat # CL) at 37 C with 5% CO 2. The trypsin solution was aspirated and tissue washed 3X with 1X HBSS. The 1X HBSS solution was replaced by medium containing Modified Eagle s medium (MEM, Cellgro, Cat# CV) with L-glutamine,10 % FBS (Invitrogen, Cat # ), 6 mg/ml glucose and 1X penicillin/streptomycin (Invitrogen, Cat# ) and tissue was triturated using flame-polished Pasteur pipettes. Dissociated cells were plated onto coverslips coated with 0.2 mg/ml poly-dl-ornithine (Sigma, Cat # O- 2250) at a density of 3X 10 6 cells/ml medium. The mixed glial and neuronal cultures were maintained in vitro for days and electrophysiological recordings were performed using neurons which displayed pyramidal cell morphology. AMPAR-mediated mepsc recording. AMPAR-mediated mepscs were isolated from CA1 neurons in ACSF plus 1 mm TTX (Tetrodotoxin), 50 mm picrotoxin and 25 mm CGP using whole-cell voltage-clamp techniques (Xiang and Tietz, 2007). Patch pipettes (3-6 MΩ) were filled with internal solution containing (in mm): Csmethanesulfonate, 132.5; CsCl, 17.5; HEPES, 10; EGTA, 0.2; NaCl, 8; Mg-ATP, 2; Na3- GTP, 0.3; QX-314, 2; ph 7.2 adjusted with CsOH. Resting membrane potential (RMP) 101

108 was measured immediately upon cell break-in. Neurons were voltage-clamped (V H = -80 mv) in continuous mode (csevc) using an Axoclamp 2A amplifier (Axon Instr., Union City, CA). mepsc activity was recorded for 5 min and analyzed with MiniAnalysis software (Synaptosoft Inc., Leonia, NJ) as previously described (Xiang and Tietz, 2007). L-type VGCC-dependent Ca 2+ current recording: Currents were recorded under whole-cell voltage-clamp conditions at room temperature using patch pipettes of 3-6 MΩ resistance. Neurons with a bright and smooth appearance, pyramidal shape with at least one moderate to large apical dendrite and were selected for recording. Presumptive interneurons, i.e. very large pyramidal neurons or bipolar neurons, as well as elliptical shaped neurons were excluded. The external solution contained (in mm): NaCl 110; HEPES 10; TEA chloride 25; KCl 5.4; CaCl 2 5; 4-AP 5; MgCl 2 1; D-glucose 25; TTX 1µM ph 7.4. The electrode solution contained (in mm): CsF 110; TEA chloride 25; phosphocreatine 20; phosphocreatine kinase 50 units/ml; EGTA 10; HEPES 10; NaCl 5; MgCl 2 2; CaCl 2 0.5; BaCl 2 0.5; MgATP 2; NaATP 0.1; PH 7.3. Currents were recorded with an Axoclamp 200B amplifier (Axon Instr.) using a Digidata 1200B AD/DA converter and pclamp 9.2 acquisition software (Axon Instr.). After establishing the whole-cell configuration, cells were allowed to stabilize for 10 min before current recording protocols were initiated. Neurons were voltage clamped at V H = -65 mv. Peak calcium currents (pa) were activated using 200 ms voltage steps to voltage levels between -70 and +40 mv in 10 mv increments, preceded by a 3 s hyperpolarization pre-pulse to -80 mv. The steady-state inactivation of calcium currents was evaluated with a 200 ms test pulse to 10 mv, preceded by a 1500 ms conditioning pre-pulse ranging from -80 mv to 10 mv, in 10 mv increments. Currents were corrected 102

109 for linear, non-specific leak currents and capacitive transients. Series resistance was compensated (80-90%). Current density was calculated as a function of cell membrane capacitance (pf), an estimate cell size. Ca 2+ current analysis: Whole-cell calcium currents were analyzed off-line using Clampfit 9.2 software (Axon Instr.). Current density (pa/pf) was defined as peak current amplitude divided by membrane capacitance and was plotted as a function of membrane potential (V). Activation curves were constructed using individual conductance values for each cell using the equation G=I/(Vt-Vr), where Vt was the test voltage, Vr was the cell reversal potential and I, the measured current, then normalizing the data to G max, the maximum conductance. Inactivation curves were constructed by normalizing the current values to the largest value. The data points for the conductance G were fitted with a Boltzmann equation: G/G max = 1/ {1+exp[(V-V h )/ k]}, where G max was the maximum Ca 2+ conductance, V h was the potential where G was the half-maximal G max and k was a factor proportional to the slope at V h. Data points were normalized to the maximal conductance, averaged and plotted. Inactivation curves were constructed by normalizing the current values to the largest value then fitted with the Boltzmann equation I (V) / I max=1/{1+exp[(v-v h )/k]}. Concentration- and use-dependent inhibition of Ca 2+ currents by FZP: The concentration-dependent effect of FZP on L-type VGCC currents was assessed in DIV (day in vitro) hippocampal cultured cells using whole-cell techniques as described above. After a 200 ms baseline test pulse (V Hold -80 to +10 mv), trains of eight equivalent depolarizing pulses were applied at 1 min inter-train intervals to allow complete recovery from calcium dependent inactivation. Trains were applied at 103

110 frequencies of 1, 2, 3 and 4 Hz followed within 10 ms by another test pulse. A recovery test pulse, without prior train stimulation was given at the end of the recording to exclude run-down of Ca 2+ channel function or seal degradation. Benzodiazepine ( μm) effects to inhibit Ca 2+ current test pulses were evaluated after 5 min preincubation during the train (FZP and DZP) or no train condition (FZP). The effect of increasing concentrations of FZP to inhibit Ca 2+ currents in acutely isolated CA1 neurons was also examined. Statistics: Data are reported as mean ± standard error of the mean (SEM). Data were analyzed by MANOVA with post hoc comparison of means by the method of Scheffé. Asterisks denote significant differences between control and FZP-treated groups, p<0.05. Drug solutions: Drugs used for superfusion during whole-cell recording were dissolved at 100 times their final concentration and added to the perfusate with a syringe pump (Razel; World Precision Instr., Inc., Sarasota, FL) at a rate of 25 to 75 μl/min to achieve their final concentration. Nimodipine was dissolved in 0.5 % Tween-80 solution and kept in a light-tight vial. All other drugs were dissolved in dh 2 O. QX-314 (lidocaine N-ethyl bromide quaternary salt), picrotoxin, APV, TEA (tetraethylammonium chloride), 4-AP (4-aminopyridine), Nimodipine were all from Sigma-Aldrich Chemical Co. (St Louis, MO). Tetrodoxin (TTX) was obtained from Alamone Laboratories (Jerusalem, Israel). CGP was purchased from Tocris Bioscience (Ellisville, MO). FZP dihydrochloride was supplied by the National Institutes of Health Drug Supply Program. 104

111 RESULTS L-type VGCC-dependent Ca 2+ currents in acutely isolated hippocampal CA1 pyramidal neurons. To study L-type Ca 2+ channel function following chronic FZP administration, Ca 2+ currents were recorded in acutely isolated hippocampal CA1 pyramidal neurons from rats after different lengths of FZP treatment and withdrawal. To isolate calcium currents, potassium currents were blocked with TEA, 4-AP and Cs + and sodium currents with TTX. Figure 1 shows representative Ca 2+ current traces recorded in CA1 neurons from 2-day FZP-withdrawn rats as described in the methods. Membrane holding potential (V H ) was -65 mv. Activation pulses were composed of 200 ms depolarizing voltage steps ranging from -70 to 40 mv in 10 mv increments, preceded by a 3 s pre-pulse to -80mV ( Fig 1A). Elicited currents were pharmacologically confirmed as predominantly L-type Ca 2+ currents by inhibition with verapamil (40 µm), a selective, water-soluble, L-type VGCC antagonist (Fig 1B). Figures 1C and 1D show representative calcium current traces elicited in neurons isolated from a control (CON) and a 2-day FZP-withdrawn rat. Temporal regulations of L-type VGCC function. Previously we described a transient AMPAR synaptic plasticity in hippocampal CA1 neurons after chronic FZP treatment (Van Sickle and Tietz 2002; Van Sickle et al, 2004; Xiang and Tietz, 2007). To evaluate if chronic FZP administration would modify L-type VGCC function and serve as a Ca 2+ signaling source for AMPAR plasticity, L- type VGCC-dependent Ca 2+ currents were recorded immediately (0 days) after ending 3- day FZP administration, or 2 or 4 days after 7-day FZP administration. The latter time period overlaps the window of AMPAR plasticity previously described (Van Sickle et al, 105

112 2004). Ca 2+ currents were normalized by the membrane capacitance of individual neurons. Current density (pa/pf) was plotted versus membrane potential. Current/voltage (IV) plots and peak current density in control neurons was similar to that previously described in dissociated CA1 neurons (Kortekaas and Wadman, 1997; Gorter et al., 2002) and was similar across all control groups (Figs 2A-D). To examine if the enhancement of Ca 2+ current occurred during chronic FZP administration, rats were treated with FZP for 3 days (Fig 2A). There was no significant increase in Ca 2+ current density in CA1 neurons isolated from rats immediately after (0-day) 3 day FZP treatment (open circles, n=9. p>0.05) compared to that from matched control rats (n=9), although there was a tendency toward enhancement (~1.3X) of peak current density. There was a significant increase (~1.8X) in Ca 2+ current density in CA1 neurons isolated from rats immediately after (0-day) withdrawal from 7-day FZP-treatment (n=12. *p<0.05) compared to that from matched control rats (n=10) (Fig 2B). An equivalent increase in Ca 2+ current density was found in neurons from 2-day FZP-withdrawn rats following 7- day FZP-treatment (n=25. *p<0.05) in comparison to control neurons (n=25) (Fig 2C). L-type Ca 2+ currents were also examined 4 days after FZP withdrawal, when increased AMPAR function is no longer apparent. There were no significant differences in Ca 2+ current density in neurons from 4-day FZP-withdrawn rats following 7 day FZPtreatment (n=8, p>0.05) in comparison to controls (n=7) (Fig 2D). To demonstrate that enhanced L-type VGCC function during FZP withdrawal period was specific to chronic benzodiazepine exposure, an acute dose of the FZP active metabolite, desalky-fzp (2.5 mg/kg, p.o.), was given to another group of rats, a dose which results in comparable benzodiazepine levels to that after 7 day FZP treatment (Van 106

113 Sickle et al., 2004). Control rats received only the emulsion vehicle. There were no significant differences in Ca 2+ current density in neurons from acute desalkylfzp treated rats (n=8. p>0.05) in comparison to controls (n=10) (data not shown). The temporal patterns of changes in L-type VGCC function in CA1 neurons following different periods of BZ treatment and withdrawal are summarized in Fig. 3. Peak Ca 2+ current density at V H = 0 mv is shown (* p<0.05). Voltage-dependence of Ca 2+ current activation is altered after FZP administration. Current amplitudes elicited by voltage steps from individual neurons were transformed to conductance, normalized to the maximal conductance and fitted with Boltzmann s equation as described in the methods. There was a significant negative shift of the activation curves derived from 0-day and 2-day FZP-withdrawn rats compared to neurons from matched control rats (Fig 4A). The voltages of half-maximal activation (V half ) were significantly decreased (Fig 4B) in neurons from 0-day FZP-withdrawn rats (-11.7 ± 1.2 mv, n=11) compared to their matched controls (-5.8 ± 2.0 mv, n=9; *p<0.05), and 2-day FZP-withdrawn rats (-13.6 ± 1.7 mv, n=25) compared to their controls (-4.2 ± 1.5 mv, n=25, *p<0.05). There was no significant difference between the slopes of the activation curves (dv) between any FZP-withdrawn group and their matched control group (Fig 4C. 0 day: FZP: 16.4 ± 0.8 mv; CON: 17.1 ± 1.4 mv, p> day: FZP: 16.1 ± 1.0 mv; CON: ± 0.93, p>0.05). There was no difference in voltage-dependence of Ca 2+ current activation immediately after acute desalkylfzp or 3 day FZP treatment, or 4 days after 7 day FZP treatment in comparison to controls (data not shown). Voltage dependence of Ca 2+ current steady-state inactivation remained unchanged. Voltage dependence of Ca 2+ current steady-state inactivation was measured by a 200 ms 107

114 test-pulse to 10 mv, preceded by a 1.5 s conditioning pre-pulse. Pre-pulse voltage steps ranged from -80 to 10 mv in 10-mV increments (Fig 5A). V Hold was -65 mv. Representative steady-state inactivation Ca 2+ current traces from CA1 neurons isolated from a control and a 2-day FZP-withdrawn rat are shown in Figs 5B and C. Boltzmann inactivation curves derived from individual fits of Ca 2+ currents elicited from neurons isolated from control and FZP-withdrawn rats are shown. There were no significant shifts in the inactivation curves from either 0 day or 2 day FZP-withdrawn neurons compared to matched control neurons (Fig 5D). The voltages of half-maximal inactivation (V half ) were not significantly different from that of FZP-withdrawn rats compared to their matched control rats (Fig 5E. 0 day: FZP: ± 2.0 mv, n= 11; CON: ± 1.16 mv, n=11. p> day: FZP: ± 1.3 mv; n=13; CON: ± 1.0 mv, n=17. p>0.05). There were also no significant differences between the slopes of the inactivation curves (dv) between any FZP-withdrawn group and their matched control group (Fig 5F. 0 day: FZP: 11.4 ± 0.3 mv, n=11; CON: 10.9 ± 0.2 mv, n=11, p> days: FZP: 10.2 ± 0.5 mv, n=13; CON: 11.8 ± 0.3, n=17, p>0.05). There was no difference in the voltagedependence of Ca 2+ current steady-state inactivation immediately after acute desalkylfzp or 3 day FZP treatment, or 4 days after 7 day FZP treatment in comparison to controls (data not shown). Prior nimodipine injection reverses the increased in AMPAR-mediated mepscs in neurons isolated from 2-day FZP-withdrawn rats. The temporal pattern of enhancement of L-type VGCC function overlaps the window of AMPAR plasticity and was postulated to regulate enhanced AMPAR function (Xiang and Tietz, 2007). It was previously demonstrated that the L-type VGCC antagonist 108

115 nimodipine (10 mg/kg, i.p.) injected immediately after FZP withdrawal could reverse the increase of AMPAR-mediated mepsc amplitude and the associated FZP-withdrawal anxiety (Xiang and Tietz, 2007). This effect is not through the direct interaction between nimodipine and AMPAR, as nimodipine has no direct effects on AMPAR-mediated mepscs (Xiang and Tietz, 2007). Since the increase in AMPAR-mediated mepsc amplitude also appears in 2-day FZP withdrawn rats, rats were injected systemically with nimodipine (10 mg/kg, i.p.) or vehicle (0.5% Tween-80, 2 ml/kg, i.p.) 1 day after FZP removal. Hippocampal slices were isolated from 2-day FZP-withdrawn rats the following day and AMPAR-mediated mepscs were evaluated. Fig 6A shows representative mepsc current traces isolated from neurons in 2-day FZP-withdrawn (FZP) and control (CON) rat injected with nimodipine (NIM) or vehicle (VEH). Average amplitudes of AMPAR-mediated mepscs (V H = -80 mv) in CA1 neurons isolated from control rats and 2-day FZP-withdrawn rats following either vehicle or nimodipine injection were compared. Fig. 6B shows that prior vehicle injection again had no effect on basal AMPAR mepsc amplitude in control neurons or their enhancement during FZP withdrawal (CON/VEH: 8.5 ± 1.4 pa, n=8; FZP/VEH: 11.2 ±0.9 pa, n = 8. *p<0.05). However, following nimodipine injection of 1-day FZP-withdrawn rats the mean AMPAR mepsc amplitude returned to control levels in CA1 neurons from 2-day FZPwithdrawn rats (CON/NIMO: 9.26 ± 1.5 pa, n=7; FZP/NIMO: 9.0 ±1.6, n=7. p> 0.05). As reported previously (Van Sickle et al., 2004, Xiang and Tietz, 2007), there were no differences in resting membrane potential (CON/VEH: ± 2.1 mv, FZP/VEH: ±1.8 mv; CON/NIM: ± 1.7 mv, FZP/NIM: ± 1.9 mv, p>0.05), current decay (CON/VEH: 16.7 ± 2.8 ms; FZP/VEH: 15.6 ± 1.8 ms; CON/NIMO: 16.3 ±1.5 ms; 109

116 FZP/NIMO: 17.1 ± 2.1 ms, p>0.05) or in mepsc frequency (CON/VEH: 0.26 ± 0.04 Hz; FZP/VEH: 0.30 ± 0.07 Hz; CON/NIMO: 0.27 ± 0.06 Hz; FZP/NIMO: 0.27 ± 0.05 Hz. p>0.05) in neurons from control vs. FZP-withdrawn rats. Concentration- and Use-dependent inhibition of L-type VGCC-dependent Ca 2+ current by FZP and DZP. The concentration- and use- dependent effects of FZP on L-type VGCC currents were evaluated in DIV hippocampal cultured cells using whole-cell techniques. After a 200-mS baseline test pulse (V Hold -80 to +10 mv), trains of eight equivalent depolarizing pulses, was applied at frequencies of 1, 2, 3 and 4 Hz followed within 10 ms by another test pulse. Each train was applied at 1 min inter-train interval to allow complete recovery from calcium-dependent inactivation. A recovery test pulse, without prior train stimulation was given at the end of the recording to exclude run-down of Ca 2+ channel function or seal degradation. FZP (1 μm) showed few inhibitory effects on L-type VGCC-dependent Ca 2+ current without train stimulation (Fig. 7A). However, with the train consecutive depolarizing pulses, which allow the channel to be at open state, there was ~ 45% inhibition of L-type Ca 2+ current density by 1 μm FZP compared to only the presence of vehicle. A recovery test pulse, without prior train stimulation was given at the end of the recording to exclude run-down of Ca 2+ channel function or seal degradation (Fig. 7A). Concentration-dependent effects of benzodiazepines on L-type Ca 2+ current were also investigated. Benzodiazepine ( μm) effects to inhibit Ca 2+ current test pulses were evaluated after 5 min pre-incubation during the train (FZP and DZP) or no train condition (FZP). As shown in Fig. 7B, FZP ( μm) had a concentrationdependent inhibition effects on L-type Ca 2+ current and 100 μm FZP completely blocked 110

117 L-type Ca 2+ current. DZP (1 and 10 μm) had a similar, but less potent, inhibitory effects on L-type Ca 2+ current (Fig 7B). Discussion A transient enhancement of L-type VGCC function in hippocampal pyramidal CA1 neurons, was represented by an increase in Ca 2+ current density during withdrawal from chronic benzodiazepine administration. The properties of L-type VGCC-dependent Ca 2+ currents were investigated using whole-cell voltage-clamp techniques in acutely isolated CA1 pyramidal neurons (Fig 1). An emergent increase in Ca 2+ current density was apparent even following relatively short periods (3 days) of FZP treatment (Fig 2A). Ca 2+ current density progressively increased until a significant, ~2-fold increase was evident upon cessation of 7-day FZP treatment (Fig 2B) which persisted at least 2 days, but not 4 days, after FZP withdrawal (Fig 2C and D). There was a negative shift in the voltage dependence of Ca 2+ channel activation in 0-day and 2-day FZP-withdrawn rats (Fig 4) without a change in the voltage dependence of steady-state inactivation (Fig 5). The temporal pattern of enhanced L-type VGCC function (Fig 3), preceded and overlapped the transient increase in AMPAR-current potentiation in CA1 neurons described previously (Van Sickle et al, 2004). Systemic nimodipine administration a day prior to hippocampal slice recording prevented the enhancement of AMPAR-mediated mepsc amplitude (Fig 6), further supporting a role for L-type VGCCs in regulating increased glutamatergic strength during FZP withdrawal (Xiang and Tietz, 2007). Moreover, both FZP and diazepam showed a use- and concentration-dependent inhibition 111

118 of L-type Ca 2+ current, indicating a potential modulation of calcium channel function by chronic benzodiazepine administration. L-type VGCCs are virtually ubiquitous in the CNS and radiolabelled antagonists intensely label binding sites in the hippocampus, amygadala and cortex (Gould et al., 1985; Hobom et al., 2000). Numerous investigations have suggested that disturbances of neuronal Ca 2+ homeostasis including enhanced Ca 2+ entry associated with upregulation of L-type VGCCs, is involved in ethanol, barbiturate, morphine and nicotine dependence (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999, Katsura et al., 2002). Our findings that chronic, but not acute, benzodiazepine administration enhances L-type VGCC-mediated Ca 2+ currents suggest that a similar effect may occur in hippocampal CA1 pyramidal neurons associated with benzodiazepine dependence (Fig 2). More importantly, the temporal pattern of modifications of L-type VGCC function overlapped the window of AMPAR synaptic plasticity, previously demonstrated to correlate with benzodiazepine withdrawal-anxiety in rats. This finding also strengthens the role of L-type VGCC-mediated Ca 2+ influx in regulating some forms of activity dependent plasticity. While the mechanism underlying the up-regulation of L-type VGCC function remains to be clearly elucidated, our studies suggest that an increase in the expression of the α1d subunit of L-type VGCCs, is at least in part responsible for the enhanced VGCC function in CA1 neurons (Fig 4). VGCCs are composed of up to four distinct subunits (α1, β, α2-δ, γ) and two α1 subunits (α1c and α1d) have been identified in the brain (Catterall, 2000). In neurons, α1c and α1d are often found in the same general neuronal compartments, particularly dendrites, although their subcellular distributions appear 112

