MOLECULAR MECHANISMS OF SYNAPTIC PLASTICITY · 2016. 3. 7. · Synaptic plasticity is a highly regulated process, emerging from complex interactions not only at the synapse itself

MOLECULAR MECHANISMSOF SYNAPTIC PLASTICITY

Implications for immediate early gene expression and learning

Die Deutsche Bibliothek – CIP-Einheitsaufnahme

Nijholt, Ingrid Maria:Molecular mechanisms of synaptic plasticity: Implications forimmediate early gene expression and learning / door Ingrid MariaNijholt. –1. Aufl. – Göttingen : Cuvillier, 2002

Zugl.: Groningen, Univ., Diss., 2002ISBN 3-89873-493-5

The research described in this thesis was carried out at theMax Planck Institute for Experimental Medicine in theDepartment of Molecular Neuroendocrinology, Göttingen,Germany.

The author was a Ph.D. student at the Graduate School forBehavioral and Cognitive Neurosciences of the University ofGroningen, the Netherlands.

© CUVILLIER VERLAG, Göttingen, 2002 Nonnenstieg 8, 37075 Göttingen Telefon: 0551-54724-0 Telefax: 0551-54724-21 www.cuvillier.de

Alle Rechte vorbehalten. Ohne ausdrückliche Genehmigung desVerlages ist es nicht gestattet , das Buch oder Teile daraus auffotomechanischem Weg (Fotokopie, Mikrokopie) zu vielfältigen.1. Auflage, 2002Gedruckt auf säurefreiem PapierISBN 3-89873-493-5

RIJKSUNIVERSITEIT GRONINGEN

MOLECULAR MECHANISMS OF SYNAPTIC PLASTICITY

Implications for immediate early gene expression and learning

PROEFSCHRIFT

ter verkrijging van het doctoraat in deWiskunde en Natuurwetenschappenaan de Rijksuniversiteit Groningen

op gezag van deRector Magnificus, dr. F. Zwarts,in het openbaar te verdedigen op

maandag 7 oktober 2002om 16.00 uur

door

INGRID MARIA NIJHOLT

geboren op 19 juni 1973

te Drachten

Promotores: Prof. dr. J. M. KoolhaasProf. dr. J. Spiess

Referent: Dr. T. Blank

Beoordelingscommissie: Prof. dr. P.G.M. LuitenProf. dr. M. JoëlsProf. dr. H.W.G.M. Boddeke

ISBN 3-89873-493-5

Mit Steinen,die einem in den Weg

gelegt werden,kann man auch

was schönes bauen.

CONTENTS

page

Chapter 1: General introduction 1

1. Synaptic plasticity 11.1 Basic principles of synaptic transmission 21.2 Second messenger pathways 3

1.2.1 Protein kinases 4 1.2.2 Protein phosphatases 5 1.2.3 Phosphoproteins 6

1.3 Glutamate receptors 7 1.3.1 NMDA receptor 7

2. Synaptic plasticity and gene expression 92.1 CREB 102.2 FOS 11

3. Synaptic plasticity and learning and memory 123.1 Hippocampus 123.2 Long-term potentiation and depression as a model for learning 12

and memory3.3 Influence of stress on synaptic plasticity and learning and memory 14

3.3.1 Corticotropin-releasing factor as mediator of the stress 15 response

Chapter 2: Modulation of hypothalamic NMDA receptor function by cyclic 33AMP-dependent protein kinase and phosphatases

Chapter 3: The phosphoprotein DARPP-32 mediates cAMP-dependent 45potentiation of striatal N-methyl-D-aspartate responses

Chapter 4: In vivo NMDA/dopamine interaction resulting in FOS production 62in the limbic system and basal ganglia of the mouse brain.

Chapter 5: In vivo CREB phosphorylation mediated by dopamine and NMDA 79receptor activation in mouse hippocampus and caudate nucleus.

Chapter 6: The corticotropin-releasing factor receptor 1 antagonist CP-154,526 87markedly enhances associative learning and paired- pulsefacilitation immediately after a stressful experience.

Chapter 7: Priming of long-term potentiation in mouse hippocampus by 101corticotropin-releasing factor and acute stress: implications forhippocampus-dependent learning.