119 distinct (Hell et al., 1993; Westenbroek et al., 1998). α1d subtype L-VGCCs open at relatively hyperpolarized membrane potentials, activated at -55 mv, approximately mv more hyperpolarized as compared with α1c subtype L-VGCCs, and mediate subthreshold calcium signaling (Xu and Lipscombe, 2001; Lipscombe et al., 2004). α1d subunit protein levels increase in hippocampal CA1 neurons of aged rats (Vern et al., 2003). In single CA1 neurons, increases in α1d subunit transcript levels were positively correlated with increased VGCC activity (Chen et al., 2000). They also mediate consolidation, but not extinction, of contextually conditioned fear in mice (McKinney and Murphy, 2006). Thus, the increase in α1d subunit protein in CA1 mini-slices might well explain the observation of a negative shift in voltage dependence of Ca 2+ current activation and may be associated with increased current density (Fig 4). It would be of significant interest to investigate the expression of α1d subunit protein in hippocampal CA1 region following chronic FZP administration in rats. Increased expression of α1d subunit-containing VGCCs may also have significant physiological importance following chronic benzodiazepine administration. As a consequence of prolonged GABA A receptor activation during FZP administration numerous time-dependent changes occur at the GABA A receptor, some of which measurably influence CA1 neuron hyperexcitability (Van Sickle et al., 2004) including a bicarbonate-driven Cl - accumulation reflected in a shift in the Cl - reversal potential and the appearance of a bicuculline-sensitive depolarizing potential (Zeng et al, 1995; Zeng and Tietz, 1996; Zeng and Tietz, 2000). Moreover, spectral analysis of hippocampal electrical activity during withdrawal from 1-week FZP reveals increased power of a 7 Hz (theta) peak (Poisbeau et al., 1997) that has been suggested to be associated with 113

120 GABAergic post-synaptic depolarization and a shift of reversal potential from Cl - toward HCO - 3 (Sun et al., 2001). Thus lower threshold VGCCs would be more likely activated by these fast, sub-threshold depolarizing driving forces and trigger sustained Ca 2+ influx. In fact, several studies have shown an increased accumulation of intracellular calcium through L-type VGCCs following prolonged GABA A receptor activation that leads to bicarbonate-driven Cl - entry and Cl - accumulation (Reichling et al, 1994; Lyons et al., 2001; Chavas et al, 2004; Marty and Liano, 2006). In addition to their well-characterized effects at GABA A receptors, benzodiazepines can also inhibit neuronal VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Reuveny et al, 1993; Watabe et al., 1993; Ishizawa et al, 1997). Based on the findings that micromolar benzodiazepine binding sites were shown to mediate inhibition of depolarization-dependent Ca 2+ uptake in synaptosomes by a series of benzodiazepines, Taft and colleagues proposed that the concentrations of diazepam achieved during intravenous administration for status epileticus (>10 μm) was sufficient to block L-type VGCCs (Taft and DeLorenzo, 1984). However, nifidipine was shown to block the hypnotic, but not anticonvulsant or anxiolytic actions of flurazepam in rats (Mendelson et al., 1984). Together these findings suggest that at least some benzodiazepines clinical actions may be mediated by L-VGCCs. Nonetheless, the potency of FZP to inhibit Ca 2+ channel currents in hippocampal cultures was significantly greater when Ca 2+ channels were activated by a prior depolarizing train (Fig. 7B) suggesting that benzodiazepine inhibition of VGCC-mediated currents is use-dependent, as previously shown with chlordiazepoxide (Reuveny et al, 1993). However, the capacity of chronic flurazepam treatment to enhance VGCC current amplitude and a1d subunit 114

121 level, is in direct contrast to the ability of prolonged nimodipine administration to decrease a1d subunit levels in aged rats (Veng et al., 2003). Thus, the low micromolar benzodiazepine brain levels attained during chronic treatment would be expected to have their primary action on GABA A receptors to enhance CA1 neuron L-type VGCC activity by enhancing GABA A receptor-mediated depolarization or be driven by emergent increases in theta activity. On the other hand, the greater potency of FZP than DZP to block VGCC-mediated Ca 2+ currents raises the prospect that FZP may also have a selflimiting protective effect to modulate the increase in L-type VGCC activity related to its action on the GABA A receptor, and thus the associated withdrawal symptoms. This possibility may, at least in part, provide an explanation for the lack of withdrawal seizures noted following chronic FZP, as opposed to chronic DZP treatment (Rosenberg et al., 1991; Ramsey-Williams et al., 1994; Rosenberg 1995). Oral flurazepam treatment and other common diazepam treatments result in rat brain levels of benzodiazepine metabolites in the low micromolar range (0.6 μm in diazepam equivalents) (Gallager et al, 1985; Xie and Tietz, 1992). L-type VGCCs have been reported to play role in various activity-dependent models of neuronal plasticity (Morgan and Teyler, 1999; Borroni et al., 2000; Rajadhyksha and Kosofsky, 2005). Activation of neuronal L-type VGCCs triggers a sustained influx of Ca 2+ upon depolarization which, via diverse soluble messengers and transcription factors, initiates long-term processes related to synaptic plasticity in the hippocampus and amygdala (Dolmetsch et al., 2001) In fact, during ethanol withdrawal and following psychostimulant administration, activation of L-type VGCCs can result in the influx of intracellular calcium and downstream activation of Ca 2+ /CaM-activated 115

122 kinase and phosphatase pathways, eventually involving CREB-induced gene expression, also important to neuronal activity and experience-dependent plasticity (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005). The finding that prior systemic injection of the L-type VGCC antagonist, nimodipine prevented the upregulation of CA1 neuron AMPAR function in 2-day FZP-withdrawn rats (Fig 6) strengthens the previous finding that nimodipine prevents FZP-withdrawal anxiety in 1-day withdrawn rats by preventing the related enhancement of CA1 neuron AMPAR currents. This effect is not through a direct interaction between the VGCC antagonist and AMPARs, as nimodipine had no direct effect on AMPAR-mediated mepscs (Xiang and Tietz, 2007). This finding may also explain the ability of L-type calcium channel blockers like nimodipine, nifedipine and verapamil to block a variety of BZ withdrawal signs including hyperkinesia, hyperthermia, hyperaggression and audiogenic seizures (Chugh et al, 1992; Gupta et al, 1996; Ganouni et al, 2004). Thus, the accumulated evidence supports the possibility that calcium influx through L-type VGCCs, may activate diverse intracellular messengers or transcription factors to increase glutamatergic strength during benzodiazepine withdrawal and contribute to benzodiazepine dependence. Previous studies in FZP-withdrawn rats demonstrated that hippocampal CA1 neuron AMPAR-mediated hyperexcitability is an essential component of a functional anatomic circuit associated with expression of benzodiazepine withdrawal-anxiety (Xiang and Tietz, 2007). Coupled with the previous findings, the evidence provided in this report suggests that an L-type VGCC dependent Ca 2+ signaling mechanism may be central to CA1 neuron AMPAR synaptic plasticity. Altered L-type VGCC-mediated Ca

123 homeostasis may also serve as a generalized mechanism for dependence on a variety of drugs of abuse and point to a common site for pharmacological interventions to prevent drug dependence. 117

124 Figure 1. A mv 40 0 B +/- Verapamil mv -80 mv 3 s 200 ms C CON D FZP 0.2 na 100 ms 118

125 Figure 1 Representative Ca 2+ current traces. Whole-cell currents were elicited from acutely isolated CA1 neurons. A) Activation pulse protocol: 200 ms depolarizing voltage steps ranging from -70 to 40 mv in 10 mv increments were preceded by a 3 s pre-pulse to -80 mv. Membrane holding potential (V H ) was -65 mv. B) Example traces of maximal Ca 2+ currents evoked in a control CA1 neuron at 0 mv in the presence and absence of 40 µm verapamil, a selective L-type VGCC antagonist. C) and D) Representative calcium current traces elicited in neurons isolated from a control (CON) and a 2-day FZPwithdrawn rat. 119

126 Figure 2. A mv DCON 0D -40 3DFZP 0D -50 pa/pf C mv B mv DCON 0D 7DFZP 0D pa/pf D mv DCON 2D 7DFZP 2D DCON 4D 7DFZP 4D -50 pa/pf -50 pa/pf 120

127 Figure 2 Current-voltage (IV) relationship of Ca 2+ currents elicited from isolated CA1 neurons. Ca 2+ currents were normalized by the membrane capacitance of individual neurons and were represented as current density (pa/pf). A) There was no significant increase in Ca 2+ current density in CA1 neurons isolated from rats immediately after (0- day) 3 day FZP treatment (3D 0D; open circles, n=9. p>0.05) compared to that from matched control rats (closed circles, n=9). B) There was a significant increase (~1.8X) in Ca 2+ current density in CA1 neurons isolated from rats immediately after (0-day) withdrawal from 7 day FZP-treatment (7D 0D; open circles, n=12. *p<0.05) compared to that from matched control rats (close circles, n=10). C) A similar increase in Ca 2+ current density was found in neurons from 2-day FZP-withdrawn rats following 7 day FZP-treatment (7D 2D; open circles, n=25. *p<0.05) in comparison to control neurons (closed circles, n=25). D) There were no significant differences in Ca 2+ current density in neurons from 4-day FZP-withdrawn rats following 7 day FZP-treatment (7D 4D; closed circles, n=8. p>0.05) in comparison to controls (open circles, n=7). 121

128 Current Density (-pa/pf) CONdFZP CON FZP CON FZP CON FZP CON FZP Acute * * 3D 0D 7D 0D 7D 2D 7D 4D 122

129 Figure 3 Temporal pattern of changes in CA1 neuron Ca 2+ currents during acute and chronic FZP treatment and withdrawal. Peak Ca 2+ current density at V H = 0 mv is shown. For acute treatment rats were offered saccharin water for 7 days followed by gavage (2.5 mg/kg, p.o.) with the primary FZP metabolite desalkylfzp (df, n=8, open bars), a dose which results in comparable BZ levels to that after 7 day FZP treatment. Control rats received only the emulsion (C: n=10, closed bars). Another group rats were offered 3 day oral FZP (100 mg/kg) in saccharin solution (3D 0D; C: n=9; F: n=9). All other rat groups were offered 7-day oral FZP treatment (100 mg/kg X 3 days and 150 mg/kg X 4 days) and were withdrawn immediately, 0 days (7D 0D; C: n=10; F: n=12); 2 days (7D 2D; C: n=25; F: n=25); or 4 days (7D 4D C: n=8; F: n=7) after ending 7 day FZP treatment. 123

130 A G/G max DCON 0D 7DFZP 0D 7DCON 2D 7DFZP 2D * mv B Vhalf (-mv) * * 0 CON FZP CON FZP 0 Day 2 Day C dv (mv) CON FZP CON FZP 0 Day 2 Day 124

131 Figure 4 Voltage-dependence of Ca 2+ current activation. A). Current amplitudes elicited by voltage steps from individual neurons were transformed into conductance, normalized to the maximal conductance and fitted with Boltzmann s equation. There was a significant negative shift of the activation curves derived from FZP-withdrawn rats (0 day: 7D 0D, open circles; 2 day: 7D 2D, open squares) compared to neurons from matched control rats (0 day, close circles; 2 day, open circles). B) The voltages of half-maximal activation (V half ) were significantly decreased in neurons from both 0 day FZP-withdrawn rats (open bars, n=11) compared to controls (solid bars, n=9; *p<0.05), as well as 2-day FZP-withdrawn rats (open bars, n=25) compared to their controls (solid bars, n=25, *p<0.05). C). There was no significant difference between the slopes of the activation curves (dv) between any FZP-withdrawn group (open bars) and their matched control group (solid bars, p>0.05). 125

132 A mv 10 D 7DCON 0D 7DFZP 0D I/Imax ms 200 ms -65 mv 7DCON 2D 7DFZP 2D mv B CON E na 500 ms Vhalf (-mv) CON FZP CON FZP 0 Day 2 Day C FZP F 15 dv (mv) CON FZP CON FZP 0 Day 2 Day 126

133 Figure 5 Voltage dependence of steady-state inactivation. A) Inactivation pulse protocol: A 200 ms test-pulse to 10 mv was preceded by a 1.5 s conditioning pre-pulse. Voltage steps ranged from -80 to 10 mv in 10-mV increments. V Hold was -65 mv. B-C) Representative steady-state inactivation Ca 2+ current traces from B) a control CA1 neuron and C) a CA1 neuron isolated from a 2-day FZP-withdrawn rat. D) Boltzmann inactivation curves derived from individual fits of Ca 2+ currents elicited from neurons isolated from control and FZP-withdrawn rats. There were no significant shifts in the inactivation curves from FZP-withdrawn neurons (0 day, open circles; 2 day, open squares) compared to their matched control neurons (0 day, closed circles; 2 day, closed squares). E) The voltages of half-maximal inactivation (V half ) were not significantly different in from that of FZPwithdrawn rats (0 day, open bars; 2 day, open bars) compared to their matched control rats (0 day, solid bars; 2 day, open bars). F) There were also no significant differences between the slopes of the inactivation curves (dv) between any FZP-withdrawn group (open bars) and their matched control group (solid bars). 127

134 A CON/VEH FZP/VEH CON/NIMO FZP/NIMO 20 pa 150 ms B Amplitude (-pa) * 0 CON FZP CON FZP Vehicle Nimodipine 128

135 Figure 6 Prior nimodipine injection prevents the increased in AMPAR-mediated mepscs in neurons isolated from 2-day FZP-withdrawn rats. Rats were injected systemically with the L-type VGCC antagonist nimodipine (10 mg/kg, i.p.) or vehicle (0.5% Tween- 80, 2 ml/kg, i.p.) 1 day after FZP removal. CA1 neurons were isolated from 2-day FZPwithdrawn rats the following day. A) Representative mepsc current traces isolated from 2-day FZP-withdrawn (FZP) and control neurons (CON) injected with nimpodipine (NIM) or vehicle (VEH). B) Average amplitudes of AMPAR-mediated mepscs (V H = - 80 mv) in CA1 neurons isolated from control rats (black bars) and 2-day FZP-withdrawn rats (open bars) following either vehicle or nimodipine injection were compared. As previously reported (Van Sickle et al., 2004) there was a significant increase in mepsc amplitude (39%, p<0.05) in CA1 neurons from 2-day FZP-withdrawn rats (n=8) in comparison to those isolated from control rats (n=8). However, there was no difference in mepsc amplitude in CA1 neurons from FZP withdrawn rats (n=5) given a systemic nimodipine injection 24 hr prior to recording, compared to those isolated from the matched control group (n=7). 129

136 A. Ca 2+ Current Inhibition (%) No Train FZP 1 μm Train B B R Train Frequency (Hz) Ca 2+ Current Inhibtion (%) DZP FZP log [BZ] (μm) 130

137 Figure 7 Concentration- and Use-dependent inhibition of L-type VGCC-dependent Ca 2+ current by FZP and DZP. The concentration- and use- dependent effect of FZP on L- type VGCC currents were evaluated in DIV hippocampal cultured cells using whole-cell techniques. (A). Without a train of consecutive depolarizing pulses, 1 μm FZP showed little inhibitory effect on L-type VGCC-dependent Ca 2+ current. With the presence of train application at frequencies of 1, 2, 3 and 4 Hz, there was about 45% inhibition by 1 μm FZP of L-type Ca 2+ current density compare to the presence of vehicle. Notice there is a Ca 2+ dependent inactivation on L-type VGCC function after train of consecutive depolarizing pulses. A recovery test pulse, without prior train stimulation was given at the end of the recording to exclude run-down of Ca 2+ channel function or seal degradation. (B). Benzodiazepine ( μm) effects to inhibit Ca 2+ current test pulses were evaluated after 5 min pre-incubation during the train (FZP and DZP) or no train condition (FZP). There is a concentration-dependent inhibition of FZP and DZP on L-type Ca 2+ current. Also notice that DZP appears to have a less potency than FZP on inhibition of L-type Ca 2+ current. 131

138 REFERENCES: Alves SH, Pinheiro G, Motta V, Landeira-Fernandez J, Cruz AP. (2004) Anxiogenic effects in the rat elevated plus-maze of 5-ht (2C) agonists into ventral but not dorsal hippocampus. Behavior pharmacology. 15(1): Bateson AN. (2002). Basic pharmacologic mechanisms involved in benzodiazepine tolerance and withdrawal. Curr Pharm Des. 8(1): Berridge M.J. (1998) Neuronal Calcium Signaling. Neuron. 21: Borroni AM, Fichtenholtz H, Woodside BL, Teyler TJ (2000). Role of voltage-dependent calcium channel long-term potentiation (LTP) and NMDA LTP in spatial memory. J Neurosci 20: Catterall WA (2000). Structure and regulation of Voltage-gated Calcium channels. Annu. Rev. Cell Dev. Biol. 16: Chavas J, Forero ME, Collin T, Liano I and Marty A (2004). Osmotic tension as a possible link between GABA A receptor activation and intracellular calcium elevation. Neuron 44: Chen KC, Blalock EM, Thibault O, Kaminker P, Landfield PW (2000). Expression of alpha 1D subunit mrna is correlated with L-type Ca2+ channel activity in single neurons of hippocampal "zipper" slices. Proc Natl Acad Sci U S A. 97(8): Chugh Y, Saha N, Sankaranarayanan A., Sharma PL (1992). Effect of peripheral administration of cinnarizine and verapamil on the abstinence syndrome in diazepamdependent rats. Psychopharmacology 106:

139 Cui xy, Zhao X, Chu QP, Chen BQ and Zhang YH. (2007). Influence of diltiazem on the behavior of zolpidem-treated mice in the elevated-plus maze test. J Neural Transm. 114(2): Degroot A, Kashluba S, Treit D. (2001) Septal GABAergic and hippocampus cholinergic systems modulate anxiety in the plus-maze and shock-probe tests. Pharmacol Biochem behav. 69 (3-4): Degroot A, Treit D. (2002). Dorsal and ventral hippocampus cholinergic systems modulate anxiety in the plus-maze and shock-probe tests. Brain Research, 949(1-2): Dolmetsch RE, Pajvani U, Fife K, Spotts JM, Greenberg ME (2001). Signaling to the nucleus by an L-type calcium channel-calmoduline complex through the MAP kinase pathway. Science, 294: El Ganouni S, Hanoun N, Boni C, Tazi A, Hakkou F, Hamon M. (2004). Prevention of diazepam withdrawal syndrome by nifedipine-behavioral and neurochemical studies. Pharmacol Biochem Behav. 79(2): Gallager DW, Malcolm AB, Anderson SA, Gonsalves SF (1985). Continuous release of diazepam: electrophysiological, biochemical and behavioral consequences. Brain Res 342: Ganouni SE, Hanoun N, Boni C, Tazi A, Hmon F (2004). Prevention of diazepam withdrawal syndrome by nifedipine behavioral and neurochemical studies. Pharmacology, Biochemistry and Behavior 79: Gershon E (1992). Effect of benzodiazepine ligands on Ca 2+ channel currents in Xenopus oocytes injected with rat heart RNA. J Basic Clin Physiol Pharmacol 3:

140 Gorter JA, Borgdorff A, van Vliet EA, da Silva FHL, Wadman WJ (2002). Differential and long-lasting alterations of high-voltage activated calcium currents in CA1 and dentate granule neurons after status epilepticus. European Journal of neuroscience, 16: Gould RJ, Murphy KM, Snyder SH. (1985). Autoradiographic localization of calcium channel antagonists receptors in rat brain with [ 3 H]nitrendipine. Brain Res. 330: Griffiths RR and Johnson MW. (2005) Relative abuse liability of hypnotic drugs: a conceptual framework and algorithm for differentiating among compounds. J Clin Psychiatry. 66 Suppl 9: Groth RD, Dunbar RL, Mermelstein PG (2003). Calcineurin regulation of neuronal plasticity. Biochem Biophys Res Commun 311: Gupta MB, Nath C, Patnaik GK, Saxena RC. (1996) Effect of calcium channel blockers on withdrawal syndrome of lorazepam in rats. Indian J Med Res. 103: Hayashida S, Katsura M, Totigoe F, Tsujimura A, Ohkuma S. (2005) Increased expression of L-type high voltage-gated calcium channel alpha1 and alpha2/delta subunits in mouse brain after chronic nicotine administration. Brain Res Mol Brain Res. 135 (1-2): Hell JW, Westenbroek RE, Warner C, Ahlijanian MK, Prystay W, Gilbert MM, Snutch TP, Catterall WA. (1993). Identification and differential subcellular localization of the neuronal class C and D L-type calcium channel alpha 1 subunits. J Cell Biol 123: Hobom M., Dai S., Marais E., Lacinová L., Hofman F., Klugbauer N. (2000) Neuronal distribution and functional characterization of the calcium channel α 2 σ-2 subunit. Eur. J. Neurosci. 12:

141 Hofmann F, Lacinova L, Klugbauer N. (1999) Voltage-dependent calcium channels: from structure to function. Rev Physiol Biochem Pharmacol. 139: Ishizawa Y, Furuya K, Yamagishi S and Dohi S (1997). Non-GABAergic effects of midazolam, diazepam and flumazenil on voltage-dependent ion currents in NG cells. NeuroReport 8: Izzo E, Auta J, Impagnatiello F, Pesold C, Guidotti A, Costa E (2001). Glutamic acid decarboxylase and glutamate receptor changes during tolerance and dependence to benzodiazepines. Proc Natl Acad Sci USA 98: Johansen J, Taft WC, Yang J, Kleinhaus AL, DeLorenzo RJ (1985). Inhibition of Ca2+ conductance in identified leech neurons by benzodiazepines. Proc Natl Acad Sci U S A. 82(11): Kagamiishi Y, Yamamoto T, Watanabe S. (2003) Hippocampal serotonergic system is involved in anxiety-like behavior induced by corticotrophin-releasing factor. Brain research. Nov 21; 991(1-2): Katsura M, Mohri Y, Shuto K, Hai-Du Y, Amano T, Tsujimura A, Sasa M, Ohkuma S (2002). Up-regulation of L-type voltage-dependent calcium channels after long term exposure to nicotine in cerebral cortical neurons. J Biol Chem. 277(10): Kelly KM, Ikonomovic MD, Abrahamson EE, Kharlamov EA, Hentosz TM, Armstrong DM. (2003) Alterations in hippocampal voltage-gated calcium channel a 1 subunit expression patterns after kainate-induced status epilepticus in aging rats. Epilepsy Research. 57: Kelley AE. (2004) Memory and addiction: Shared neural circuitry and molecular mechanisms. Neuron, 44 :