Chapter 8: Modulation of neuronal excitability and associative learning by 117different corticotropin-releasing factor-signaling cascades inmouse hippocampus

Chapter 9: Summary and general discussion 135

1. Brain regional differences in the modulation of the NMDA 135 receptors by phosphorylation events

1.1 DARPP-32 1361.2 Possible physiological relevance of PKA-mediated NMDA receptor 138

modulation in the hypothalamus1.3 Direct versus indirect phosphorylation of the NMDA receptor 138

2. Brain regional differences in gene expression 1392.1 Specificity in signaling pathways 140

3. Synaptic plasticity and learning and memory 1433.1 Metaplasticity 1433.2 Correlation between synaptic plasticity and learning 146

3.2.1 Possible role of calcium/calmodulin-dependent protein 147 kinase II

3.2.2 Comparison between two mouse inbred strains Balb/c and 148 C57BL/6N

4. Concluding remarks 149

Nederlandse samenvatting 159

Supplement: NS-257, a novel competitive AMPA receptor antagonist, 165 interacts with kainate and NMDA receptors.

List of publications 179

Dankwoord 183

Curriculum vitae 185

GENERAL INTRODUCTION

1. Synaptic plasticity

Ever since the recognition of the basic organization of the brain during thelast century, hypotheses about the mechanisms of information storage have focussedon the connection between neurons as the most likely site of action. The earliest ofthese connectionistic hypotheses were those of Ramon y Cajal (Ramon y Cayal,1911) and Tanzi (Tanzi, 1893), who proposed that utilization promoted theformation of new circuits or that existing circuits were strengthened. The moremodern formulation, which underlies much of our current thinking, was that of Hebb(Hebb, 1949), who proposed more specific rules for modification of the connection.These rules have now proved to be remarkably prophetic.

The interconnection between neurons is named synapse. The synapse is themost fundamental unit of information transmission in the nervous system.Information storage, including all forms of memory and behavioral adaptation, arethus believed to emerge from changes in neuronal transmission, both in the short-term and the long-term, a property known as synaptic plasticity. Synaptic plasticityis a highly regulated process, emerging from complex interactions not only at thesynapse itself. It can also be the result of a specific neuronal interplay at themolecular, cellular and system level. Thus, understanding the mechanismsunderlying synaptic plasticity may help to apprehend general learning and memoryprocesses.

Changes in synaptic plasticity are achieved by changes in inhibitory orexcitatory neurotransmission or both. The first part of this thesis deals with themodulation of excitatory neurotransmission. The principal excitatoryneurotransmitter in the brain is glutamate. The regulation of glutamate-mediatedexcitatory neurotransmission has been shown to play a critical role in many aspectsof synaptic plasticity. The phosphorylation of glutamate receptors has beendemonstrated to alter their function, suggesting that they may be targets of variouskinases and phosphatases during the induction and maintenance of synapticplasticity (reviewed in Roche et al., 1994). The aim of this part of the thesis is toinvestigate brain regional differences in the modulation of one of the members of theglutamate receptor family, namely the N-methyl-D-aspartate (NMDA) receptor. Inan expression system the modulation of the NMDA receptor by phosphorylation isstudied. Thereafter the possible physiological role of NMDA receptor modulation isassessed in vivo by monitoring immediate early gene expression.

The second part of the thesis focuses on the possible correlation betweensynaptic plasticity and learning and memory. In particular, the significance of theneuropeptide corticotropin-releasing factor (CRF) and of acute stress onhippocampal synaptic plasticity and learning is investigated.

The following paragraphs will provide more background information and adiscussion of the existing literature on the topics studied in this thesis.

Chapter 1

2

1.1 Basic principles of neurotransmission (Fig. 1)Synapses are the elementary building blocks of neuronal communication.

There are two types of synapses either electrical or chemical, although electricalsynapses are less common than chemical synapses in the brain. At electricalsynapses transmission occurs by means of current flow through gap junctions whichconnect the cytoplasm of the pre- and postsynaptic cells. Transmission across theseelectrical synapses is extremely rapid.

Small molecules, the neurotransmitters, mediate information transferbetween neurons at chemical synapses. Before release, the transmitters are stored insmall membraneous organelles, the synaptic vesicles. Upon arrival of an actionpotential, the synaptic membrane depolarizes causing an opening of presynapticcalcium channels. The resulting rise in intracellular calcium triggers exocytosis ofsynaptic vesicles, which releases the transmitter into the synaptic cleft. Thetransmitter molecules diffuse across the synaptic cleft, which separates thetransmitting cell from the receiving (postsynaptic) cell. They bind to specificmembrane receptors at the presynaptic cell, the autoreceptors, which can regulate therate of transmitter release, or to receptors on the membrane of the postsynaptic cell.Postsynaptic receptors binding neurotransmitter can open Na+ or Ca2+ channels atthe postsynaptic cell, thus causing a depolarization and carrying on the signal to thenext cell. Alternatively, the presynaptic cell can inhibit firing of the postsynaptic cellby release of inhibitory neurotransmitters. There are two types of postsynapticreceptors named ionotropic and metabotropic receptors. Ionotropic receptors aredirectly coupled to ion channels and are responsible for fast chemical transmission.As soon as the neurotransmitter binds to the receptor, the ion channel opens.Common neurotransmitters that mediate signaling of this type are glutamate andacetylcholine, which are usually excitatory neurotransmitters, and glycine and γ-aminobutyric acid (GABA), which are usually inhibitory neurotransmitters. Themetabotropic receptors in the postsynaptic membrane that bind neurotransmitters inthe slow chemical transmission pathway are not directly coupled to ion channels butaffect them or alter the level of intracellular second messengers like adenosine 3',5'-cyclic monophosphate (cAMP), Ca2+ and diacylglycerol (DAG), throughintermediary G-proteins.

If the neurotransmitter would remain bound to the receptor, the postsynapticcell would be in a state of constant depolarization or hyperpolarization. Therefore,neurotransmitters are enzymatically degraded or they can be transported back to thepresynaptic cell via transporters (reuptake). The transmitter-receptor complex canalso be taken back into the cell via invagination of membrane which pinches off toform a vesicle, a process called internalisation. The vesicle then fuses with theendosome and inside the endosome transmitter dissociates from the complex and istransferred to the lysosome for degradation and the receptor is recycled to themembrane. The receptor itself can also become desensitized. After prolongedexposure to its own transmitter a receptor can become refractory to later applicationof the same transmitter.

General introduction

3

Figure 1. Overview of the processes involved in synaptic transmission. The elementsdirectly addressed in this thesis are written italic.

1.2 Second messenger pathwaysChanges in the level of second messengers by metabotropic receptors elicit a

cascade of events leading, among others, to the activation of enzymes controllingprotein phosphorylation. Phosphorylation of proteins is an important mechanism forthe modulation of their function, and is thought to play an important role in synapticplasticity in different brain regions.

Many hormone- and neurotransmitter-stimulated signaling pathways alterthe activities of target proteins including receptors, ion channels, synaptic vesicleproteins and nuclear proteins by reversibly phosphorylating serine and threonineresidues.

CREBP

POSTSYNAPSE

synthesis

storage

degradation

inactivationreuptake

release

autoreceptors

ionotropic receptors

metabotropic receptors(e.g. dopamine or CRF)

Ca2+Na+

second messenger pathways(e.g. cAMP, PKC, CaMKII)

neurotransmittersynaptic vesicle

PRESYNAPSE

Na+

NMDA

AMPA/KA

NUCLEUS

Ca2+

voltage-gated channels

genes (c-fos)

transcription

translation

Na+

Chapter 1

4

1.2.1 Protein kinasesTarget proteins such as the NMDA receptor channel complex, which are

involved in synaptic plasticity, can be phosphorylated by serine/threonine proteinkinases. cAMP-dependent protein kinase (PKA), protein kinase C (PKC), andcalcium/calmodulin-dependent protein kinase (CaMK) are the three principalserine/threonine kinases, which catalyze protein phosphorylation in response tosecond messengers such as cAMP, Ca2+ and/or DAG.

Mammalian PKA includes four regulatory (RIα, RIβ, RIIα, RIIβ) and threecatalytic (Cα, Cβ, Cγ) subunits, each encoded by a unique gene. PKA consists of aninactive heterotetramer of two catalytic subunits bound to two regulatory subunits(Taylor et al., 1990). Together the regulatory and catalytic subunits form thefunctional enzyme, the holoenzyme. Subunit assembly in the holoenzyme is likely todiffer because each subunit shows a distinct expression pattern across the brainregions (Cadd and McKnight, 1989). PKA is activated by cAMP, which is generatedby adenylyl cyclase via G protein-coupled receptor activation. When cAMP binds tothe regulatory subunit of PKA, the holoenzyme dissociates to yield a regulatorysubunit dimer and two activate catalytic subunits.

Increases in intracellular Ca2+ concentrations via influx from theextracellular space and via mobilization from intracellular stores byinositoltriphosphate activate PKC and CaMK. PKC is a group of Ca2+ andphospholipid-dependent serine/threonine kinases. It exists of a family of isozymesthat differ in structure, cofactor requirement and substrate specificity (Nishizuka,1988; Dekker and Parker, 1994). These multiple isoforms of PKC also show distinctpatterns of tissue expression and subcellular localization (Tanaka and Saito, 1992).Activation of PKC requires an increase in intracellular Ca2+ and DAG and/orunsaturated fatty acids depending on the isoform involved. Binding of DAG incombination with Ca2+ and phosphatidylserine deinhibits the kinase by producing aconformational change that releases the catalytic site from a pseudosubstratesequence. Activation of several isoforms is also characterized by a translocationfrom cytoplasm to membrane.