142 Kortekaas P, Wadman WJ (1997). Development of HVA and LVA calcium currents in pyramidal CA1 neurons in the hippocampus of the rats. Developmental brain research, 101: Lamprea MR, Cardenas FP, Silveira R, Walsh TJ, Morato S. (2003). Effects of septal cholinergic lesion on rat exploratory behavior in an open-field. Braz J Med Biol Res. 36 (2): Lipscombe D, Helton TD, Xu W (2004). L-type calcium channels: the low down. J Neurophysiol 92: Lyons HR, Land MB, Gibbs TT, Farb DH (2001). Distinct signal transduction pathways for GABA-induced GABA(A) receptor down-regulation and uncoupling in neuronal culture: a role for voltage-gated calcium channels. J Neurochem. 78(5): Marty A. and Liano I (2006). Excitatory effects of GABA in established brain networks. Trends Neurosci 28: McKinney BC, Murphy GG (2006). The L-Type voltage-gated calcium channel Cav1.3 mediates consolidation, but not extinction, of contextually conditioned fear in mice. Learn Mem. 13(5): Mendelson WB, Owen C, Skolnick P, Paul SM, Martin JV, Ko G, Wagner R (1984). Nifedipine blocks sleep induction by flurazepam in the rat. Sleep. 7(1):64-8. Morgan SL, Teyler TJ (1999).VDCCs and NMDARs underlie two forms of LTP in CA1 hippocampus in vivo. J Neurophysiol 82: Nestler EJ (2005). Is there a common molecular pathway for addiction? Nat Neurosci 8:

143 Rajadhyksha AM and Kosofsky BE (2005) Psychostimulants, L-type calcium channels, kinases, and phosphatases. The Neuroscientist 11: Reuveny E, Twombly DA and Narahashi T (1993). Chlordiazepoxide block of two types of calcium channels in neuroblastoma cells. J Pharmacol Exp Ther 264: Podhorna J. (2002). The experimental pharmacotherapy of benzodiazepine withdrawal. Current Pharmaceutical Design. 8: Poisbeau P, William SR, Mody I (1997). Silent GABA A synapses during flurazepam withdrawal are region-specific in the hippocampal formation. J Neurosci, 17: Pourmotabbed A, Motamedi F, Fathollahi Y, Mansouri FA, Semnanian S. (1998). Involvement of NMDA receptors and voltage-dependent calcium channels on augmentation of long-term potentiation in hippocampal CA1 area of morphine dependent rats. Brain Research 804: Rabbani M, Little HJ. (1999). Increase in neuronal Ca 2+ flux after withdrawal from chronic barbiturate treatment. European Journal of Pharmacology 364: Rajadhyksha AM and Kosofsky BE. (2005) Psychostimulants, L-type calcium channels, Kinases, and phosphatases. The Neuroscientist, 11(5): Ramsey-Williams VA, Wu Y, Rosenberg HC (1994). Comparison of anticonvulsant tolerance, crosstolerance, and benzodiazepine receptor binding following chronic treatment with diazepam or midazolam. Pharmacol Biochem Behav. 48(3): Reichling DB, Kyrosis A, Wang J and MacDermott AB (1994). Mechanisms of GABA and glycine depolarization-induced calcium transients in rat dorsal horn neurons. J Physiol 476:

144 Reuveny E, Twombly DA and Narahashi T (1993). Chlordiazepoxide block of two types of calcium channels in neuroblastoma cells. J Pharmacol Exp Ther 264: Rosenberg HC, Tietz EI, Chiu TH (1991). Differential tolerance to the antipentylenetetrazol activity of benzodiazepines in flurazepam-treated rats. Pharmacol Biochem Behav. 39(3): Rosenberg HC (1995). Differential expression of benzodiazepine anticonvulsant crosstolerance according to time following flurazepam or diazepam treatment. Pharmacol Biochem Behav. 51(2-3): Shinnick-Gallagher P, McKernan MG, Xie JG and Zinebi F. (2003) L-Type Voltage-Gated Calcium Channels Are Involved in the in Vivo and in Vitro Expression of Fear Conditioning. Annals of the New York Academy of Sciences. 985: Sun MK, Zhao WQ, Nelson TJ and Alkon DL (2001). Theta rhythm of hippocampal CA1 neuron activity: gating by GABAergic synaptic depolarization. J Neurophysiol 85: Taft WC, DeLorenzo RJ (1984). Micromolar-affinity benzodiazepine receptors regulate voltage-sensitive calcium channels in nerve terminal preparations. Proc Natl Acad Sci 81: Van Sickle B, Tietz E (2002). Selective enhancement of AMPA receptor-mediated function in hippocampal CA1 neurons from chronic benzodiazepine-treated rats. Neuropharmacology 43: Van Sickle BJ, Cox As, Schak k, Greenfield Jr.LJ, Tietz EI. (2002) chronic benzodiazepine administration alters hippocampus CA1 neuron excitability: NMDA receptor function and expression. Neurophamacology 43:

145 Van Sickle B., Kun Xiang, E.I. Tietz. (2004) Transient plasticity of hippocampus CA1 neuron glutamate receptors contributes to benzodiazepine withdrawal anxiety. Neuropsychopharmacology. 29: Wafford KA. (2005). GABAA receptor subtypes: any clues to the mechanism of benzodiazepine dependence? Curr Opin Pharmacol. 5(1): Walter HJ, Messing MO. (1999). Regulation of neuronal voltage-gated calcium channels by ethanol. Neurochemistry international 35: Watabe S, Yoshii M, Ogata N, Tsunoo A, Narahashi T (1993). Differential inhibition of transient and long-lasting calcium channel currents by benzodiazepines in neuroblastoma cells. Brain Res. 606(2): Watson WP, Little HJ (2002). Selectivity of the protective effects of dihydropyridine calcium channel antagonist against the ethanol withdrawal syndrome. Brain Research 930: Westenbroek RE, Hoskins L, Catterall WA (1998). Localization of Ca 2+ channel subtype on rat spinal motor neurons, interneurons, and nerve terminals. J Neurosci 18: Whittington MA, Little HJ. (1993). Changes in voltage-operated calcium channels modify ethanol withdrawal hyperexcitability in mouse hippocampal slices. Exp Physiol. 78(3): Xia Z, Storm DR. (2005). The role of calmodulin as a signal integrator for synaptic plasticity. Nat Rev Neurosci. 6: Xiang K, Tietz EI (2007). Benzodiazepine-induced hippocampal CA1 neuron AMPA receptor plasticity linked to severity of withdrawal-anxiety: Differential role of voltagegated calcium channels and NMDA receptors. Behav Pharmacol in press. 139

146 Xie, X.H., Tietz, E.I. (1992). Reduction in potency of selective γ-aminobutyric acid A agonist and diazepam in CA1 region of in vitro hippocampal slices from chronic flurazepam-treated rats. J Pharmacol ExpTher 262, Xu W, Lipscombe D (2001). Neuronal Ca(V)1.3 alpha(1) L-type channels activate at relatively hyperpolarized membrane potentials and are incompletely inhibited by dihydropyridines. J Neurosci 21: Veng LM, Mesches MH, Browning MD (2003). Age-related working memory impairment is correlated with increases in the L-type calcium channel protein alpha1d (Cav1.3) in area CA1 of the hippocampus and both are ameliorated by chronic nimodipine treatment. Mol Brain Res. 110(2): Zeng X, Xie XH and Tietz EI (1995). Reduction of GABA-mediated inhibitory post-synaptic potentials in hippocampus CA1 pyramidal neurons following oral flurazepam administration. Neuroscience 66: Zeng X and Tietz EI (1997). Depression of early and late monosynaptic inhibitory postsynaptic potentials in hippocampal CA1 neurons following prolonged benzodiazepine administration: role of reduction in Cl - driving force. Synapse 25: Zeng,X., Tietz,E.I., (1999). Benzodiazepine tolerance at GABAergic synapses on hippocampal CA1 pyramidal cells. Synapse 31: Zeng X and Tietz EI (2000). Role of bicarbonate ion in mediating decreased synaptic conductance in benzodiazepine tolerant hippocampal CA1 pyramidal neurons. Brain Res 868:

147 MANUSCRIPT III Nimodipine reversal of postsynaptic GABA A receptor-mediated mipsc changes in hippocampal CA1 pyramidal neurons following chronic flurazepam administration Kun Xiang and Elizabeth I. Tietz*, Department of Physiology and Pharmacology and the Cellular and Molecular Neurobiology Program, University of Toledo College of Medicine, Health Science Campus, Toledo, OH

148 ABSTRACT One week oral flurazepam (FZP) administration in rats results in reduced GABA A receptor-mediated synaptic transmission in hippocampal CA1 pyramidal neurons associated with benzodiazepine tolerance. Nimodipine (10 mg/kg, i.p.) or the vehicle 0.5 % Tween 80 (2 ml/kg) was injected 1 day after ending 1-week FZP treatment and 24 hours prior to hippocampal slice preparation to evaluate the role of L-type VGCC in regulating benzodiazepine-induced CA1 neuron GABA A receptor-mediated function. Nimodipine reversed the reduced mipsc amplitude 2 days after drug removal. Prior nimodipine injection and also reversed the reduction of GABA A receptor unitary channel conductance, estimated using peak-scaled non-stationary fluctuation analysis. However, nimodipine injection failed to reverse in vitro tolerance to zolpidem s ability to prolong mipsc decay in FZP-treated neurons, suggesting multiple mechanisms may be involved in regulating GABA A receptor-mediated synaptic transmission following chronic FZP administration. Nimodipine superfusion inhibited mipsc amplitude and prolonged mipsc decay only at high concentrations (> 30 μm), significantly greater than attained in vivo (1 μm) 45 min after a single acute injection. The findings suggest that the effects of nimodipine to reverse GABA A receptor functional changes are through interruption of L- type VGCC activity, supporting the role of L-type VGCC Ca 2+ -mediated signaling pathways in the modulation of GABA A receptor synaptic function following chronic FZP administration. 142

149 INTRODUCTION As a group of wildly prescribed medications for their potent hypnotic, anxiolytic and anticonvulsant actions, benzodiazepines exert their therapeutic effects through potentiating fast, γ-aminobutyric acid type A receptor-mediated inhibitory neurotransmission. Nevertheless, prolonged administration of benzodiazepines results in functional tolerance to many of their pharmacological effects including their sedative and anticonvulsant effects, as well as a characteristic withdrawal syndrome on cessation of treatment (for review see Hutchinson et al., 1994; Griffiths and Weerts, 1997). GABA receptors, the main locus of benzodiazepine actions, have been extensively studied to elucidate the neuroadaptive process underlying benzodiazepine tolerance and withdrawal. Native GABA receptors are pentameric assemblies of subunit proteins (α 1-6, β 1-4, γ 1-3, ρ 1-3, σ, π, ε and θ) with an integral Cl - channel (Macdonald and Olsen, 1994; Costa et al., 2002). Changes in GABA receptor structure, function, and pharmacology associated with prolonged benzodiazepine exposure include region-specific decreases in the number of benzodiazepine binding sites, reduced GABA agonist efficacy, consistent reductions in allosteric coupling between GABA and benzodiazepine binding sites, modulation of GABA A receptor subunit mrna and protein expression and reduced GABA A receptormediated synaptic inhibition (Heninger et al., 1990; Gallager et al., 1991; Xie and Tietz, 1992; Zhao et al., 1994; Barnes, 1996; Hutchinson et al., 1996; Impagnatiello et al., 1996; Pesold et al., 1997; Posibeau et al., 1997; Zeng and Tietz, 1999; Costa et al., 2002). 143

150 Although the exact neural mechanisms underlying the development of benzodiazepine tolerance and dependence remain incomplete, new evidence indicates a role for L-type voltage-gated calcium channels (VGCCs) in mediating GABA A receptor changes associated with chronic benzodiazepine administration. Raising intracellular Ca 2+ and activating Ca 2+ /calmodulin-dependent protein kinase have been shown to regulate GABA A receptor-mediated Cl - current (Wang et al., 1995; Koninck and Mody, 1996; Aguayo et al., 1998; Churn and DeLorenzo et al., 1998; Alix et al., 2002). Moreover, GABA-induced GABA A receptor downregulation was suggested to be the product of a transcriptional repression of GABA A receptor subunit genes (Lyons et al., 2000; Russek et al., 2000) that depends on activation of L-type VGCC (Lyons et al., 2001; Gravielle et al., 2005). Notably, nifedipine preincubation from 18-day-old rat embryos neurons inhibited both GABA-induced increases in intracellular Ca 2+ concentration and GABA A receptor down-regulation without effects on the allosteric coupling between benzodiazepine and GABA binding sites (Lyons et al., 2001). Recent studies suggest that administration of L-type VGCC antagonists can also interrupt withdrawal symptoms after ending benzodiazepine treatment (Gupta et al, 1996; Podhorna, 2002; EI Ganouni et al, 2004, Cui et al, 2006). Benzodiazepines can directly modulate VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Gershon, 1992; Reuveny et al, 1993; Ishizawa et al, 1997) and may modify L-type VGCC function following chronic FZP administration (Xiang et al., 2007). Indeed, as reported in other forms of neuronal plasticity, acute systemic injection of an L-type VGCC antagonist, nimodipine, averted hippocampal CA1 neuron AMPA receptor current enhancement and anxiety-like behavior in benzodiazepine-withdrawn rats (Xiang and Tietz, 2007), also 144

151 pointing to a role of L-type VGCCs in mediating benzodiazepine dependence. Whether GABAergic system dysfunction underlying benzodiazepine tolerance is modulated by L- type VGCC Ca 2+ signaling mechanisms is unknown. In this study we evaluated the role of L-type VGCC in mediating GABA A receptor functional plasticity associated with chronic benzodiazepine administration. By means of pharmacological antagonism of VGCC activity one possible Ca 2+ -mediated mechanism underlying impaired hippocampal CA1 neuron GABA A receptor activity following one week flurazepam administration was explored. Whole-cell hippocampal slice-patch studies were carried out to evaluate the effect of prior systemic L-type VGCC antagonist nimodipine administration on GABA A receptor-mediated miniature inhibitory post-synaptic current (mipscs) amplitude and kinetics. In vitro tolerance to zolpidem, an α1 selective GABA A receptor ligand, to enhance mipsc decay was also examined in the same CA1 neurons. Effects of nimodipine administration on GABA A receptor singlechannel conductance were further investigated by non-stationary variance analysis. The concentration dependent effect of nimodipine to directly affect GABA receptor-mediated mipscs was also investigated. Evaluating the differential effects of an L-type VGCC antagonist on measures of GABA dysfunction might provide some insight into the role of L-type VGCCs in mediating benzodiazepine tolerance and dependence. 145

152 METHODS Experimental protocols involving the use of vertebrate animals were approved by the University of Toledo College of Medicine (formerly the Medical University of Ohio), Institutional Animal Care and Use Committee (IACUC) and conformed to National Institutes of Health ethical guidelines. Drug Treatments Chronic Flurazepam Administration. FZP treatment in rats was as previously described (Zeng and Tietz, 1999). In short, following a 2 day adaptation period when rats were offered only a 0.02% saccharin vehicle, male Sprague-Dawley rats (initial age PN22-25, Harlan, Indianapolis, IN) were offered FZP (flurazepam dihydrochloride, ph 5.8) for 1 week in saccharin solution as their only source of drinking water. The concentration of FZP was adjusted daily according to each rat s body weight and fluid consumption (100 mg/kg X 3 days and 150 mg/kg X 4 days) appropriate to flurazepam s relative potency, oral bioavailability and biotransformation resulting in benzodiazepine brain levels equivalent to other common benzodiazepine chronic treatments (Gallager et al, 1985; Lau et al, 1987; Xie and Tietz, 1992). Only rats that consumed a criterion dose of an average >120 mg/kg/day were accepted for study. Saccharin water was again provided during the 2-day withdrawal period. Unlike in humans, residual FZP and metabolites rapidly decline over the first 24 hours after drug removal and are no longer detectable in hippocampus 2 days after 1-week FZP administration (Xie and Tietz, 1992). Pair-handled control rats receive saccharin water for the same length of time. Rats were euthanized for 146

153 hippocampal slice preparation on PN The experimenter was not informed the rats treatment histories until after the data analysis were completed. Systemic Antagonist Injection. Two groups of control and FZP-treated rats were given a single intraperitoneal injection of L-type VGCC antagonist, nimodipine (10 mg/kg, i.p.) or the vehicle 0.5 % Tween 80 (2 ml/kg) 1 day after ending 1-week FZP treatment and 24 hours prior to hippocampal slice preparation. This dose of nimodipine has minimal effect on locomotion, does not produce ataxia, has no effect on seizure threshold, yet reverses behavioral signs of ethanol dependence (Watson and Little, 2002). Electrophysiology Hippocampal slice preparation. Hippocampal slices (400 μm) were prepared from rats as previously described (Van Sickle et al, 2004). Briefly, transverse dorsal hippocampal slices were cut on a Vibratome (Ted Pella, Inc.) in ice-cold, pre-gassed (95%O 2 /5% CO 2 ) artificial cerebrospinal fluid (ACSF) containing (in mm): NaCl, 120; KCl, 2.5; CaCl 2, 0.5; MgSO 4 7.0; NaH 2 PO 4 1.2; NaHCO 3, 2; D-glucose, 20; Ascorbate, 1.3, ph 7.4. Slices were maintained at room temperature (RT) for 15 min in gassed, low-calcium, high-magnesium ACSF, then transferred to normal ACSF containing (in mm): NaCl, 119; KCl, 2.5; CaCl 2, 1.8; MgSO 4 1.3; NaH 2 PO ; NaHCO 3, 26; D-glucose, 10; ph 7.4. Slices were maintained at room temperature for 1 hr in ACSF. During recording, slices were superfused at a rate of 2.5 ml/min with gassed ACSF at room temperature. GABA A receptor-mediated mipsc recording. GABA A receptor-mediated mipscs were isolated from CA1 pyramidal neurons in the presence of 1 μm Tetrodotoxin (TTX), 10 μm DNQX (6,7-dinitroquinoxaline-2,3-done) and 50 μm APV (DL-2-amino-5 147

154 phosphonovaleric acid) with or without 1 μm zolpidem. Patch pipettes for mipsc recording are filled with (in mm): CsCl, 130; HEPES, 10; EGTA, 1; CaCl 2, 0.5; MgCl 2, 2; Mg-ATP, 2; QX-314, 2; ph 7.2 adjusted with CsOH. Cells are Cl - -loaded to minimize the possible contribution of intracellular Cl - accumulation to the reduction of GABAmediated inhibition suggest by a shift in the E IPSC in previous experiments in FZP-treated neurons (Zeng and Tietz, 1997). Chloride-loading was demonstrated to reverse the usedependent shift in E Cl- due to prolonged GABA activation (Zeng and Tietz, 1999). QX- 314 (2 mm), an intracellular sodium channel blocker, is also included to block the spontaneous firing of CA1 pyramidal neurons. Resting membrane potential (RMP) was measured immediately upon cell break-in. Neurons were voltage-clamped (V H = -70 mv) in continuous mode (csevc) using an Axoclamp 2A amplifier (Axon Instr., Union City, CA). Current output was low-pass filtered (10 khz), DC-offset, amplified 10,000-fold and continuously monitored on-line (PClamp 8.0, Axon). The digitized signal (Digidata 1200A, Axon) was stored on disk for later off-line analysis. Cells in which the holding current changed by more than 20% or the seal degraded, were abandoned. mipsc activity was recorded 5 min and analyzed with MiniAnalysis software (Synaptosoft Inc., Leonia, NJ). Peak mipsc amplitude was measured from baseline. Decay kinetics and mipsc amplitude were estimated using a single exponential function: [y(t)=a*exp(-t/τ)]. Wholecell data were compared by repeated measures ANOVA with post-hoc analysis by the method of Scheffé. Zolpidem effects on mipsc amplitude and decay In a subset of FZP-treated and control cells, mipsc activity was recorded for 8 min in the presence of 1µM zolpidem after the 5 min baseline recording. The final 3 min segment in the presence of zolpidem was used 148

155 for off-line analysis of mipsc amplitude and decay kinetics. mipsc decay phase was fitted with a single exponential function y(t)=a*exp(-t/t). It was previously shown in a lager number of CA1 neurons that the proportion of control (65%) and FZP-treated (62%) neurons best fit with a mono-exponential versus bi-exponential decay were similar (Zeng and Tietz, 1999). The degree of zolpidem potentiation of mipsc decay was expressed as a fraction of the control response. Non-stationary fluctuation analysis. mipsc events from FZP-treated and control cells following vehicle or nimodipine injection were also analyzed using peak-scaled nonstationary fluctuation analysis using MiniAnalysis software. The average of mipscs was scaled to each individual event before computing the variance. Data were fitted with the equation σ 2 =ii-i 2 /N, where σ 2 is the variance, I is the mean current, N is the number of channels activated at the peak of the mean current and, i is the singlechannel current (Zeng and Tietz, 1999). Unitary channel conductance (γ) was derived from γ=i/v, where V is the driving force (V H = -70 mv, E REV = 0 mv). Nimodipine concentration-response effects on mepscs. To determine whether nimodipine had direct effects on GABAR-mediated mipscs, nimodipine ( µm) was superfused onto hippocampal slices during mipsc recording. After recording baseline mipsc activity for 5 min, nimodipine or vehicle (0.0001% to 0.1% DMSO in water) was added to the superfusate in increasing concentrations for 10 min each. mipsc amplitude and kinetics were analyzed using the final 5 min recording period at each concentration. 149

156 Radioligand binding studies. Radioligand binding study was performed in order to investigate the nimodipine concentration in the rat brain after in vivo injection. Briefly, rats were injected with nimodipine (10 mg/kg, ip) and euthanized 15, 30, 45, min or 24 hr later (n=2-3 time point). Whole brains (minus brain stem) were dissected, homogenized in 4 vols of ethanol and centrifuged at 10,000 X g for 20 min. Ethanol extracts (40 ul) were used to displace 2 nm specific [ 3 H]PN high affinity binding to triple-washed P2 membranes prepared from whole rat brain minus cerebellum (~1 mg/protein/ml, using standard techniques except that during the first resuspenion in (50 mm Tris buffer (ph 7.7) membranes were brought to 37 C for 30 min (Weiland and Oswald, 1985). Tubes (0.4 ml) were incubated for 90 min at room temperature in the dark and the reaction terminated by vacuum filtration followed by 3 X 0.5 ml buffer washes on #32 glass fiber filters (Keene, New Hampshire). Non-specific binding was in the presence of 10 μm nitrendipine. Radioactivity on filters was counted 5 min in ScintiSafe (30%, Fisher Scientific). Brain extracts were compared to a standard curve generated with 14 concentrations of nimodipine ranging from 0.01 nm to 10 mm nimodipine. Drug solutions Drugs used for superfusion during whole-cell recording were dissolved at 100 times their final concentration and added to the superfusate with a syringe pump (Razel, World Precision Instruments, Inc., Sarasota, FL) at a rate of 25 to 75 μl/min to achieve their final concentrations. For in vivo injection, nimodipine was dissolved in 0.5 % Tween-80 solution and kept in a light-tight vial. For in vitro perfusion, nimodipine was dissolved in 150

157 DMSO to make a 10 mm stock solution diluted to the final concentration as needed (from 0.1 to 100 µm). All other drugs were dissolved in dh 2 O. DNQX (6,7- dinitroquinoxaline-2,3-done), QX-314 (lidocaine N-ethyl bromide quaternary salt), APV (DL-2-amino-5 phosphonovaleric acid), FZP dihydrochloride, and nimodipine are all from Sigma-Aldrich Chemical Co. (St Louis, MO). Tetrodotoxin (TTX) was obtained from Alamone Laboratories (Jerusalem, Israel). Zolpidem was kindly provided by Synthélabo Recherche (Bagneux, France). RESULTS Prior nimodipine injection reverses the decrease in mipsc amplitude in CA1 neurons from 2-day FZP-withdrawn rats. Since inhibition of L-type VGCCs can modulate GABA A receptor downregulation after prolonged GABA exposure (Lyons et al., 2001; Gravielle et al., 2005), the potential role of L-type VGCC in modulating GABA A receptor dysfunction following chronic benzodiazepine administration was investigated. Following chronic FZP administration, rats were injected systemically with the L-type VGCC antagonist nimodipine (10 mg/kg, i.p.) or vehicle (0.5% Tween-80, 2 ml/kg, i.p.) one day before hippocampal slice preparation. One day later, hippocampal slices were prepared from 2-day FZP-withdrawn rats followed by electrophysiological recordings. GABA A receptor-mediated mipscs were recorded in CA1 neurons in hippocampal slices in the presence of 1μM TTX, 10μM DNQX and 50 μm APV at a holding potential of -70 mv (Fig. 1A and Table 1). As previously reported (Zeng and Tietz, 1999), there was a significant decrease in mipsc amplitude (43.0%, *p<0.05) in CA1 neurons from 2-day FZP-withdrawn rats (FZP/VEH: 151

158 12.9 ±1.6 pa, n = 8) in comparison to those isolated from control rats (CON/VEH: 22.7 ± 1.1 pa, n=9). Prior vehicle injection had no effect on basal GABA A receptor mipsc amplitude in control neurons or the decrease in mipsc amplitude in 2-day FZPwithdrawn rats. However, following prior nimodipine injection the mean GABA A receptor mipsc amplitude returned to control levels in neurons from 2-day FZPwithdrawn rats (CON/NIM: 22.8 ± 1.2 pa, n=8; FZP/NIM: 23.5 ± 1.1, n=9; p>0.05). There were no differences in resting membrane potential (CON/VEH: ± 1.1 mv, FZP/VEH: ±0.8 mv; CON/NIM: ± 1.2 mv, FZP/NIM: ± 1.2 mv. p>0.05), current decay (CON/VEH: 29.0 ± 1.4 ms; FZP/VEH: 26.2 ± 1.4 ms; CON/NIM: 28.7 ± 2.2 ms; FZP/NIM: 23.1 ± 2.5 ms, p>0.05) or in mepsc frequency (CON/VEH: 0.8 ± 0.1 Hz; FZP/VEH: 0.5 ± 0.1 Hz; CON/NIM: 0.6 ± 0.1 Hz; FZP/NIM: 0.6 ± 0.1 Hz. p>0.05) in neurons from control vs. FZP-withdrawn rats. Prior nimodipine injection failed to reverse the tolerance to zolpidem s ability to prolong decay of mipscs in CA1 neurons from 2-day FZP-withdrawn rats. One way in which BZ tolerance can be assessed in vitro is by the shift in zolpidem s ability to prolong the decay of mipscs recorded in hippocampal CA1 neurons (Zeng and Tietz, 1999; Tietz et al., 1999). To investigate the role of L-type VGCC in mediating in vitro BZ tolerance, the ability of 1 µm zolpidem to enhance mipsc decay and amplitude was evaluated in CA1 neurons from control and FZP-treated rats injected with vehicle or nimodipine 1 day before recording. Effects of 1 µm zolpidem to prolong mipsc decay in CA1 neurons was expressed as a percentage of the average baseline mipsc decay recorded in the absence of zolpidem. As seen in Fig. 2 and Table 1, zolpidem tolerance was indicated by a significant reduction in the ability of zolpidem to enhance mipsc 152

159 decay in FZP-VEH neurons (104.0 ± 3.9%, p>0.05, n=7) in comparison to CON-VEH neurons (131.0 ± 5.4%, *p<0.05, n=6). However, prior systemic nimodipine injection failed to prevent zolpidem tolerance in FZP-NIM neurons (101.4 ± 2.7%, p>0.05, n=7) in comparison to CON-NIM neurons (130.2 ± 3.03%, *p<0.05, n=6). The absence of an effect of nimodipine on in vitro tolerance to zolpidem in contrast to its effects to prevent the reduction in GABA mipsc amplitude suggests that L-type VGCC-mediated Ca 2+ influx might mediate some, but not other measures of GABA dysfunction in rats chronically administered benzodiazepines. As expected from previous recordings carried out at room temperature (Perrais and Ropert, 1999; Zeng and Tietz, 1999) 1 µm zolpidem also increased the amplitude of mipscs in control and FZP-treated cells from rats injected with both vehicle and nimodipine (CON-VEH: ± 3.1%, n=7; FZP- VEH: ± 2.7%, n=7; CON-NIM: ± 6.4%, N=6; FZP-NIM: ± 0.2%, N=7, p>0.05). Effects of nimodipine injection on peak-scaled non-stationary variance analysis of mipscs recorded in CA1 neurons in hippocampal slices from 2-day FZPwithdrawn rats. Previous studies have demonstrated a decrease in GABA A receptor single channel conductance, but not channel number, after chronic flurazepam treatment (Poisbeau et al., 1997; Zeng and Tietz, 1999). Based on regional changes of subunit mrna and protein expression in rat brain, a decreased channel conductance may be associated with switches in GABA receptor subunit composition (Kang and Miller, 1991; Primus and Gallager, 1992; Tietz et al., 1994; Zhao et al., 1994; Holt et al., 1996). A variety of studies have also indicated that raising intracellular Ca 2+ or activating Ca 2+ /calmodulin-dependent 153

160 protein kinase can regulate GABA A receptor-mediated Cl - current (Wang et al., 1995; Koninck and Mody, 1996; Aguayo et al., 1998; Churn and DeLorenzo et al., 1998; Alix et al., 2002). Therefore, the effects of systemic vehicle or nimodipine injection on GABA A receptor unitary channel conductance was analyzed using NSAN analysis in CA1 neurons from control or FZP-withdrawn rats. As shown in Fig 3A, there was a significant (~38.0%) decrease GABA A receptor unitary channel conductance in CA1 neurons in hippocampal slices from 2-day FZPwithdrawn rats compare to control rats ( CON-VEH: 25.9 ± 1.6 ps, n=9; FZP-VEH: 16.0 ± 0.7 ps, n=8, * p< 0.05) as expected based on previous reports (Poisbeau et al., 1987; Zeng and Tietz, 1999). Prior nimodipine injection reversed the decrease in GABA A receptor unitary channel conductance to basal levels (CON-NIM: 25.4 ± 2.7 ps, n=8; FZP-NIM: 26.3 ± 2.6 ps, n=7, p>0.05). There was no significant difference in the mean channel number (N) between control (14.9 ± 2.9, n=9) and FZP-withdrawn neurons (18.7 ± 3.6, n=8, p>0.05). Prior nimodipine injection had no effect on channel number in either group (CON-NIM: 16.2 ± 2.1, n=8; FZP-NIM: 16.7 ± 1.7, n=7, p>0.05). Effects of in vitro nimodipine superfusion on GABA A receptor-mediated mipscs. To evaluate whether systemic nimodipine injection may have a direct effect to modify GABA A receptor activity, the concentration-response profile for the effect of nimodipine on GABA A receptor-mediated mipsc characteristics was evaluated. mipscs were recorded for 5 min without nimodipine during the baseline hippocampal slice recording period. Nimodipine was then superfused in increasing concentrations ranging from 10 µm to 300 µm. Average GABA A receptor mipsc amplitude (V H = -80 mv) in CA1 neurons during vehicle or nimodipine superfusion was compared. As shown in Fig. 4A, 154

161 mipsc amplitude was decreased in a concentration-dependent manner following nimodipine superfusion (n=4) at concentrations greater than 30 µm in comparison to neurons superfused with vehicle (n=3). mipsc amplitude was inhibited ~40% in the presence of 300 µm nimodipine, the highest concentration tested, in comparison to vehicle (VEH: 21.0 ± 1.3 pa; NIM: 12.5 ± 1.9 pa, * p<0.05). There were no significant changes (p>0.05) in mipsc resting membrane potential (VEH: ± 1.1 mv; NIM: ± 2.3); rise time (VEH: 1.7 ± 0.2 ms; NIM: 1.5 ± 0.2 ms), or frequency (VEH: 0.7 ± 0.1 Hz, NIM: 0.7 ± 0.1 Hz) during nimodipine superfusion. Nimodipine also produced a concentration-dependent prolongation of mipsc decay at concentrations higher than 30 µm. At 300 µm nimodipine, the highest concentration tested, there was a ~ 58% prolongation of mipsc decay (VEH: 32.0 ± 5.1 ms; 300 µm NIM: 50.8 ± 7.6 ms, *p<0.05), while vehicle had no effects on mipsc decay (VEH: 32.6 ± 2.0 ms; 300 µm VEH: 30.7 ± 1.8 ms, p>0.05). Measurement of nimodipine concentration in the brain after single acute systemic injection A radioligand binding study was performed to measure the nimodipine concentration in rat brain after a single injection (10 mg/kg, i.p.). The nimodipine peak concentration in brain (1.1 μm) was achieved 45 min after injection equivalent to 65.8% inhibition of 2 nm [ 3 H]PN binding (Weiland and Oswald, 1985). Inhibition was negligible within 2 hrs, i.e. similar to that in the extract from a Tween 80 vehicle injected saccharintreated control rat. As expected from the nimodipine effect after 2 hrs, the value from nimodipine binding study was similar to that in control 24 hrs later when rat hippocampal slices would be prepared for mipsc recording. 155

162 DISCUSSION The goal of this study was to establish whether L-type VGCCs may play a role in mediating a variety of in vitro changes in hippocampal CA1 neuron GABA A receptor function that may underlie benzodiazepine tolerance and dependence. Prior injection of nimodipine reversed the decrease of GABA A receptor-mediated mipsc amplitude in hippocampal CA1 neurons 2 days after ending chronic FZP treatment when rats are tolerant to benzodiazepine anticonvulsant effects (Tietz et al., 1999). These findings support the possibility that Ca 2+ signaling through L-type VGCCs may contribute to the GABA A receptor functional changes following chronic FZP administration. Furthermore, nimodipine reversed the reduction in GABA A receptor unitary channel conductance, indicating that these changes may be related to the decreased mipsc amplitude and that L-type VGCC activation may also be involved in the protracted reductions in GABA A receptor single channel conductance, across the withdrawal period (Posibeau et al., 1997; Zeng and Tietz, 1999). However, prior nimodipine injection had no effects on in vitro tolerance to zolpidem s ability to prolong the GABA A receptor-mediated mipsc decay, suggesting L-type VGCC may not be involved in these processes, induced at an earlier time-point. Extensive studies have described the changes in GABA A receptor structure, function, and pharmacology associated with prolonged benzodiazepine exposure that could underlie benzodiazepine tolerance and dependence (for reviews see Bateson 2002; Allison and Pratt, 2003), yet the mechanisms underlying these changes have been poorly elucidated. A series of electrophysiological studies have demonstrated a pronounced 156

163 reduction in GABAergic inhibition in hippocampal CA1 neurons reflected in a decrease in mipsc amplitude and unitary conductance (Xie and Tietz, 1992; Zeng and Tietz, 1995; Poisbeau et al., 1997; Zeng and Tietz, 1999). Both the reduction in mipsc amplitude and the ability of benzodiazepine site ligands to modify mipsc kinetics in vitro following 1-week FZP administration were associated with benzodiazepine anticonvulsant tolerance (Zeng and Tietz, 1999; Tietz et al., 1999). The modulation of expression of several GABA A receptor subunit mrnas and proteins has been suggested to be associated with GABA A receptor functional changes following chronic benzodiazepine administration (Kang and Miller, 1991; Primus and Gallager, 1992; Tietz et al., 1994; Zhao et al., 1994; Holt et al., 1996; Chen et al., 1999; Tietz et al., 1999). GABA-induced GABA A receptor down-regulation was suggested to be due to the transcriptional repression of GABA A receptor subunit genes related to activation of L-type VGCCs (Lyons et al., 2001; Gravielle et al., 2005) since nifedipine preincubation inhibited both GABA-induced increases in intracellular Ca 2+ concentration and GABA A receptor down-regulation (Lyons et al., 2001). Similarly, L-type VGCCs have been reported to play role in various activity-dependent models of neuronal plasticity (Morgan and Teyler, 1999; Borroni et al., 2000; Rajadhyksha and Kosofsky, 2005). In fact, during ethanol withdrawal and following psychostimulant administration, activation of L-type VGCCs can result in the influx of intracellular calcium and downstream activation of Ca 2+ /CaM-activated kinase and phosphatase pathways, eventually involving CREB-induced gene expression, also important to neuronal and experience-dependent plasticity (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005). Our observation that prior systemic nimodipine 157

164 reversed decreased GABA A receptor-mediated mipsc amplitude (Fig 1 and Table 1), suggests that activation of L-type VGCC may also contribute to the regulation of reduced GABAergic function, likely through mediating GABA A receptor subunit expression and/or composition in hippocampal CA1 neurons. The ability of benzodiazepines to prolong mipsc decay in CA1 pyramidal neurons (De Koninck and Mody, 1992; Poisbeau et al., 1997) was previously observed to be decreased following chronic flurazepam administration (Zeng and Tietz, 1999; Tietz et al, 1999). Zolpidem potency to potentiate GABA-induced currents was also decreased in acutely dissociated CA1 neurons following chronic diazepam treatment in rats (Itier et al., 1996). Though nimodipine reversed the reduction of GABA A receptor -mediated mipsc amplitude (Fig 1), it failed to reverse the tolerance to zolpidem s ability to prolong mipsc decay in the same CA1 neurons (Fig 2 and Table 1). Thus it suggests that the mechanisms underlying reduction in GABA A receptor mipsc amplitude and in vitro tolerance to zolpidem prolongation of mipsc decay may be separable. This dissociation may be similar to that reported by the Farb group in which nifedipine prevented GABAinduced increases in intracellular Ca 2+ concentration and GABA A receptor downregulation (Lyons et al., 2001), yet failed to inhibit GABA-induced allosteric uncoupling, a decreased functional linkage between the GABA and BZ binding sites. Changes in GABA/benzodiazepine coupling are dependent on the selectivity and efficacy of the benzodiazepine used for chronic treatment, may be rapid (within 60 hr in vitro and 1-3 days in vivo) and need not involve changes in the numbers of GABA or benzodiazepine binding sites (Primus et al, 1996). In hippocampus, zolpidem binds to α1- α3 subunit-containing GABA A receptors (Sieghart, 1995) which show a large degree of 158

165 GABA-mediated allosteric coupling (Ruano et al., 1992). Since the ability of benzodiazepine site agonists to prolong mipsc decay is related to their ability to increase GABA affinity (Lavoie and Twyman, 1996), a mechanism similar to that underlying uncoupling, and independent of L-type VGCC activity regulation may be operative (Gravielle et al., 2005). Notably, systemic nimodipine effects were distinct from the systemic effects of the benzodiazepine antagonist flumazenil. Namely, following a similar protocol, prior systemic injection of flumazeinil reversed both the reduction in CA1 neuron mipsc amplitude and tolerance to zolpidem s ability to prolong mipsc decay in FZP-treated rats. This suggests that nimodipine antagonism of GABA A receptor -mediated function does not occur via the benzodiazepine binding site, but more likely through interruption of L-type VGCC activation. A reduction in GABA A receptor -mediated mipsc amplitude could reflect a change in channel conductance or channel number (DeKoninck and Mody, 1992; Nusser et al., 1997). Chronic FZP administration is associated with a decrease in CA1 neuron GABA receptor unitary channel conductance (Poisbeau et al., 1997; Zeng and Tietz, 1999), which may be associated with modulation of GABA A receptor subunit composition based on regional changes of subunit mrna and protein expression in rat brain following a variety of benzodiazepine chronic treatments (Kang and Miller, 1991; Primus and Gallager, 1992; Tietz et al., 1994; Zhao et al., 1994; Holt et al., 1996; Impagnetiello et al., 1996; Chen et al., 1999, Tietz et al., 1999). Acute systemic nimodipine injection the day prior to hippocampal slice recording prevented the decrease in CA1 neuron GABA A receptor unitary channel conductance estimated using peak-scaled non-stationary variance analysis. This finding suggests that activation of L-type VGCCs may contribute 159

166 to GABA A receptor single channel conductance changes following chronic FZP administration (Fig 3). Since VGCC-mediated calcium signaling has been shown to regulate gene transcription (Dolmetsch et al., 2001; Weick et al., 2003), the decrease in conductance might also be related to transcriptional repression of GABA A receptor subunit genes similar to GABA-induced GABA A receptor down-regulation (Lyons et al., 2001; Gravielle et al., 2005). Another likely explanation for the observed effects of nimodipine on GABA A receptor dysfunction is the interruption of intracellular Ca 2+ homeostasis through blockade of L-type VGCCs. Intracellular Ca 2+ is an important regulator of GABA A receptor channel function. For instance, Ca 2+ was demonstrated to have a biphasic effects on synaptic GABA A receptor channels (dekoninck and Mody, 1996; Aguayo et al., 1998). When the rapid Ca 2+ chelator 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'- tetraacetic acid (BAPTA) was in the pipette solution, the GABA-activated Cl - current amplitude decreased over time to 49-7% of control (Aguayo et al., 1998). In contrast, equimolar replacement of BAPTA with ethylenebis (oxonitrilo) tetraacetate (EGTA) caused a 10-60% increase in GABA current (Aguayo etal., 1998). These effects could be mediated by a Ca 2+ -dependent phosphorylation/dephosphorylation processes since both Ca 2+ /calmodulin-dependent protein kinase and calcineurin (PP2B) have been demonstrated to directly regulate GABA A receptor-mediated Cl - current (Wang et al., 1995; Churn and DeLorenzo et al., 1998; Lu et al., 2000; Alix et al., 2002; Wang et al., 2003). It will be of significant interest to further evaluate the role of VGCC-mediated Ca 2+ signaling in the significant reductions in GABA A receptor-mediated mipscs and 160

167 GABA A receptor single channel conductance in CA1 neurons following chronic FZP administration. The excitatory role of GABA A receptor during the early developmental period has been well described to elevate intracellular Ca 2+ concentration through L-type VGCCs (Ganguly et al., 2001; also see reviews Stein and Nicoll, 2003; Marty and Llano, 2005). The electrochemical Cl - gradient determines the excitatory versus inhibitory nature of GABA A receptor activation. However, more recent studies have shown a GABA A receptor-mediated depolarization can also occur in mature neurons (Voipio and Kaila, 2000, Isomura et al., 2003). In mature hippocampal CA1 neurons, a single stimulation of GABAergic inputs is inhibitory, whereas high-frequency trains can facilitate Cl - accumulation rendering GABAergic synapses depolarizing (Voipio and Kaila, 2000, Isomura et al., 2003). Several studies have also shown that an increased intracellular calcium levels through L-type VGCCs follows prolonged GABA A receptor activation, which leads to bicarbonate-driven Cl - entry and Cl - accumulation and eventually promotes cell depolarization (Reichling et al, 1994; Lyons et al., 2001; Chavas et al, 2004; Marty and Llano, 2005). A similar situation likely occurs with prolonged GABA A receptor activation by chronic exposure to benzodiazepines. As a consequence of FZP administration numerous time-dependent changes occur at the GABA A receptor, some of which may influence CA1 neuron hyperexcitability (Van Sickle et al., 2004), including a bicarbonate-driven Cl - accumulation reflected in a shift in the Cl - reversal potential and the appearance of a bicuculline-sensitive depolarizing potential (Zeng et al, 1995; Zeng and Tietz, 1997; Zeng and Tietz, 2000). Moreover, spectral analysis of hippocampal electrical activity during withdrawal from 1-week FZP reveals increased power of a 7Hz 161

168 (theta) peak (Poisbeau et al., 1997) suggested to be associated with GABAergic postsynaptic depolarization and a shift of reversal potential from Cl - toward HCO - 3 (Sun et al., 2001). These potential depolarizing driving forces associated with chronic FZP administration could promote the activation of L-type VGCCs and down-stream Ca 2+mediated signal transductions, including GABA A receptor synaptic changes. Dihydropiridines (DHPs) are highly selective calcium channel blockers which bind to rat, guinea pig and human brain membranes with high affinity (less than 1 nm) and concentration as low as nm were reported to block inward Ca 2+ currents (Scriabine et al, 1989). DHPs can block a variety of non-l-type calcium channels, including GABA A receptors, nicotinic acetylcholine receptors and 5-HT3A receptors (Hargreaves et al, 1996; Houlihan et al, 2000; Das et al, 2004) at IC 50 s in the micromolar to milimolar range. A variety of calcium channel antagonists were also reported to inhibit recombinant α1β2γ2 GABA A receptors (Das et al., 2004), identified by immunoprecipitation studies to be the dominant hippocampal GABA receptor subtype (McKernan and Whiting, 1996). A similar concentration of nimodipine (> 30 µm) directly inhibited CA1 neuron mipsc amplitude and prolonged mipsc decay (Fig. 4). Nevertheless, nimodipine pharmacokinetic studies indicated that nimodipine reaches maximal concentration in the brain in 15 to 30 minutes after injection and nearly clears from rat plasma within 8 hours (Suwelack et al., 1985; Scherling et al., 1991). Therefore, a negligible amount of nimpodipine is expected in rat brain 24 hours following injection and GABA A receptor-mediated mipscs in CA1 neurons were not likely influenced by residual nimodipine in hippocampal slices. In the present study following an acute injection nimodipine reached peak brain levels (<2 μm) within 45 minutes and was 162

169 negligible 2 and 24 hrs later. Since the peak nimodipine brain concentration (~1 μm) achieved is far less than that required (30 µm) to directly inhibit GABAR-mediated mipscs (Fig. 4), GABARs were not likely directly influenced by residual nimodipine in vivo or in vitro. Since in brain the density of neuronal L-type Ca 2+ channels are significantly higher than that of vascular ones (Ricci et al, 2002), nimodipine is expected to have a more pronounced neuronal than vascular effects. Considering the much higher affinity and efficacy of nimodipine for neuronal L-type VGCCs than other channels (nm vs. µm to mm), impairment of L-type VGCC mediated-signaling pathways is the most reasonable explanation for nimodipine s effects to avert the decrease in both CA1 neuron mipsc current amplitude and unitary channel conductance. In summary, the present study provides evidence that activation of L-type VGCCs may play a role in mediating the reduction in mipsc amplitude in hippocampal CA1 pyramidal neurons following chronic FZP administration. The reduction in mipsc amplitude likely reflects a reduction in GABA A receptor single channel conductance. Since GABA A receptor dysfunction has been associated with the regulation of certain GABA A receptor subunit proteins. L-type VGCCs mediated changes in Ca 2+ signaling may be associated with changes in GABA A receptor subunit composition. In contrast, in vitro tolerance to zolpidem s prolongation of mipsc decay was independent of L-type VGCC activation, and most likely dependent on a change in zolpidem potency related to that involved in GABA/benzodiazepine site uncoupling. Together these findings suggest that VGCC-mediated Ca 2+ signaling may play a significant role in mediating functional 163

170 GABA system plasticity associated with chronic activation of GABA A receptors during prolonged benzodiazepines administration. REFERENCES Aguayo LG, Espinosa F, Kunos G, Satin LS (1998). Effects of intracellular calcium on GABA A receptors in mouse cortical neurons. Pflüger Arch 435: Alix P, Grolleau F, Hue B (2002). Ca 2+ /calmodulin-dependent protein kinase regulates GABA-activated Cl - current in cockroach dorsal unpaired median neurons. J Neurophysiol 87: Barnes EM (1996). Use-dependent regulation of GABA A receptors. Int Rev Neurobiol 39: Chen S, Huang X, Zeng XJ, Sieghart W, Tietz EI (1999). Benzodiazepine-mediated regulation of α1, α2, β1-3 and γ2 GABA A receptor subunit proteins in the rat brain hippocampus and cortex. Neuroscience, 93: Churn SB and DeLorenzo RJ (1998). Modulation of GABAergic receptor function is regulated by phosphorylation in acutely dissociated guinea pig hippocampal neurons. J Physiol (Lond) 420: Costa E, Auta J, Grayson DR, Matsumoto K, Pappas GD, Zhang X and Guidotti A. (2002) GABA A receptors and benzodiazepines: a role for dendritic resident subunit mrnas. Neuropharmacology, 43: Das P, Bell-Horner CL, Huang RQ, Raut A, Gonzales EB, Chen ZL, Covey DF and Dillon GH (2004). Inhibition of type A GABA receptors by L-type calcium channel blockers. Neuroscience 124:

171 De Koninck Y, Mody I (1992). Noise analysis of miniature IPSCs in adult rat brain slices: properties and modulation of synaptic GABA A receptor channels. J Neurophysiol 71: Gallager DW, Marley RJ, Hernandez TD. (1991). Biochemical and electrophysiological mechanisms underlying benzodiazepine tolerance and dependence. Pratt J, editor. New York: Academic Press: pp Griffiths RR. and Weerts EM. (1997) Benzodiazepine self-administration in humans and laboratory animals-implications for problems of long-term use and abuse. Psychopharmacology 134: Gravielle MC, Faris R, Russek SJ, Farb DH (2005). GABA induces activity dependent delayed-onset uncoupling of GABA/Benzodiazepine site interactions in neocortical neurons. J. Biol. Chem., 280: Heninger SN, Tallman JF, Garret KM, Vitek MP, Duman RS, GAllager DW (1990). Effects of continuous diazepam administration on GABA A subunit mrna in rat brain. J Mol Neurosci. 2: Hutchinson MA, Smith PF, Darlington CL (1996). The behavioral and neuronal effects of the chronic administration of benzodiazepine anxiolytic and hypnotic drugs. Prog Neurobiol. 49: Impagnatiello F, Pesold C, Longone P, Carunchol H, Fritschy JM, Costa E, Guidotti A (1996). Modification of γ-amino butyric acid type A receptor subunit expression in rat neocortex during tolerance to diazepam. Mol Pharmacol 49:

172 Itier V, Granger P, Perrault C, Depoortere H, Scatton B, Avenet P (1996). Protracted treatment with diazepam reduces benzodiazepine receptor-mediated potentiation of γ- amino butyric acid-induced currents in dissociated rat hippocampal neurons. J Pharmacol Exp Ther 279: Koninck KD and Mody I (1996). The effects of rasing intracellular calcium on synaptic GABA A receptor-channels. Neuropharmacology 35: Lavoie AM, Twyman RE (1996). Direct evidence for diazepam modulation of GABAA receptor microscopic affinity. Neuropharmacology. 35(9-10): Lyons HR, Gibbs TT, Farb DH (2000). Turnover and down-regulation of GABA(A) receptor alpha1, beta2s, and gamma1 subunit mrnas by neurons in culture. J Neurochem. 74 (3): Lyons HR, Land MB, Gibbs TT, Farb DH (2001). Distinct signal transduction pathways for GABA-induced GABA(A) receptor down-regulation and uncoupling in neuronal culture: a role for voltage-gated calcium channels. J Neurochem. 78(5): Macdonald RL and Olsen RW (1994) GABA A receptor channels. Annu Rev Neurosci. 17: Marty A and Llano I (2005). Excitatory effects of GABA in established brain networks. Trends Neurosci. 28(6): Otis TS. Mody I (1992). Modulation of decay kinetics and frequency of GABA A receptor-mediated spontaneous hippocampal neurons. Neuroscience 49: Pesold C, Caruncho HJ, Impagnatiello F, Berg MJ, Fritschy JM, Guidotti A and Costa E (1997). Tolerance to diazepam and changes in GABA A receptor subunit expression in rat neocortical areas. Neuroscience, 79:

173 Poisbeau P, Williams SR, Mody I (1997). Silent GABAA synapses during flurazepam withdrawal are region-specific in the hippocampal formation. J Neurosci. 17(10): Ruano D, Vizuete M, Cano J, Machado A, Vitorica J (1992). Heterogeneity in the allosteric interaction between the gamma-aminobutyric acid (GABA) binding site and three different benzodiazepine binding sites of the GABAA/benzodiazepine receptor complex in the rat nervous system. J Neurochem. 58(2): Russek SJ, Bandyopadhyay S, Farb DH (2000). An initiator element mediates autologous downregulation of the human type A gamma -aminobutyric acid receptor beta 1 subunit gene. Proc Natl Acad Sci U S A. 97(15): Sieghart W (1995). Structure and pharmacology of γ-amino butyric acid type A receptor subtypes. Pharmacol Rev 47: Sun MK, Zhao WQ, Nelson TJ, Alkon DL. Theta rhythm of hippocampal CA1 neuron activity: gating by GABAergic synaptic depolarization (2001). J Neurophysiol. 85(1): Tietz EI, Huang X, Weng X, Rosenberg HC, Chiu TH (1994). Expression of α1, α5, and γ2 GABA A receptor subunit mrnas measured in situ in rat hippocampus and cortex following chronic flurazepam administration. J Mol Neurosci 4: Wang RA. Cheng G, Kolal M, Randic M (1995). α-subunit of calcium/calmodulindependent protein kinase II enhances γ-aminobutyric acid and inhibitory synaptic responses of rat neurons in vitro. J Neurophysiol. 73: Weiland GA, Oswald RE (1985). The mechanism of binding of dihydropyridine calcium channel blockers to rat brain membranes. J. Biol. Chem. 260 (14):

174 Xie XH, Tietz EI (1992). Reduction in potency of selective gamma-aminobutyric acid A agonists and diazepam in CA1 region of in vitro hippocampal slices from chronic flurazepam-treated rats. J Pharmacol Exp Ther 262: Zeng XJ and Tietz EI (1999). Benzodiazepine tolerance at GABAergic synapses on hippocampal CA1 pyramidal cells. Synapses 31: Zhao T, Chiu TH, Rosenberg HC (1994). Decreased expression of (-amino butyric acid type A/benzodiazepine receptor β subunit mrna in flurazepam-tolerant rats. J Mol Neurosci 5:

175 CON/VEH FZP/VEH CON/NIM FZP/NIM 15 ms 20 pa mipsc Amplitude (-pa) * CON FZP CON FZP Vehicle Nimodipine 169

176 Fig. 1. Prior nimodipine injection reverses the decrease in GABA A R-mediated mipscs in CA1 neurons in hippocampal slices from 2-day FZP-withdrawn rats. Rats were injected systemically with the L-type VGCC antagonist nimodipine (10 mg/kg, i.p.) or vehicle (0.5% Tween-80, 2 ml/kg, i.p.) 24 h before hippocampal slice preparation. 1 day later hippocampal slices were prepared from 2-day FZP-withdrawn rats followed by electrophysiology recordings. A) Representative mipsc current traces isolated from 2-day FZP-withdrawn (FZP) and control neurons (CON) injected with nimpodipine (NIM) or vehicle (VEH). B) Average amplitudes of GABA A receptor-mediated mipscs (V H = -70 mv) in CA1 neurons isolated from control rats (solid bars) and 2-day FZPwithdrawn rats (open bars) following either vehicle or nimodipine injection were compared. As previously reported (Zeng and Tietz, 1999) there was a significant decrease in mipsc amplitude (43.0%, *p<0.05) in CA1 neurons from 2-day FZP-withdrawn rats (n=8) in comparison to those isolated from control rats (n=9). However, there was no difference in mipsc amplitude in CA1 neurons from FZP withdrawn rats (p>0.05, n=7) given a systemic nimodipine injection 24 hr prior to recording, compared to those isolated from the matched control group (n=8), as shown in table

177 CON-VEH FZP-VEH 1 μm Zolpidem 1 μm Zolpidem 50 ms 5 pa CON-NIM FZP-NIM 1 μm Zolpidem 1 μm Zolpidem % of Control mipsc Decay * * CON FZP CON FZP Vehicle Nimodipine 171

178 Fig 2. Prior nimodipine injection failed to reverse the tolerance to zolpidem s ability to prolong decay of GABA A receptor-mediated mipscs in CA1 neurons in hippocampal slices from 2-day FZP-withdrawn rats. A). Representative mipsc traces, averaged from 130 to 170 miniature events, before and after superfusion of 1 µm zolpidem. CA1 neurons were recorded from control (CON) and FZP-treated (FZP) rats systemically injected with vehicle or nimodipine 24 h before recording. The peak amplitude of averaged mipscs after zolpidem superfusion was normalized to the peak amplitude of averaged mipscs before zolpidem application for comparison. B). Effects of 1 µm zolpidem to prolong mipsc decay in CA1 neurons was expressed as a percentage of the baseline average mipsc decay recorded in the absence of zolpidem. Zolpidem tolerance was measured as a significant reduction in the ability of zolpidem to enhance mipsc decay in FZP-VEH neurons (* p<0.05, n=7) in comparison to CON-VEH neurons (n=6). Prior systemic nimodipine injection failed to prevent zolpidem tolerance in FZP-NIM neurons (* p<0.05, n=7) in comparison to CON-NIM neurons (n=6), as shown in Table

179 20 20 Variance(σ 2 ) 10 Variance(σ 2 ) Amplitude (pa) Amplitude (pa) Unitary Conductance (ps) * CON FZP CON FZP Vehicle Nimodipine Channel Number CON FZP CON FZP Vehicle Nimodipine 173

180 Fig. 3 Effects of nimodipine injection on peak-scaled non-stationary variance analysis of mipscs recorded in CA1 neurons in hippocampal slices from 2-day FZP-withdrawn rats. A). Representative amplitude-variance plots following peak-scaled non-stationary variance analysis of mipscs in CA1 neurons from control or FZP-withdrawn rats after systemic vehicle or nimodipine injection. Data were fitted with the equation σ 2 =ii-i 2 /N, where σ 2 is the variance, I is the mean current, N is the number of channels activated at the peak of the mean current, i is the single-channel current. B). Prior nimodipine injection prevents the decrease in GABA A receptor single channel conductance in CA1 neurons in hippocampal slices from 2-day FZP-withdrawn rats. Unitary channel conductance (γ) was derived from γ=i/v, where V is the driving force (V H = -70 mv, E REV = 0 mv). As previously reported (Zeng and Tietz, 1999) there was a significant decrease in GABA A receptor single channel conductance (~38.0%, *p<0.05) in CA1 neurons from 2-day FZP-withdrawn rats (n=9) in comparison to those recorded control slices (n=8). However, there was no difference in GABA A receptor single channel conductance in CA1 neurons from FZP withdrawn rats (p>0.05, n=7) given a systemic nimodipine injection 24 hr prior to recording, compared to those isolated from the matched control group (n=8). C). There was no significant difference in the mean channel number (N) between control and FZP-withdrawn neurons. Nimodipine injection had no effect on channel number in either group (p>0.05). 174

181 mipsc Amplitude (-pa) nimo vehicle Baseline Concentration ( μm) τ (ms) nimo vehicle Baseline Concentration (μm) 175

182 Fig 4. Effects of in vitro nimodipine superfusion on GABA A receptor-mediated mipscs. Mean concentration-response profile for the effect of nimodipine on CA1 neuron GABA A receptor-mediated mipscs. A). Average GABA A receptor mipsc amplitude (V H = -70 mv) during vehicle (close circles) or nimodipine (open circles) superfusion of hippocampal slices. Baseline mipsc amplitude was recorded for 5 min without nimodipine. In comparison to vehicle superfusion (n=3), nimodipine decreased mipsc amplitude in concentration-dependent superfusion (n=4) at concentrations ranging from 10 µm to 300 µm. There were no changes in mepsc frequency or rise time during nimodipine perfusion. B). Nimodipine also produced a concentration-dependent prolongation of mipsc decay at concentrations higher than 50 µm. 176

183 Table 1. Effects of nimodipine on CA1 pyramidal cell mipscs Group CON-VEH (n=9) FZP-VEH (n=8) mipscs Group Amplitude τ (-pa) (ms) 22.68± ±1.39 CON-VEH (n=6) 12.86± ±1.37 FZP-VEH (n=7) Zolpidem Potentiation Amplitude Τ (%pa) (%ms) ± 3.1 p= ± 2.7 p= ±5.42 * p= ±3.88 p=0.73 CON-NIM (n=8) 22.82± ±2.22 CON-NIM (n=6) ± 6.4 p= ±3.03 * p=0.04 FZP-NIM (n=7) 23.47± ±1.53 FZP-NIM (n=7) ± ±2.66 * p=0.91 p=0.20 P value * p<0.001 P>0.05 p>0.05 * p<0.001 Values are means ± S.E.; VEH: i.p. injection of vehicle; NIM: i.p. injection of nimodipine. p values indicated in the last row is between groups analyzed by one-way ANOVA. Zolpidem potentiation (%) is expressed as a percentage of the baseline value in the absence of zolpidem. p value indicated in each group is from comparison between absence and presence of zolpidem. 177

184 DISCUSSION/SUMMARY Benzodiazepines, widely prescribed for the treatment of anxiety and insomnia, are limited by their clinical side-effects including tolerance, a decrease in the ability of the drug to produce the same degree of pharmacological effect during prolonged exposure, and dependence, in which discontinuation of the drug following repeated exposure generates a characteristic withdrawal syndrome. Benzodiazepines are a group of positive allosteric modulators of GABA A receptors. Therefore, changes in GABA A receptor function and expression have been the primary focus for seeking mechanisms to explain benzodiazepine tolerance and dependence. Changes in GABA A receptor structure, function, and pharmacology associated with prolonged benzodiazepine exposure include region-specific decreases in the number of benzodiazepine binding sites, reduced GABA agonist efficacy, consistent reductions in allosteric coupling between GABA and benzodiazepine binding sites, modulation of GABA A receptor subunit mrna and protein expression (Heninger et al., 1990; Gallager et al., 1991; Zhao et al., 1994; Barnes, 1996; Hutchinson et al., 1996; Impagnatiello et al., 1996; Pesold et al., 1997; Costa et al., 2002). Indeed, following 1-week oral FZP treatment tolerance to benzodiazepine anticonvulsant actions has been demonstrated in rat in vivo (Tietz et al, 1999) as well as significant impairment of GABA A receptor-mediated synaptic transmission in CA1 pyramidal neurons in in vitro hippocampal slices (Xie and Tietz, 1992; Zeng et al., 1995; Zeng and Tietz, 1999). However, the precise mechanism underlying GABA A receptor changes in benzodiazepine tolerance and dependence still remains inconclusive. 178

185 Recent reports suggest that different neural mechanisms mediated by distinct neural pathways might underlie the different components of the benzodiazepine withdrawal syndrome (Allison et al, 1999; Podhorna 2002; Allison and Pratt, 2003; Millan 2003; Van Sickle et al, 2004; Ganouni et al, 2004). Glutamate receptors, mainly AMPA receptors and NMDA receptors, attract a lot of attention, in part because of the close neuroanatomical interrelationship between GABAergic and glutamatergic neurons in many brain regions and their interactions in relation to mediating neuronal adaptive processes and synaptic plasticity. Moreover, their involvement mediating dependence to a variety of drugs of abuse, including many CNS depressants, make glutamate receptor mechanisms potential candidates underlying benzodiazepine dependence as well (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999; Podhorna 2002; Ganouni et al, 2004). New evidence has indicated that L-type VGCCs may play an important role in benzodiazepine tolerance and dependence. L-type VGCCs have been shown to be involved in mediating withdrawal signs to a variety of CNS depressants (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999; Podhorna 2002; Ganouni et al, 2004). L-type VGCC activity has also been implicated in activity-dependent synaptic plasticity (Morgan and Teyler, 1999; Rajadhyksha and Kosofsky, 2005). Similar phenomena have been observed to regulate GABA A receptor, AMPA receptor and NMDA receptor and functional plasticity in hippocampal CA1 pyramidal neurons (Zeng and Tietz, 1999; Van sickle and Tietz, 2002; Van Sickle et al., 2002). 179

186 Thus, using the model of oral 1-week FZP administration in rats and synaptic changes of GABA A receptor, AMPA receptor and NMDA receptor function in rat hippocampal CA1 pyramidal neurons, the goals of these dissertation studies were to : 1) evaluate the role of L-type VGCC activity in glutamate receptor synaptic plasticity in CA1 neurons associated with FZP-withdrawal anxiety in rats; 2) understand what role AMPA receptor plasticity may have to modify NMDA receptor function; 3) examine the functional regulation of L-type VGCCs in CA1 neurons during and following chronic FZP administration in rats and 4) to investigate the role of L-type VGCCs in mediating synaptic GABA A receptor changes in CA1 neurons following chronic FZP administration. Temporal patterns of both AMPA receptor and NMDA receptor functional regulation were demonstrated and paralleled anxiety-like behavior in rats, measured in the elevated plus-maze, during withdrawal from 1-week oral FZP treatment. More specifically, increased AMPA receptor-mediated mepsc amplitude was demonstrated in rats exhibiting anxiety 1 day after FZP withdrawal. The next day, while AMPA receptormediated mepsc amplitude was still increased, NMDA receptor-mediated eepsc amplitude was decreased, and rats did not show anxiety-like behavior (Van Sickle et al., 2004). It was hypothesized that CA1 neuron increased AMPA receptor function contributes to FZP-withdrawal anxiety in rats, while decreased NMDA receptor function counter-balanced the increased CA1 neuron AMPA receptor function, masking the anxiety the next day (Van Sickle et al., 2004). This was supported by evidence that prior systemic MK-801 injection prevented the decreased NMDA receptor function leaving 180

187 increased AMPA receptor function unchanged, thus unmasking anxiety-like behavior in 2-day FZP-withdrawn rats (Van Sickle et al., 2004). However, several questions remain to be answered concerning correlations between modulation of ion channel function in CA1 neurons and rat behavioral changes. The first question addressed was whether ion channel functional changes produce significant changes at the neuronal level. In other words, was CA1 neuron excitability altered in 1-day FZP-withdrawn rats thus the output signal to other brain areas, and eventually the behavioral manifestation in whole animal. To address this question, CA1 neuron hyperexcitability was examined during superfusion with 4-aminopyridine (4-AP) in hippocampal slices from 1-day FZP withdrawn rats. More frequent discharges were recorded in the CA1 region supporting the hypothesis that the hippocampus is an important locus associated with BZ withdrawal symptoms. The next question concerned the relationship between CA1 neuron AMPA receptor functional upregulation and anxiety behavior in rats. Regulation of AMPA receptor function was antagonized by systemic GYKI injection the day prior to electrophysiological recording. FZP-withdrawal anxiety was measured in vivo and CA1 neuron AMPA receptor-mediated mepscs were examined in vitro after prior GYKI injection. A significant positive correlation was detected between increased CA1 neuron AMPA receptor-mediated mepsc amplitude and anxiety measured in rats in the elevated-plus maze; GYKI (0.5 mg/kg, i.p.) abolished this correlation. This study further strengthened a contributory role for hippocampal CA1 neuron AMPA receptor- 181

188 mediated excitation in mediating benzodiazepine withdrawal symptoms. The finding is also consistent with other evidence of neuroadaptive changes in the glutamatergic system associated with benzodiazepine withdrawal (Stephens, 1995; Izzo et al, 2001; Allison and Pratt, 2003). Many studies support an important role for non-nmdar glutamatergic mechanisms in drug dependence underlying dependence on ethanol (Molleman and Little, 1995; Sanchis-Segura et al, 2006) and morphine (Vekovischeva et al, 2001; Zhong et al, 2006). An increased glutamatergic strength in mesolimbic reward pathways has also been implicated in addictive behaviors (Tzschentke and Schmidt, 2003; Nestler, 2005). Since down-regulation of NMDA receptor function was hypothesized to counterbalance AMPA receptor functional upregulation and mask the appearance of anxiety in 2-day FZP-withdrawn rats, and given the close anatomical and physiological co-regulation of these two receptor families (Watt AJ et al, 2000; Watt AJ et al, 2004; Perez-Otano and Ehlers, 2005), the next question addressed was whether NMDA receptors were down-regulated following up-regulation of AMPA receptor function. Based on the prior observation that AMPA receptor functional upregulation preceded NMDAR functional downregulation it was postulated that downregulation of NMDA receptor function may play a preventive role to protect neuronal over-excitation. To test this hypothesis, GYKI was injected immediately after FZP withdrawal and NMDA receptor-mediated eepscs, as well as AMPA receptor-mediated mepscs were examined 2 days after FZP withdrawal. The observation that down-regulation of CA1 neuron NMDA receptors in 2-day FZP-withdrawn rats was averted by prior AMPA receptor antagonist injection suggested that downregulation of NMDA receptor-mediated 182

189 function may be secondary to enhanced AMPA receptor function. Moreover, these findings suggest that NMDA receptor regulation may play a compensatory, feedback role to prevent CA1 neuron hyperexcitability, thus dampening hippocampal output activity and preventing benzodiazepine-induced withdrawal anxiety. These studies support a neuroadaptive mechanism by which the imbalance in excitatory glutmatergic receptors during benzodiazepine withdrawal is initially AMPA receptor-dependent and is subsequently counterbalanced by down-regulation of CA1 neuron NMDA receptor function, represented by a reduction in both NMDA receptor-mediated eepsc amplitude and the levels of NR2B mrna and protein in the CA1 region (Van Sickle et al, 2002). To look for the potential mechanism that may underlie AMPA receptor functional changes, and considering the well-described roles of NMDA receptors and L-type VGCCs in mediating Ca 2+ signaling associated with activity-dependent synaptic plasticity, AMPA receptor mepscs were evaluated following systemic treatment, with MK801 and nimodipine, respectively (Shi et al, 1999; Malinow 2003; Morgan and Teyler, 1999; Rajadhyksha and Kosofsky, 2005). Nimodipine, but not MK801 prevented the increase in AMPA receptor-mediated mepsc amplitude and anxiety in rats, as well as the correlation between these measures. Since nimodipine is a highly-selective calcium channel blocker (Scriabine et al, 1989), and L-type VGCC-dependent Ca 2+ signaling may mediate strengthening of AMPA receptor activity with dependence on other drugs of abuse including ethanol and psychostimulants (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005), it is likely that L-type VGCC mediated Ca 2+ signaling contributed to AMPA receptor functional augmentation. On the basis of 183

190 evidence to date, a model of the proposed cellular mechanisms underlying benzodiazepine withdrawal hyperexcitability in hippocampal CA1 neurons was outlined at the end of first manuscript which provides a framework for the hypotheses that directed additional mechanistic studies described in this dissertation. Numerous investigations have suggested that disturbances of neuronal Ca 2+ homeostasis including enhanced Ca 2+ entry associated with upregulation of L-type VGCCs, is involved in ethanol, barbiturate, morphine and nicotine dependence (Whittington and Little, 1993; Pourmotabbed et al, 1998; Rabbani and Little, 1999, Katsura et al., 2002). Based on the evidence that benzodiazepines can directly inhibit VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Gershon, 1992; Reuveny et al, 1993; Ishizawa Y et al, 1997), the second manuscript postulated that chronic benzodiazepine administration modifies L-type VGCC function and contributes to AMPA receptor plasticity in CA1 pyramidal neurons in benzodiazepine-withdrawn rats. The properties of L-type VGCC-dependent Ca 2+ currents were investigated using whole-cell voltage-clamp techniques in acutely isolated CA1 neurons. The temporal pattern of L-type VGCC regulation indicated an emergent increase in Ca 2+ current density following relatively short periods (3 days) of FZP treatment. Ca 2+ current density progressively increased until a significant, ~2-fold increase was evident upon cessation of 7-day FZP treatment which persisted at least 2 days, but not 4 days, after FZP withdrawal. An acute dose of the FZP active metabolite, desalky-fzp (2.5 mg/kg, p.o.) had no effects on Ca 2+ current density, demonstrating that the enhanced L-type VGCC 184

191 function during the FZP withdrawal period was specific to chronic benzodiazepine exposure. While the mechanism underlying the up-regulation of L-type VGCC function remains to be clearly elucidated, an increase in α1d subunit protein in CA1 mini-slices might well explain the observation of a negative shift in voltage dependence of Ca 2+ current activation and may be associated with increased current density. α1d subtype L- VGCCs open at relatively hyperpolarized membrane potentials, are activated at -55 mv approximately mv more hyperpolarized as compared with α1c subtype L-VGCCs, and mediate subthreshold calcium signaling (Xu and Lipscombe, 2001; Lipscombe et al., 2004). α1d subunit protein levels increase in hippocampal CA1 neurons of aged rats (Vern et al., 2003). In single CA1 neurons, increases in α1d subunit transcript levels were positively correlated with increased VGCC activity (Chen et al., 2000). They also mediate consolidation, but not extinction, of contextually conditioned fear in mice (McKinney and Murphy, 2006). The expression of α1d subunit protein in hippocampal CA1 region of 2-day FZP withdrawn rats is currently under the investigation. Augmentation of L-type VGCC function may have considerable physiological significance. L-type VGCCs have been reported to play a role in various activitydependent models of neuronal plasticity (Morgan and Teyler, 1999; Borroni et al, 2000; Rajadhyksha and Kosofsky, 2005). Activation of neuronal L-type VGCCs triggers a sustained influx of Ca 2+ upon depolarization which, via a diverse soluble messengers and transcription factors, initiates long-term processes related to synaptic plasticity in the 185

192 hippocampus and amygdala (Dolmetsch et al., 2001). In fact, during ethanol withdrawal and following psychostimulant administration, activation of L-type VGCCs can result in the influx of intracellular calcium and downstream activation of Ca 2+ /CaM-activated kinase and phosphatase pathways, eventually involving CREB-induced gene expression, also important to neuronal and experience-dependent plasticity (Groth et al, 2003; Xia and Storm 2005; Nestler, 2005; Rajadhyksha and Kosofsky, 2005). The finding that prior systemic injection of nimodipine prevented the upregulation of CA1 neuron AMPA receptor function in 2-day FZP-withdrawn rats, strengthens the previous finding that nimodipine prevents FZP-withdrawal anxiety in 1-day withdrawn rats by averting the related enhancement of CA1 neuron AMPA receptor currents. This effect is not through a direct interaction between the VGCC antagonist and AMPA receptors, as nimodipine had no direct effect on AMPA receptor-mediated mepscs (Xiang and Tietz, 2007). This finding may also explain the ability of L-type calcium channel blockers like nimodipine, nifedipine and verapamil to block a variety of benzodiazepine withdrawal signs including hyperkinesia, hyperthermia, hyperaggression and audiogenic seizures (Chugh et al, 1992; Gupta et al, 1996; Ganouni et al, 2004). Thus, the accumulated evidence supports the physiological significance of augmentation of L-type VGCC function in mediating glutamatergic strength during benzodiazepine withdrawal, contributing to benzodiazepine dependence. The evidence that benzodiazepines have both concentration- and use-dependent inhibitory effects on L-type VGCC-mediated Ca 2+ current also pointed to another potential mechanism able to explain the modification of L-type VGCC function 186

193 following chronic FZP treatment. In addition to their well-characterized effects on GABA A receptors, benzodiazepines can also inhibit neuronal VGCC-mediated Ca 2+ flux (Taft and DeLorenzo, 1984; Reuveny et al, 1993; Watabe et al., 1993; Ishizawa et al, 1997). Oral flurazepam treatment and other common diazepam treatments result in rat brain levels of benzodiazepine metabolites in the low micromolar range (0.6 μm in diazepam equivalents) (Gallager et al, 1985; Xie and Tietz, 1992). The potency of FZP to concentration-dependently inhibit Ca 2+ channel currents in hippocampal cultures was significantly greater when Ca 2+ channels were activated by a prior depolarizing train suggesting that benzodiazepine inhibition of VGCC-mediated currents is use-dependent, as previously shown with chlordiazepoxide (Reuveny et al, 1993). Thus, the low micromolar benzodiazepine brain levels attained during chronic treatment might be expected to modulate VGCCs. At least some benzodiazepines clinical actions may be mediated by L-VGCCs, in addition to their primary action on GABA A receptors. Based on evidence of L-type VGCC functional augmentation following chronic benzodiazepine treatment, and its role in mediating activity-driven synaptic plasticity, including AMPA receptor functional plasticity in hippocampal CA1 neurons associated with withdrawal symptoms, the third manuscript explored the role of L-type VGCC in mediating GABA A receptor synaptic plasticity following chronic FZP administration. Evidence of participation of L-type VGCCs in mediating GABAergic system changes following chronic benzodiazepine administration is less abundant. L-type VGCCs have been long tied to modulation of gene transcription (Dolmetsch et al., 2001; Weick et al., 2003). Nifedipine was reported to inhibit both GABA-induced increases in intracellular 187

194 Ca 2+ concentration and GABA A receptor down-regulation, likely through gene transcriptional gene repression (Lyons et al., 2001). Past benzodiazepine tolerance and dependence studies have demonstrated correlations between modulation of expression of several GABA A receptor subunit mrnas and protein and GABA A receptor functional changes (Kang and Miller, 1991; Primus and Gallager, 1992; Tietz et al., 1994; Zhao et al., 1994; Holt et al., 1996). The finding that prior systemic nimodipine injection averted the decreased GABA A receptor-mediated mipsc amplitude, suggested that activation of L-type VGCCs, likely through modulation of gene transcription, may contribute to the reduced GABAergic function in hippocampal CA1 neurons. The result that nimodipine failed to prevent the decreased ability of zolpidem to prolong the mipsc decay, even though it averted the reduction of mipsc amplitude, following chronic FZP administration is intriguing. This may suggest that L-type VGCC may mediate altered GABA A receptor function following chronic benzodiazepine administration but is not correlated with benzodiazepine tolerance either in vitro or in vivo. This is distinct from the effects of the benzodiazepine antagonist flumazenil. Prior injection of flumazenil, following a similar protocol as nimodipine administration, reversed both the decreased GABA A receptor-mediated mipsc amplitude and tolerance to zolpidem s ability to prolong mipsc decay (Tietz et al. 1999). This suggests that nimodipine antagonism in GABA A receptor-mediated function is not through direct antagonism of the benzodiazepine binding sites, but more likely through interruption of L-type VGCC activation. Since the ability of zolpidem to enhance mipsc decay is related to an increase in GABA affinity a mechanism similar to that responsible for a 188

195 decreased allosteric coupling between the GABA and BZ binding sites might be involved. Uncoupling between GABA and benzodiazepine binding sites has been suggested to be independent of L-type VGCC activity and likely a result of a posttranscriptional regulatory mechanism (Gravielle MC et al., 2005). Uncoupling was observed in our 7-day FZP treatment protocol (Chen et al, 1995; Tietz, personal communication). It would be of significant interest to evaluate if nimodipine could reverse the uncoupling phenomenon in 2-day FZP-withdrawn rats. Systemic nimodipine injection also reversed the decreased GABA A receptor unitary channel conductance estimated using peak-scaled non-stationary variance analysis in CA1 neurons in hippocampal slices from 2-day FZP-withdrawn rats. This finding suggested that enhanced L-type VGCC function may also contribute to decreases in GABA A receptor single channel conductance following chronic FZP administration. Thus L-type VGCC-dependent signaling may modulate GABA A receptor subunit composition or number, as GABA A receptor subunit mrna and protein expression were changed following chronic FZP administration. It would be helpful in future studies to investigate whether nimodipine administration could prevent changes in GABA A receptor subunit mrna and protein expression in the CA1 region of FZP withdrawn rats. It would also be of interest to evaluate the role of VGCC-mediated changes in intracellular Ca 2+ in mediating CA1 neurons mipsc amplitude and GABA A receptor single channel conductance following chronic FZP administration. Intracellular Ca 2+ homeostasis is important to the regulation GABA A receptor channel kinetics and both Ca 2+ /calmodulindependent protein kinase and calcineurin have been demonstrated to directly regulate 189

196 GABA A receptor-mediated Cl - current (Wang et al., 1995; dekoninck and Mody, 1996; Aguayo et al., 1998; Churn and DeLorenzo et al., 1998; Lu et al., 2000; Alix et al., 2002; Wang et al., 2003). Thus L-type VGCC-mediated Ca 2+ homeostasis could mediate GABA A receptor single channel conductance, which may underly GABA A receptor functional changes in CA1 neurons following chronic FZP administration. High concentrations of nimodipine (> 30 µm) can directly inhibit GABA A receptor-mediated mipsc amplitude and prolong mipsc decay but have no effect on AMPA receptor-mediated mepscs. Based on previous nimodipine pharmacokinetic studies in animals (Suwelack et al., 1985; Scherling et al., 1991) the nimodipine levels were expected to be negligible 24 hr after injection. In fact, radioligand binding studies confirmed that the peak concentration in brain (~1 μm) was achieved within 45 min after injection of 10 mg/kg, i.p. and was negligible 2 hours thereafter. Thus, the nimodipine concentration was far less than that required for direct inhibition of GABA A receptors. Considering the much higher affinity and efficacy of nimodipine on L-type VGCCs than other channels (nm vs. µm to mm), impairment of L-type VGCC mediated-signaling pathways is the most reasonable explanation for nimodipine effects on AMPA receptor and GABA A receptors. A potential approach to confirm this explanation would be the use of an L-type VGCC agonist to generate an opposing effect of nimodipine on AMPA receptor and GABAR function associated with benzodiazepine tolerance and dependence. Another possible study would be to confirm whether an L-type VGCC-mediated downstream signaling pathway, such as CaMKII activity or gene transcription, directly regulates AMPA receptor and GABA A receptor function. 190

197 Another important question which needs to be addressed as a result of the findings of these dissertation studies is how the L-type VGCCs might be activated following chronic benzodiazepine administration. Currently the most reasonable explanation would be GABA A receptor-dependent depolarization. As discussed in all three manuscripts, the excitatory role of GABA A receptor during the early developmental period to elevate intracellular Ca 2+ concentration through L-type VGCCs has been well described (Ganguly et al., 2001; also see reviews Stein and Nicoll, 2003; Marty and Llano 2005). The Cl - electrochemical gradient, which depends on the intra- and extracellular Cl - concentration, determines the excitatory versus inhibitory nature of GABA A receptors. However, newer studies have shown that a GABA A receptor-mediated depolarization can also occur in mature neurons. In mature hippocampal pyramidal CA1 neurons, while a single stimulation of GABAergic inputs is inhibitory, high-frequency trains can switch GABAergic synapses to from hyperpolarizing to depolarizing due to Cl - accumulation (Voipio and Kaila, 2000, Isomura et al., 2003). Several other studies have shown that an increased concentration of intracellular calcium through L-type VGCCs following prolonged GABA A receptor activation can lead to bicarbonate-driven Cl - entry and Cl - accumulation and eventually promotes cell depolarization (Reichling et al, 1994; Lyons et al., 2001; Chavas et al, 2004; Marty and Llano, 2005). A similar situation likely occurs with prolonged GABA A receptor activation by chronic exposure of benzodiazepines. As a consequence, during FZP administration numerous timedependent changes occur at the GABA A receptor, some of which influence CA1 neuron hyperexcitability (Van Sickle et al., 2004) including a bicarbonate-driven Cl - accumulation reflected in a shift in the Cl - reversal potential and the appearance of a 191

198 bicuculline-sensitive depolarizing potential (Zeng et al, 1995; Zeng and Tietz, 1997; Zeng and Tietz, 2000). Moreover, spectral analysis of hippocampal electrical activity during withdrawal from 1-week FZP revealed increased power of a 7Hz (theta) peak (Poisbeau et al., 1997) that was suggested to be associated with GABAergic postsynaptic depolarization and a shift of reversal potential from Cl - toward HCO - 3 (Zeng et al, 1995; Zeng and Tietz, 1997; Sun et al., 2001). These potential depolarizing driving forces following chronic FZP administration could promote the activation of L-type VGCCs and down-stream Ca 2+- mediated signal transduction. In summary, these dissertation studies continued to explore the mechanisms underlying glutamate receptor synaptic plasticity in hippocampal CA1 pyramidal neurons associated with BZ withdrawal anxiety in rats, based on previous studies in the lab. In the first manuscript, a role for L-type VGCCs in mediating glutamate receptor functional changes was identified. The second manuscript studied the functional regulation of L- type VGCC in hippocampal CA1 pyramidal neurons following chronic FZP administration. This study indicated a transient functional enhancement of L-type VGCCs associated with chronic FZP administration, parallel to the window in which other glutamate receptors and GABA A receptor changes were demonstrated previously (Zeng and Tietz, 1999; Van Sickle and Tietz, 2002; Van Sickle et al., 2004). Thus this study provided evidence to suggest the enhanced Ca 2+ influx through L-type VGCC during chronic FZP administration may contribute to ion channel plasticity. The third manuscript examined the role of L-type VGCC in mediating GABA A receptor functional changes. The evidence in this study suggests that L-type VGCC contribute at least in part 192

199 to synaptic GABA A receptor changes in CA1 neurons following chronic FZP administration. Together these dissertation studies explored and demonstrated functional changes of L-type VGCC following chronic FZP administration and their roles in mediating glutamate receptor and GABA A receptor synaptic plasticity in hippocampal CA1 pyramidal neurons associated with benzodiazepine tolerance and dependence. 193

200 BIBLIOGRAPHY Andersen P, Bliss TVP, Skrede KK (1971). Unit analysis of hippocampal population spikes. Exp. Brain Res., 13: Aguayo LG, Espinosa F, Kunos G, Satin LS (1998). Effects of intracellular calcium on GABA A receptors in mouse cortical neurons. Pflüger Arch 435: Ahlijanian MK, Westenbroek RE, Catterall WA (1990). Subunit structure and localization of dihydropyridine-sensitive calcium channels in mammalian brain, spinal cord, and retina. Neuron 4: Alix P, Grolleau F, Hue B (2002). Ca 2+ /calmodulin-dependent protein kinase regulates GABA-activated Cl - current in cockroach dorsal unpaired median neurons. J Neurophysiol 87: Allison C, Dunworth SJ, Stephens DN, Pratt JA (1999). Involvement of AMPA-receptors in benzodiazepine withdrawal and sensitization processes. Behav Pharmacol 10 (Suppl. 1): S2. Allison C, Claase L-A, Pratt JA (2002). Diazepam withdrawal-induced anxiety and place aversion in the rat: differential effects of two chronic diazepam treatment regimes. Behav Pharmacol 13: Allison C, Pratt JA (2003). Neuroadaptice processes in GABAergic and glutamatergic systems in benzodiazepine dependence. Pharmacol Ther 98: Amaral DG, Witter MP (1995). Hippocampal formation. In: Paxinos G., ed., The rat nervous system. New York Academic Press:

201 Angelotti TP, Tan F, Macdonald RC (1992). Molecular and electrophysiological characterization of an allelic variant of the rat α 6 GABA A receptor subunit. Mol Brain Res, 16: Ascher P and Nowak L (1988). The role of divalent cations in the N-methyl-D-aspartate responses of mouse central neurons in culture. J Physiol, 399: Avery RB, Johnson D (1996). Multiple channel types contribute to the low-voltageactivated calcium current in hippocampal CA3 pyramidal neurons. J Neurosci 16: Banke TG, Bowie D, Lee H, Huganir RL, Schousboe A, Traynelis SF (2000). Control of GluR1 AMPA receptor function by camp-dependent protein kinase. J Neurosci 20: Barnes EM (1996). Use-dependent regulation of GABA A receptors. Int Rev Neurobiol 39: Barnes EM (2000). Intracellular trafficking of GABA(A) receptors. Life Sci. 66(12): Barria A, Muller D, Derkach V, Griffith LC, Soderling TR (1997). Regulatory phosphorylation of AMPA-type glutamate receptors by CaM-KII during long-term potentiation. Science 276, Bech-Hansen NT, Naylor MJ, Maybaum TA, Pearce WG, Koop B, Fishman GA, Mets M, Musarella MA, Boycott KM (1998). Loss-of-function mutations in a calciumchannel alpha-subunit gene in Xp11.23 cause incomplete X-linked congenital stationary night blindness. Nat Genet 19:

202 Bigge CF (1999). Inotropic glutamate receptors. Curr Opin Chem Biol 3: Birnbaumer L, Campbell KP, Catterall WA, Harpold MM, Hofmann F, Horne WA, Mori Y, Schwartz A, Snutch TP, Tanabe T, et al (1994). The naming of voltage-gated calcium channels. Neuron. 13(3): Blahos J, Wenthold RJ (1996). Relationship between N-methyl-D-asparate receptor NR1 splice variants and NR2 subunits. J Biol Chem, 271: Bliss TV and Lomo T (1973). Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J Physiol, 232: Bliss TV, Collingridge GL (1993). A synaptic model of memory: long-term potentiation in the hippocampus. Nature 361: Boehm J, Kang MG, Johnson RC, Esteban J, Huganir RL, Malinow R (2006). Synaptic incorporation of AMPA receptors during LTP is controlled by a PKC phosphorylation site on GluR1. Neuron. 51(2): Borroni AM, Fichtenholtz H, Woodside BL, Teyler TJ (2000). Role of voltage-dependent calcium channel long-term potentiation (LTP) and NMDA LTP in spatial memory. J Neurosci 20: Boulter J, Hollmann M, O Shea-Greenfield A, Hartley M, Deneris C, Maron C, Heinemann S. (1990). Molecular cloning and functional expression of glutamate receptor subunit genes. Science 249: Brett RR, Pratt JA (1995). Changes in BZ-GABA receptor coupling in an accumbenshabenula circuit after chronic diazepam treatment. Br J Pharmacol 116:

203 Catterall WA (1999). Structure and function of neuronal Ca 2+ channels and their role in neurotransmitter release. Cell Calcium 24: Catterall WA (2000). Structure and regulation of Voltage-gated Calcium channels. Annu. Rev. Cell Dev. Biol. 16: Chan, C.Y. and Farb, D.H. (1985). Modulation of neurotransmitter action: control of the γ-amino butyric acid response through the benzodiazepine receptor. J. Neurosci., 5(9): Chazot PL and Stephenson FA (1997). Molecular dissection of native mammalian forebrain NMDA receptors containing the NR1 C2 exon: direct demonstration of NMDA receptors comprising NR1, NR2A and NR2B subunits within the same complex. J Neurochem 69: Chen L, Chetkovich DM, Petralia RS, Sweeney NT, Kawasaki Y, Wenthold RJ, Bredt DS, Nicoll RA (2000). Stargazin regulates synaptic targeting of AMPA receptors by two distinct mechanisms. Nature 408: Chen S, Cox A, Tietz EI (1995). Temporal regulation of GABA A receptor binding and coupling in flurazepam tolerant rats. Soc. Neurosci. Abstr., 21: Chen S, Huang X, Zeng XJ, Sieghart W, Tietz EI (1999). Benzodiazepine-mediated regulation of α1, α2, β1-3 and γ2 GABA A receptor subunit proteins in the rat brain hippocampus and cortex. Neuroscience, 93: Chugh Y, Saha N, Sankaranarayanan A., Sharma PL (1992). Effect of peripheral administration of cinnarizine and verapamil on the abstinence syndrome in diazepamdependent rats. Psychopharmacology 106:

204 Churchill L, Swanson CJ, Urbina M, Kalivas PW (1999). Repeated cocaine alters glutamate receptor subunit levels in the nucleus accumbens and ventral tegmental area of rats that develop behavioral sensitization. J Neurochem 72: Churn SB and DeLorenzo RJ (1998). Modulation of GABAergic receptor function is regulated by phosphorylation in acutely dissociated guinea pig hippocampal neurons. J Physiol (Lond) 420: Collingridge GL, Herron CE, Lester RA (1988). Synaptic activation of N-methyl-Daspartate receptors in the Schaffer collateral-commissural pathway of rat hippocampus. J Physiol. 399: Connolly CN, Kittler JT, Thomas P, Uren JM, Brandon NJ, Smart TG, Moss SJ. (1999). Cell surface stability of gamma-aminobutyric acid type A receptors. Dependence on protein kinase C activity and subunit composition. J Biol Chem. 274(51): Costa E., Guidotti A, Toffano G (1978). Molecular mechanisms mediating the action of diazepam on GABA receptors. Br J Pharmacol 133: Costa, E.; Auta, J.; Grayson, D.R.; Matsumoto, K.; Pappas, G.D.; Zhang, X.; and Guidotti, A. (2002). GABA(A) receptors and benzodiazepines: a role for dendritic resident subunit mrnas. Neurophamacology, 43: Crestani F, Martin JR, Mohler H, Rudolph U (2000). Mechanism of action of the hypnotic zolpidem in vivo. Br J Pharmacol, 131: Cui xy, Zhao X, Chu QP, Chen BQ and Zhang YH. (2007). Influence of diltiazem on the behavior of zolpidem-treated mice in the elevated-plus maze test. J Neural Transm. 114(2): Dale Purves et al. Neuroscience. (1st edition): chapter

205 Davis M and Gallager DW (1988). Continuous slow release of low levels of diazepam produces tolerance to its depressant and anxiolytic effects on the startle reflex. Eur. J. Pharmacol., 150: Derkach V, Barria A, Soderling TR (1999). Ca 2+ / calmodulin-kinase II enhances channel conductance of alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate type glutamate receptors. Proc Natl Acad Sci USA 96: Dingledine R, Borges K, Bowie D, Traynelis SF (1999). The glutamate receptor ion channels. Pharmacol Rev 51: Dolin SJ, Path TL, Rabbani M, Siarey RJ, Bowhay AR Little HJ (1990). Nitrendipine decreases benzodiazepine withdrawal seizures but not the development of benzodiazepine tolerance or withdrawal signs. Br J Pharmacol 101: Dolmetsch RE, Pajvani U, Fife K, Spotts JM, Greenberg ME (2001). Signaling to the nucleus by an L-type calcium channel-calmoduline complex through the MAP kinase pathway. Science, 294: Ducic, I.; Caruncho, H.J.; Zhu, W.J.; Vicini, S.; and Costa, E. (1995). γ-amino butyric acid gating of Cl - Channels in recombinant GABA(A) receptors. J Pharmacol Exp Ther, 272: Dunworth SJ, Stephens DN (1998). Sensitization to repeated withdrawal, in mice treated chronically with diazepam, is blocked by an NMDA receptor antagonist. Psychopharmacology 136: Elliott EM, Malouf AT, Catterall WA (1995). Role of calcium channel subtypes in calcium transients in hippocampal CA3 neurons. J Neurosci 15(10):

206 El Ganouni S, Hanoun N, Boni C, Tazi A, Hakkou F, Hamon M. (2004). Prevention of diazepam withdrawal syndrome by nifedipine-behavioral and neurochemical studies. Pharmacol Biochem Behav. 79(2): Farb DH, Borden LA, Chan CY, Czajkowski CM, Gibbs TT, Schiller GD (1984). Modulation of neuronal function through benzodiazepine receptors: biochemical and electrophysiological studies of neurons in primary monolayer cell culture. Ann N.Y. Acad. Sci. 435: Fernandes C, File SE (1999). Dizocilpine does not prevent the development of tolerance to the anxiolytic effects of diazepam in rats. Brain Res 815: File, S.E. (1985). Tolerance to the behavior actions of benzodiazepines. Neurosci Biobehav Rev, 9: Felix R (1999). Voltage-dependent Ca 2+ channel α 2 δ auxiliary subunit: structure, function and regulation. Receptors Channels, 6: Felix R (2005). Molecular regulation of voltage-gated Ca 2+ channels. J Receptor and Signal Transduction, 25: File SE, Fernandes C (1994). Dizocilpine prevents the development of tolerance to the sedative effects of diazepam in rats. Pharmacol Biochem Behav 47: Fitzgerald LW, Ortiz J, Hamedani AG, Nestler EJ (1996). Drugs of abuse and stress increase the expression of GluR1 and NMDAR1 glutamate receptor subunits in the rat ventral tegmental area: common adaptations among cross-sensitizing agents. J Neurosci 16: Gallager DW, Lakoski JM, Gonsalves SF, Rauch SL (1984). Chronic benzodiazepine treatment decreases postsynaptic GABA sensitivity. Nature, 308:

207 Gallager DW, Rauch SL, Malcolm AB (1984b). Alternations in a low affinity GABA recognition site following chronic benzodiazepine treatment. Eur J Pharmacol 98: Gallager DW, Malcolm AB, Anderson SA, Gonsalves SF (1985). Continuous release of diazepam: electrophysiological, biochemical and behavioral consequences. Brain Res, 342: Gallager DW, Marley RJ, Hernandez TD. (1991). Biochemical and electrophysiological mechanisms underlying benzodiazepine tolerance and dependence. Pratt J, editor. New York: Academic Press: pp Ganguly K, Schinder AF, Wong ST, Poo M (2001). GABA itself promotes the developmental switch of neuronal GABAergic responses from excitation to inhibition. Cell. 105(4): Ganouni SE, Hanoun N, Boni C, Tazi A, Hmon F (2004). Prevention of diazepam withdrawal syndrome by nifedipine behavioral and neurochemical studies. Pharmacology, Biochemistry and Behavior 79: Gershon E (1992). Effect of benzodiazepine ligands on Ca 2+ channel currents in Xenopus oocytes injected with rat heart RNA. J Basic Clin Physiol Pharmacol 3: Giap BT, Jong CN, Ricker JH, Cullen NK, Zafonte RD (2000). The hippocampus: anatomy, pathophysiology, and regenerative capacity. J Head Trauma Rehabilitation, 15(3): Gilbert ME (1988). The NMDA-receptor antagonist, MK-801, suppresses limbic kindling and kindled seizures. Brain Res 463:

208 Gravielle MC, Faris R, Russek SJ, Farb DH (2005). GABA induces activity dependent delayed-onset uncoupling of GABA/Benzodiazepine site interactions in neocortical neurons. J. Biol. Chem., 280: Griffiths, R.R., Weerts, E.M. (1997). Benzodiazepine self-administration in humans and laboratory animals-implications for problems of long-term use and abuse. Psychopharmacology 134: Griffiths, R.R. and Johnson, M.W. (2005). Relative abuse liability of hypnotic drugs: a conceptual framework and algorithm for differentiating among compounds. J Clin Psych 66 (Suppl 9): Groth RD, Dunbar RL, Mermelstein PG (2003). Calcineurin regulation of neuronal plasticity. Biochem Biophys Res Commun 311: Gupta MB, Nath C, Patnaik GK, Saxena RC. (1996) Effect of calcium channel blockers on withdrawal syndrome of lorazepam in rats. Indian J Med Res. 103: Haefely W (1985). The biological basis of benzodiazepine action. In: Smith DE and Wesson DR, eds. The benzodiazepines: current standards for medical practice. Lancaster, MTP Press, Hanlon MR, Wallace BA (2002). Structure and function of voltage-dependent ion channel regulatory β subunits. Biochemistry, 41: Hell JW, Westenbroek RE, Warner C, Ahlijanian MK, Prystay W, Gilbert MM, Snutch TP, Catterall WA (1993). Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel alpha 1 subunits. J Cell Biol, 123:

209 Hell JW, Westenbroek RE, Breeze LJ, Wang KKW, Chavkin C, Catterall WA (1996). N- methyl-d-aspartate receptor-induced proteolytic conversion of postsynaptic class C L-type calcium channels in hippocampal neurons. Proc Natl Acad Sci USA 93: Heninger SN, Tallman JF, Garret KM, Vitek MP, Duman RS, GAllager DW (1990). Effects of continuous diazepam administration on GABA A subunit mrna in rat brain. J Mol Neurosci. 2: Hestrin S, Nicoll RA, Perkel DJ, Sah P (1990). Analysis of excitatory synaptic action in pyramidal cells using whole-cell recording from rat hippocampal slices. J Phyiol, 442: Holt RA, Bateson AN, Martin IL (1996). Chronic treatment with diazepam or abecarnil differentially affects the expression of GABA A receptor subunit mrnas in the rat cortex. Neuropharmacology 35: Holt RA, Bateson AN, Martin IL (1999). Decreased GABA enhancement of benzodiazepine binding after a single dose of diazepam. J Neurochem. 72(5): Hu S, Ticku MK (1994). Chronic benzodiazepine agonist treatment produces functional uncoupling of the γ-amnobutyric acid-benzodiazepine receptor ionophore complex in cortical neurons. J Pharmacol Exp Ther 45: Huang X, Chen S, Tietz EI (1996). Immunocytochemical detection of regional protein changes in rat brain sections using computer-assisted image analysis. J Histochem Cytochem, 44:

210 Hunkeler W, Moehler H, Pieri L (1981). Selective antagonist of benzodiazepines. Nature 290: Hutchinson MA, Smith PF, Darlington CL (1996). The behavioral and neuronal effects of the chronic administration of benzodiazepine anxiolytic and hypnotic drugs. Prog Neurobiol. 49: Impagnatiello F, Pesold C, Longone P, Carunchol H, Fritschy JM, Costa E, Guidotti A (1996). Modification of γ-amino butyric acid type A receptor subunit expression in rat neocortex during tolerance to diazepam. Mol Pharmacol 49: Ishizawa Y, Furuya K, Yamagishi S and Dohi S (1997). Non-GABAergic effects of midazolam, diazepam and flumazenil on voltage-dependent ion currents in NG cells. NeuroReport 8: Isomura Y, Sugimoto M, Fujiwara-Tsukamoto Y, Yamamoto-Muraki S, Yamada J, Fukuda A (2003). Synaptically activated Cl- accumulation responsible for depolarizing GABAergic responses in mature hippocampal neurons. J Neurophysiol. 90(4): Izzo E, Auta J, Impagnatiello F, Pesold C, Guidotti A, Costa E (2001). Glutamic acid decarboxylase and glutamate receptor changes during tolerance and dependence to benzodiazepines. Proc Natl Acad Sci USA 98: Jackson A, Mead AN, Stephens DN (2000). Behavioral effects of alpha-amino-3- hydroxy-5-methyl-4-isoxazolepropionate-receptor antagonists and their relevance to substance abuse. Pharmacol Ther 88:

211 Johansen J, Taft WC, Yang J, Kleinhaus AL, DeLorenzo RJ (1985). Inhibition of Ca2+ conductance in identified leech neurons by benzodiazepines. Proc Natl Acad Sci U S A. 82(11): Johnston D, Fisher R, Gray R (1992). Voltage- gated calcium channels in adult hippocampal neurons. Ion channel, 3 (2): Kang L, Miller LG (1991). Decreased GABA A receptor subunit mrna concentrations following chronic lorazepam administration. Br J Pharmacol, 103: Kang MG, Campbell KP (2003). Gamma subunit of voltage-activated calcium channels. J Biol Chem, 278: Kelley AE. (2004) Memory and addiction: Shared neural circuitry and molecular mechanisms. Neuron, 44: Klugbauer N, Marais E, Hofmann F (2003). Calcium channel α 2 δ subunit: differential expression, function, and drug binding. J Bioengerg Biomembr 35: Knowles WD (1992). Normal anatomy and neurophysiology of the hippocampal formation. J. Clin. Neurophysiol. 9: Koff JM, Pritchard GA, Greenblatt DJ, Miller LG (1997). The NMDA receptor competitive antagonist CPP modulates benzodiazepine tolerance and discontinuation. Pharmacololgy 55: Koninck KD and Mody I (1996). The effects of rasing intracellular calcium on synaptic GABA A receptor-channels. Neuropharmacology 35: Koob GF, Bloom FE (1988). Cellular and molecular mechanisms of drug dependence. Science 242:

212 Koschak A, Reimer D, Walter D, Hoda JC, Heinzle T, Grabner M, Striessnig J (2001). Alpha 1D (Cav1.3) subunits can form L-type Ca 2+ channels activating at negative voltages. J Biol Chem 276: Lavoie AM, Twyman RE (1996). Direct evidence for diazepam modulation of GABAA receptor microscopic affinity. Neuropharmacology. 35(9-10): Liao D, Hessler NA, Malinow R (1995). Activation of postsynaptically silent synapses during pairing-induced LTP in CA1 region of hippocampal slice. Nature 375: Liao D, Scannevin RH, Huganir R (2001). Activation of silent synapses by rapid activitydependent synaptic recruitment of AMPA receptors. J Neurosci 21: Lipscombe D, Helton TD, and Xu WF. (2004) L-Type Calcium Channels: The Low Down. J Neurophysiol 92: Lipton SA, Rosenberg PA (1994). Excitatory amino acids as a final common pathway for neurological disorders. N Engl J Med 330: Lledo PM, Zhang X, Sudhof TC, Malenka RC, Nicoll RA (1998). Postsynaptic membrane fusion and long-term potentiation. Science 279: Loscher W (1998). Pharmacology of glutamate receptor antagonists in the kindling model of epilepsy. Prog Neurobiol 54: Lopes da Silva FH, Arnolds DEAT (1978). Physiology of the hippocampus and related structures. Ann. Rev.Physiol., 40: Low K, Crestani F, Keist R, Benke D, Brunig I, Benson JA, Fritschy JM, Rulicke T, Bluethmann H, Mohler H, Rudolph U (2000). Molecular and neuronal substrate for selective attenuation of anxiety. Science, 290:

213 Lu WX, Wolf ME (1999). Repeated amphetamine administration alters AMPA receptor subunit expression in rat nucleus accumbens and medial prefrontal cortex. Synapse 32: Lu YM, Mansuy IM, Kandel ER, Roder J (2000). Calcineurin-mediated LTP of GABAergic inhibition underlies the increased excitability of CA1 neurons associated with LTP. Neuron 26: Luo J, Wang Y, Yasuda RP, Dunah AW, Wolfe BB (1997). The majority of N-methyl-Daspartate receptor complexes in adult rat cerebral cortex contain at least three different subunits (NR1/NR2A/NR2B). Mol Pharmacol, 51: Lyons HR, Gibbs TT, Farb DH (2000). Turnover and down-regulation of GABA(A) receptor alpha1, beta2s, and gamma1 subunit mrnas by neurons in culture. J Neurochem. 74 (3): Lyons HR, Land MB, Gibbs TT, Farb DH (2001). Distinct signal transduction pathways for GABA-induced GABA(A) receptor down-regulation and uncoupling in neuronal culture: a role for voltage-gated calcium channels. J Neurochem. 78(5): Macdonald, R.L. and Olsen, R.W. (1994). GABA (A) receptor channels. Annu Rev Nature, 305: Marty A and Llano I (2005). Excitatory effects of GABA in established brain networks. Trends Neurosci. 28(6): Malenka RC (1994). Synaptic plasticity in the hippocampus: LTP and LTD. Cell 78: Malenka RC and Nicoll RA. (1999). Long-Term Potentiation-A Decade of Progress? Science 285,

214 Malenka RC. (2003) Synaptic plasticity and AMPA receptor trafficking. Ann N Y Acad Sci. 1003:1-11. Malinow R. (2003). AMPA receptor trafficking and long-term potentiation. Philos Trans R Soc Lond B Biol Sci. 29;358(1432) : Mattson RH (1974). Other antiepileptic drugs: the benzodiazepines. In: Woodbury DM, Penry JK, Schmidt RP, ed., Antiepileptic drugs. New York, Raven Press: Mayer ML, Westbrook GL, Guthrie PB (1984). Voltage-dependent block by Mg 2+ of NMDA responses in spinal cord neurons. Nature, 309: McGlade-McCulloh E, Yamamoto H, Tan SE, Brickey DA, Soderling TR (1993). Phosphorylation and regulation of glutamate receptors by calcium/calmodulindependent protein kinase II. Nature 362: McKernan RM and Whiting PJ (1996). Which GABA A receptor subtypes really occur in the brain? Trends Neurosci 19: Meyer DK, Olenik C, Hofmann F, Barth H, Leemhuis J, Brunig I, Aktories K, Norenberg W (2000). Regulation of somatodendritic GABAA receptor channels in rat hippocampal neurons: evidence for a role of the small GTPase Rac1. J Neurosci. 20(18): Millan MJ (2003). The neurobiology and control of anxious states. Prog Neurobiol 70: Miller LG, Greenblatt DJ, Barnhill JG, Shader RI (1988). Chronic benzodiazepine administration. I. Tolerance is associated with benzodiazepine receptor downregulation and decreased γ-aminobutyric acid A receptor function. J Pharmacol Exp Ther, 246:

215 Monyer H, Burnashev N, Laurie DJ, Sakmann B, Seeburg PH (1994). Development and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron, 12: Morgan SL, Teyler TJ (1999).VDCCs and NMDARs underlie two forms of LTP in CA1 hippocampus in vivo. J Neurophysiol 82: Mori Y, Niidome T, Fujita Y, Mynlieff M, Dirksen RT, Beam KG, Iwabe N, Miyata T, Furutama D, Furuichi T, Mikoshiba K (1993). Molecular diversity of voltagedependent calcium channel. Ann NY Acad Sci, 707: Moser MB and Moser EI (1998). Functional differentiation in the hippocampus. Hippocampus 8: Molher H, Okada T (1977). Benzodiazepine receptors demonstrated in the central nervous system. Science 198: Molher, H.; Fritschy, J.M.; and Rudolph, U. (2002) A new benzodiazepine pharmacology. J Pharmacol Exp Ther, 300: 2-8. Molleman A, Little HJ (1995). Increases in non-n-methyl-d-aspartate glutamatergic transmission, but no change in gamma-aminobutyric acid B transmission, in CA1 neurons during withdrawal from in vivo chronic ethanol treatment. J Pharmacol Exp Ther 274: Monyer H, Seeburg PH, Wisden W (1991). Glutamate-operated channels: developmentally early and mature forms arise by alternative splicing. Neuron. 6(5):

216 Monyer H, Burnashev N, Laurie DJ, Sakmann B, Seeburg PH (1994). Developmental and regional expression in the rat brain and functional properties of four NMDA receptors. Neuron, 12: Nayak A, Zastrow DJ, Lickteig R, Zahniser NR, Browning MD (1998). Maintenance of late-phase LTP is accompanied by PKA-dependent increase in AMPA receptor synthesis. Nature 394: Nestler EJ (2001). Neurobiology. total recall the memory of addiction. Science 292: Nestler EJ (2005). Is there a common molecular pathway for addiction? Nat Neurosci 8(11): O Brien RJ, Kamboj S, Ehlers MD, Rosen KR, Fischbach GD, Huganir RL (1998). Activity-dependent modulation of synaptic AMPA receptor accumulation. Neuron 21: Olsen RW, Tobin AJ (1990). Molecular biology of GABA A receptors. FASEB J 4: Perez-Otano I, Ehlers MD. (2005) Homeostatic plasticity and NMDA receptor trafficking. Trends in Neuroscience. 28(5) : Pesold C, Caruncho HJ, Impagnatiello F, Berg MJ, Fritschy JM, Guidotti A and Costa E (1997). Tolerance to diazepam and changes in GABA A receptor subunit expression in rat neocortical areas. Neuroscience, 79: Podhorna J. (2002). The experimental pharmacotherapy of benzodiazepine withdrawal. Current Pharmaceutical Design. 8:

217 Polc P (1988). Electrophysiology of benzodiazepine receptor ligands: multiple mechanisms and sites of action. Prog Neurobiol 31: Poisbeau P, Williams SR, Mody I. Silent GABAA synapses during flurazepam withdrawal are region-specific in the hippocampal formation (1997). J Neurosci. 17(10): Pourmotabbed A, Motamedi F, Fathollahi Y, Mansouri FA, Semnanian S. (1998). Involvement of NMDA receptors and voltage-dependent calcium channels on augmentation of long-term potentiation in hippocampal CA1 area of morphine dependent rats. Brain Research 804 : Prichett DB, Sontheimer H, Shivers BD, Ymer S, Kettenmann H, Schofield PR, Seeburg PH (1989). Importance of a novel GABA A receptor subunit for benzodiazepine pharmacology. Nature 338: Primus RJ, Gallager DW (1992). GABA A receptor subunit mrna levels are differentially influenced by chronic FG 7142 and diazepam exposure. Eur J Pharmacol--Mol Pharmacol Sect, 226: Primus RJ, Yu J, Xu J, Hartnett C, Meyyappan M, Kostas C, Ramabhadran TV, Gallager DW (1996). Allosteric uncoupling after chronic benzodiazepine exposure of recombinant gamma-aminobutyric acid(a) receptors expressed in Sf9 cells: ligand efficacy and subtype selectivity. J Pharmacol Exp Ther. 276(3): Purves D, Augustine GJ, Fitzpatrick D, Katz LC, LaMantia AS, McNamara JO (1997). Neuroscience, Sinauer Associates, INC. Publishers. Rabbani M, Little HJ. (1999). Increase in neuronal Ca 2+ flux after withdrawal from chronic barbiturate treatment. European Journal of Pharmacology 364:

218 Rajadhyksha AM and Kosofsky BE (2005) Psychostimulants, L-type calcium channels, kinases, and phosphatases. The Neuroscientist 11: Reichling DB, Kyrosis A, Wang J and MacDermott AB (1994). Mechanisms of GABA and glycine depolarization-induced calcium transients in rat dorsal horn neurons. J Physiol 476: Reuveny E, Twombly DA and Narahashi T (1993). Chlordiazepoxide block of two types of calcium channels in neuroblastoma cells. J Pharmacol Exp Ther 264: Ricci A, Sabbatini M, Tomassoni D, Mignini F, Petrelli C, Amenta F (2002). Neuronal populations of rat cerebral cortex and hippocampus expressed a higher density of L- type Ca2+ channel than corresponding cerebral vessels. Clin Exper Hyper 24: Robinson TE, Berridge KC (1993). The neural basis of drug craving: an incentivesensitization theory of addiction. Brain Res Rev 18: Roca DJ, Rozenberg I, Farrant M, Farb DH (1990). Chronic agonist exposure induces down-regulation and allosteric uncoupling of the γ-amino butyric acid/benzodiazepine receptor complex. Mol Pharmacol 37: Rogawski MA, Kurzman PS, Yamaguchi SI, Li H (2001). Role of AMPA and GluR5 kainate receptors in the development and expression of amygadala kindling in the mouse. Neuropharmacology 40: Rosenberg HC, Chiu TH (1981). Tolerance during chronic benzodiazepine treatment associated with decreased receptor binding. Eur J Pharmacol, 70: Rosengberg HC, Tietz EI, Chiu TH (1985). Tolerance to the anticonvulsant action of benzodiazepines: relationship to decreased receptor density. Neuropharmacology 24:

219 Rosenberg HC, Tietz EI, Chiu TH (1989). Tolerance to anticonvulsant effects of diazepam, clonazepam and clobazam in amygdale-kindled rats. Epilepsia, 30: Rosenberg, H.C. (1995). Differential expression of benzodiazepine anticonvulsant crosstolerance according to time following flurazepam or diazepam treatment. Pharmacol Biochem Behav, 51: Ruano D, Vizuete M, Cano J, Machado A, Vitorica J (1992). Heterogeneity in the allosteric interaction between the gamma-aminobutyric acid (GABA) binding site and three different benzodiazepine binding sites of the GABAA/benzodiazepine receptor complex in the rat nervous system. J Neurochem. 58(2): Rudolph U, Crestani F, Mohler H (2001). GABA A receptor subtypes: dissecting their pharmacological functions. Trends Pharmacol Sci 22: Russek SJ, Bandyopadhyay S, Farb DH (2000). An initiator element mediates autologous downregulation of the human type A gamma -aminobutyric acid receptor beta 1 subunit gene. Proc Natl Acad Sci U S A. 97(15): Sanchis-Segura C, Borchardt T, Vengetina V, Zghoul T, Bachteler D, Gass P, Sprengel R, Spanagel R (2006). Involvement of the AMPA receptor-c subunit in alcoholseeking behavior and relapse. J Neurosci 26: Sato K, Morimoto K, Okamoto M (1988). Anticonvulsant action of a non-competitive antagonist of NMDA receptors (MK-801) in the kindling model of epilepsy. Brain Res 463: Schofield PR (1989). The GABA A receptor: molecular biology reveals a complex picture. Trends Pharmacol Sci 10:

220 Schweizer E and Rickels K. (1998). Benzodiazepine dependence and withdrawal: a review of the syndrome and its clinical management. Acta Psychiatr Scand Suppl, 393: Semyanov A (2003). Cell type specificity of GABA A receptor mediated signaling in the hippocampus. Curr Drug Target CNS Neurol Disord, 2: Sieghart W, Sperk G (2002). Subunit composition, distribution and function of GABA A receptor subtypes. Current Topics in Medicinal Chemistry, 2: Sommer B, Keinanen K, Verdoorn TA, Wisden W, Burnashev N, Herb A, Kohler M, Takagi T, Sakmann B, Seeburg PH (1990). Flip and flop: a cell-specific functional switch in glutamate operated channels of the CNS. Science 249: Sommer B, Kohler M, Sprengel R, Seeburg PH (1991). RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell 67: Sommer B, Seeberg PH (1992). Glutamate receptor channels-novel properties and new clones. Trend Pharmacol Sci 13: Stein V and Nicoll RA (2003). GABA generates excitement. Neuron 37(3): Stephens DN (1995). A glutamatergic hypothesis of drug dependence. Behav Pharmacol 6: Steppuhn KG, Turski L (1993). Diazepam dependence prevented by glutamate antagonists. Proc. Natl Sci USA 90: Study, R.E. and Barker, J.L. (1981). Diazepam and (-)-pentobarbital: Fluctuation analysis reveals different mechanisms for potentiation of GABA responses in cultured central neurons. Proc. Natl. Acad. Sci., 78:

221 Sun MK, Zhao WQ, Nelson TJ, Alkon DL. Theta rhythm of hippocampal CA1 neuron activity: gating by GABAergic synaptic depolarization (2001). J Neurophysiol. 85(1): Taft WC, DeLorenzo RJ (1984). Micromolar-affinity benzodiazepine receptors regulate voltage-sensitive calcium channels in nerve terminal preparations. Proc Natl Acad Sci 81: Tallman JF, Thomas JW, Gallager DW (1978). GABAergic modulation of benzodiazepine binding site sensitivity. Nature 274, Taft WC, DeLorenzo RJ (1984). Micromolar-affinity benzodiazepine receptors regulate voltage-sensitive calcium channels in nerve terminal preparations. Proc Natl Acad Sci 81: Tan SE, Wenthold RJ, Soderling TR (1994). Phosphorylation of AMPA-type glutamate receptors by calcium/calmodulin-dependent protein kinase II and protein kinase C in cultured hippocampal neurons. J Neurosci 14: Timmermann DB, Westenbroek RE, Schousboe A, Catterall WA (2002). Distribution of high-voltage-activated calcium channels in cultured gamma-aminobutyric acidergic neurons from mouse cerebral cortex. J Neurosci. Res. 67: Tietz EI, Rosenberg HC, Chiu TH (1986). Autoradiographic localization of benzodiazepine receptor downregulation. J Pharmacol Exp Ther, 236: Tietz EI, Chiu TH, Rosenberg HC (1989). Regional GABA/benzodiazepine receptor/chloride channel coupling after acute and chronic benzodiazepine treatment. Eur J Pharmacol 167:

222 Tietz EI, Huang X, Weng X, Rosenberg HC, Chiu TH (1994). Expression of α1, α5, and γ2 GABA A receptor subunit mrnas measured in situ in rat hippocampus and cortex following chronic flurazepam administration. J Mol Neurosci 4: Tietz EI, Zeng XJ, Chen S, Lilly SM, Rosenberg HC, Kometiani P (1999a). Antagonistinduced reversal of functional and structural measures of hippocampal benzodiazepine tolerance. J Pharmacol ExpTher, 291: Tietz EI, Huang X, Chen S, Ferencak WFr (1999b). Temporal and regional regulation of α1, β2, and β3, but not α2, α4, α5, α6, β1 or γ2 GABA A receptor subunit messenger RNAs following one-week oral flurazepam administration. Neuroscience, 91: Trujillo KA, Akil H (1995). Excitatory amino acids and drugs of abuse: a role for N- methyl-d-aspartate receptors in drug tolerance, sensitization and physical dependence. Drug Alcohol Depend 38: Turrigiano GG (2000). AMPA receptors unbound: membrane cycling and synaptic plasticity. Neuron 26: 5-8. Tzschentke and Schmidt (2003). Glutamatergic mechanisms in addiction. Mol Psychiatry. 8: Vandenberghe W, Nicoll RA, Bredt DS (2005). Stargazin is an AMPA receptor auxiliary subunit. Proc Natl Acad Sci USA 102: Van Sickle B, Tietz EI (2002). Selective enhancement of AMPA receptor-mediated function in hippocampal CA1 neurons from chronic benzodiazepine-treated rats. Neuropharmacology 43:

223 Van Sickle B, Cox A, Schak K, John GL, Tietz EI (2002). Chronic benzodiazepine administration alters hippocampal CA1 neuron excitability: NMDA receptor function and expression. Neuropharmacology 43: Van Sickle B, Xiang K Tietz EI (2004). Transient plasticity of hippocampus CA1 neuron glutamate receptors contributes to benzodiazepine withdrawal anxiety. Neuropsychopharmacology 29: Vekovischeva OY, Zamanillo D, Echenko O, Seppala T, Uusi-Oukari M, Honkanen A, et al. (2001). Morphine-induced dependence and sensitization are altered in mice deficient in AMPA-type glutamate receptor-a subunits. J of Neurosci 21: Vicini S, Wang JF, Li JH, Zhu WJ, Wang YH, Luo JH, Wolfe BB, Grayson DR (1998). Functional and pharmacological differences between recombinant N-methyl-Daspartate receptors. J Neuroohysiol, 79: Voipio J, Kaila K (2000). Gabaergic excitation and K + -mediated volume transmission in the hippocampus. Prog. Brain Res. 125: Walter HJ, Messing RO (1999). Regulation of neuronal voltage-gated calcium channels by ethanol. Neurochemistry international 35: Wang RA. Cheng G, Kolal M, Randic M (1995). α-subunit of calcium/calmodulindependent protein kinase II enhances γ-aminobutyric acid and inhibitory synaptic responses of rat neurons in vitro. J Neurophysiol. 73: Wang J, Liu SH, Haditsch U, Tu WH, Cochrane K, Ahmadian G, Tran L, Paw J, Wang YT, Mansuy I, Salter MM, Lu YM (2003). Interaction of calcineurin and type-a GABA receptor γ2 subunits produces long-term depression as CA1 inhibitory synapses. J Neurosci, 23(3):

224 Watabe S, Yoshii M, Ogata N, Tsunoo A, Narahashi T (1993). Differential inhibition of transient and long-lasting calcium channel currents by benzodiazepines in neuroblastoma cells. Brain Res. 606(2): Watson WP, Little HJ (2002). Selectivity of the protective effects of dihydropyridine calcium channel antagonist against the ethanol withdrawal syndrome. Brain Research 930: Watt AJ, van Rossum MC, MacLeod KM, Nelson SB, Turrigiano GG.(2000). Activity coregulates quantal AMPA and NMDA currents at neocortical synapses. Neuron.26: Watt AJ, Sjostrom PJ, Hausser M, Nelson SB, Turrigiano G (2004) A proportional but slower NMDA potentiation follows AMPA potentiation in LTP. Nature Neuroscience. 7: Wei X, Neely A, Lacerda AE, Olcese R et al (1994). Modification of Ca 2+ channel activity by deletions at the carboxyl terminus of the cardiac α1 subunit. J Biol Chem 269: Weick JP, Groth RD, Isaksen AL, Mermelstein PG (2003). Interactions with PDZ proteins are required for L-type calcium channels to activate camp response element-binding protein-dependent gene expression. J Neurosci 23: Weiland GA, Oswald RE (1985). The mechanism of binding of dihydropyridine calcium channel blockers to rat brain membranes. J. Biol. Chem. 260 (14): Wenthold RJ, Yokotani N, Doi K, Wada K (1992). Immunochemical characterization of the non-nmda glutamate receptor using subunit-specific antibodies. Evidence for a hetero-oligoeric structure in rat brain. J Biol Chem 267:

225 Wenthold RJ, Petralia RS, Blahos J II, Niedzielski AS (1996). Evidance for multiple AMPA receptor complexes in hippocampal CA1/CA2 neurons. J Neurosci 16: Westenbroek RE, Ahlijanian MK, Catterall WA (1990). Clustering of L-type Ca 2+ channels at the base of major dendrites in hippocampal pyramidal neurons. Nature 347: Westenbroek RE, Hoskins L, Catterall WA (1998). Localization of Ca 2+ channel subtype on rat spinal motor neurons, interneurons, and nerve terminals. J Neurosci 18: Whetsell WOJ (1996). Current concepts of excitotoxicity. J Neuropathol Exp Neurol 55: Whittington MA, Little HJ (1991). A calcium channel antagonist stereo selectively decreases ethanol withdrawal hyper-excitability, but not that due to bicuculline, in hippocampal slices. Br. J. Pharmaco. 103: Whittington MA, Little HJ. (1993). Changes in voltage-operated calcium channels modify ethanol withdrawal hyperexcitability in mouse hippocampal slices. Exp Physiol. 78(3): Wolf ME (1998). The role of excitatory amino acids in behavioral sensitization to psychomotor stimulants. Prog Neurobiol 54: Wyllie DJ, Manabe T, Nicoll RA(1994). A rise in postsynaptic Ca2+ potentiates miniature excitatory postsynaptic currents and AMPA responses in hippocampal neurons. Neuron. 12(1):

226 Wyllie DJ, Nicoll RA(1994). A role for protein kinases and phosphatases in the Ca(2+)- induced enhancement of hippocampal AMPA receptor-mediated synaptic responses. Neuron. 13(3): Xia Z, Storm DR. (2005). The role of calmodulin as a signal integrator for synaptic plasticity. Nat Rev Neurosci. 6: Xiang K, Giovannucci DR, Greenfield Jr, LJ, Tietz EI. (2006) Enhanced calcium entry through L-type voltage-gated calcium channels contributes to AMPA receptor plasticity in hippocampal CA1 neurons after chronic benzodiazepine treatment. Abstract Viewer/Itinerary Planner. Washington, DC: Soc for Neurosci., Online. Xiang K, Tietz EI (2007). Benzodiazepine-induced hippocampal CA1 neuron AMPA receptor plasticity linked to severity of withdrawal-anxiety: Differential role of voltage-gated calcium channels and NMDA receptors. Behav Pharmacol in press. Xie XH, Tietz EI (1992). Reduction in potency of selective gamma-aminobutyric acid A agonists and diazepam in CA1 region of in vitro hippocampal slices from chronic flurazepam-treated rats. J Pharmacol Exp Ther 262: Xu W, Lipscombe D (2001). Neuronal Ca(V)1.3alpha(1) L-type channels activate at relatively hyperpolarized membrane potentials and are incompletely inhibited by dihydropyridines, J Neurosci, 21: Yasuda R, Sabatini BL, Svoboda K (2003). Plasticity of calcium channels in dendritic spines. Nat Neurosci, 6:

227 Zamanillo D, Sprengel R, Hvalby O, Jensen V, Burnashev N, Rozov A, Kaiser KM, Koster HJ, Borchardt T, Worley P, Lubke J, Frotscher M, Kelly PH, Sommer B, Andersen P, Seeburg PH, Sakmann B (1999). Importance of AMPA receptors for hippocampal synaptic plasticity but not for spatial learning. Science 284: Zeng X, Xie XH and Tietz EI (1995). Reduction of GABA-mediated inhibitory postsynaptic potentials in hippocampus CA1 pyramidal neurons following oral flurazepam administration. Neuroscience 66: Zeng X and Tietz EI (1997). Depression of early and late monosynaptic inhibitory postsynaptic potentials in hippocampal CA1 neurons following prolonged benzodiazepine administration: role of reduction in Cl - driving force. Synapse 25: Zeng XJ and Tietz EI (1999). Benzodiazepine tolerance at GABAergic synapses on hippocampal CA1 pyramidal cells. Synapses 31: Zeng X and Tietz EI (2000). Role of bicarbonate ion in mediating decreased synaptic conductance in benzodiazepine tolerant hippocampal CA1 pyramidal neurons. Brain Res 868: Zhao T, Chiu TH, Rosenberg HC (1994). Decreased expression of (-amino butyric acid type A/benzodiazepine receptor β subunit mrna in flurazepam-tolerant rats. J Mol Neurosci 5: Zhong W, Dong Z, Tian M, Cao J, Xu T, Xu L, Luo J (2006). Opiate withdrawal induces dynamic expressions of AMPA receptors and its regulatory molecule CaMKII alpha in hippocampal synapses.2006, Life Sci 79:

228 APPENDICES Transient Plasticity of Hippocampal CA1 Neuron Glutamate Receptors Contributes to Benzodiazepine Withdrawal-Anxiety Bradley J Van Sickle, Kun Xiang and Elizabeth I Tietz Neuropsychopharmacology (2004) 29, , advance online publication, 21 July 2004; 222

229 RESULTS Effect of FZP Withdrawal on CA1 Neuron Excitability To determine whether enhanced AMPA receptor-mediated currents may contribute to CA1 neuron hyperexcitability and expression of withdrawal-anxiety in 1-day FZP-withdrawn rats, prior to downregulation of NMDA receptor-mediated currents, 4-AP-induced endogenous glutamate release was examined by its capacity to increase CA1 neuron spike discharges in hippocampal slices. CA1 neurons are not spontaneously active, and thus baseline spike discharges were infrequent during ACSF superfusion (Figure 5). In the CA1 region of control slices, 4-AP increased spike discharges from 2-3 per 5 min to 8-13 per 5 min across the 40 min drug superfusion (Figure 5). While spontaneous CA1 neuron activity was similar in slices from control and 1-day FZP withdrawn rats, there was a significant two-fold (20-24 spikes) increase in frequency of 4-AP-induced spiking in slices from 1-day FZP-withdrawn rats beginning 5 min after start of 4-AP superfusion and persisting for the duration of drug superfusion (Figure 5b). Upon drug washout, the CA1 neuron population response returned to baseline in slices from both control and 1-day FZP-withdrawn groups. GYKI Reverses Upregulation of AMPA receptor Function and Prevents Anxiety-Like Behavior during FZP Withdrawal Since anxiety in 1-day FZP-withdrawn rats corresponded with increased AMPA receptormediated function in CA1 pyramidal neurons, we hypothesized that antagonism of AMPA 223

230 receptor activation at the onset of FZP withdrawal may prevent upregulation of AMPA receptor function in CA1 neurons and alter expression of benzodiazepine withdrawalanxiety. To evaluate this hypothesis, control and FZP-withdrawn rats were injected with 1% TWEEN-20 vehicle or GYKI (0.5 mg/kg) immediately following removal of FZP from the drinking water. FZP-withdrawn rats were then tested 1 day later in the elevated plus-maze prior to hippocampal slice preparation for studies of AMPAR function. In studies of AMPA receptor mepscs in CA1 neurons, there was no difference in resting membrane potential (CON/VEH: -60.3±1.0 mv, n=9 cells/eight rats; FZP/VEH: -60.1±1.2 mv, n=8 cells/six rats; CON/GYKI: -58.6±1.2 mv, n=9 cells/seven rats; FZP/GYKI: ±0.9 mv, n=8 cells/five rats), 10-90% rise time, decay or frequency of AMPA receptormediated events between control and 1-day FZP-withdrawn rats that received vehicle or GYKI injection the previous day. However, there was a significant interaction between FZP and GYKI treatments with respect to mepsc amplitude (F 1,33 =6.73, p=0.014). Similar to other groups of 1-day FZP withdrawn rats, AMPAR mepsc amplitude in CA1 neurons was upregulated in comparison to controls and unmodified by vehicle injection (Figure 7a; CON/VEH: -8.7±0.5 pa, FZP/VEH: -11.3±0.4 pa). GYKI injection (Figure 7a) at the end of the drug treatment period had no effect on AMPAR mepsc amplitude in control neurons; however, average AMPA receptor mepsc amplitude returned to control levels in neurons from 1-day FZP-withdrawn rats (CON/GYKI: -9.5±0.8 pa; FZP/GYKI: -9.0±0.6 pa). 224

231 Fig

232 CA1 neuron excitability during FZP withdrawal. Extracellular recordings of 4- aminopyridine (4-AP)-induced spike discharges were made in the CA1 pyramidal cell layer of hippocampal slices from control and 1-day FZP-withdrawn rats. A 5 min baseline recording during ACSF superfusion was followed by 40 min superfusion with 55 M 4-AP. Spike discharges were recorded for an additional 15 min during drug washout with ACSF. The frequency of spike discharges was measured per 5 min epoch. (a) Representative traces of spike discharges in slices from control (top) and 1-day FZPwithdrawn (bottom) rats taken during the first 5-10 min epoch of 4-AP superfusion. (b) CA1 neuron hyperexcitability in 1-day FZP-withdrawn rats. There was a significant increase in 4-AP-induced spike discharges in slices from FZP-withdrawn rats in comparison to control rats suggesting hyperexcitability of the hippocampal CA1 neuron population in 1-day FZP-withdrawn rats. Data were analyzed by repeated measures ANOVA with post hoc analysis of drug effect by the method of Scheffé. Asterisks denote significant differences between control and FZP-withdrawn groups, p

233 Fig