When the intracellular Ca2+ concentration rises, the concentration ofCa2+/calmodulin (CaM) complex increases. CaM binds to an autoinhibitory domainon the CaM kinases and thereby activates these enzymes. Four different types ofCaMK (I, II, III, IV) have been described but main attention concentrates onCaMKII which is also called multifunctional CaMK. CaMKII constitutes up to 2%of the total protein content in certain brain areas such as the hippocampus (Eronduand Kennedy, 1985) and is localized in postsynaptic densities of excitatory synapsesas the most abundant protein and as such its participation in mechanisms of synapticplasticity has been widely discussed (Kennedy, 1989; Soderling, 2000). CaMK ismade up of a multiple gene family, in which each of the four distinct classes ofCaMK (α, β, γ, δ) is encoded by a separate gene (Tobimatsu and Fujisawa, 1989).Within a class several isoforms have been identified and both homomultimers andheteromultimers exist (Bennett et al., 1983; Kanaseki et al., 1991). The holoenzymeis arranged in a so called hub-and-spoke pattern, which enables uniqueautoregulatory functions (Kanaseki et al., 1991). Autophosphorylation at threonine-286 (Thr286) increases the affinity of the molecule to calmodulin (Meyer et al.,1992) which makes it Ca2+ independent. Moreover, phosphorylation of the


5

autonomy site is sufficient to disrupt the autoinhibitory domain, through which thekinase stays partially active even after calmodulin dissociates. Hence, assay ofautonomous CaMK activity from control and treated groups gives a direct measureof the fraction of the kinase that has been activated by stimulation of the cells.

1.2.2 Protein phosphatasesAlthough former studies mainly concentrated on the role of protein kinases

in the modulation of target proteins by phosphorylation, it has recently becomeapparent that dephosphorylation of proteins by phosphatases plays an equallyimportant role in alterations of synaptic transmission. Serine/threonine phosphatasescan be divided in two major classes according to similarities in amino-acidsequence. The first class shares a common phosphatase domain and includesphosphatase 1 (PP1), protein phosphatase 2A (PP2A) and calcineurin (PP2B).Recently, molecular cloning identified several more protein serine/threoninephosphatases (PP4, PP5, PP6, and PP7) belonging to this class. The proteinphosphatase 2C (PP2C) class consists of several closely related isoforms that havevery little sequence homology with the first class (Cohen, 1997).

PP1 has emerged as a prominent regulatory element in synaptic plasticity. Itis a multifunctional enzyme that controls the phosphorylation status and activity of avariety of downstream effector molecules that are known to govern synaptic strength(Greengard et al., 1999). These include NMDA (Snyder et al., 1998) and alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) (Yan et al., 1999)glutamate receptors, plus additional components of the calcium signaling cascade,such as CaMKII (Strack et al., 1997) and cAMP response element-binding protein(CREB; Bito et al., 1996). The native structure of PP1 is a 1:1 complex between thecatalytic and a number of different regulatory/targeting subunits. Although presentthroughout the brain, at least four isoforms of the PP1 catalytic subunit areexpressed at varying levels in different regions. For example, PP1α and PP1γ1 areabundant in striatum whereas PP1β is much less prevalent (Da Cruz e Silva et al.,1995). Within neurons, PP1 is highly enriched in dendritic spines and is thereforeappropriately localized for the regulation of excitatory synaptic transmission(Ouimet et al., 1995). The substrate specificity may be controlled by a number ofdifferent regulatory subunits both in different tissues and within the same tissue.

PP2A is composed of a common core enzyme comprising a scaffold subunitthat is always associated with the PP2A catalytic subunit. The major native forms ofPP2A are heterotrimers in which the core enzyme associates with one of a variety ofregulatory subunits that are expressed in a cell- and tissue-specific manner. Theprecise roles of PP2A in dephosphorylation of brain proteins are not clearlydelineated. However, it was shown to dephosphorylate voltage-sensitive sodiumchannels (Chen et al., 1995) and it selectively dephosphorylated solubleautophosphorylated CaMKII (Strack et al., 1997). On the basis of this finding PP2Awas even proposed to be a possible regulator of protein kinases (Barnes et al., 1995;Millward et al., 1999).

PP2B or calcineurin is unique in its requirement for Ca2+ and calmodulinfor activation. It is a heterodimeric phosphatase composed of a catalytic subunit anda regulatory subunit. Binding to the CaM complex (Klee et al., 1998) activates theenzyme. It is widely distributed within the brain, with the highest concentrations in

Chapter 1

6

the hippocampus and caudate putamen (Klee et al., 1988). Interestingly, calcineurinbecomes activated at much lower Ca2+ concentrations than CaMKII. Onceactivated, it has a rather restricted substrate specificity. Phosphoproteins and alsoCaMKII are preferentially dephosphorylated by calcineurin.

Although PP2C is present in the brain (Kasahara et al., 1999; Strack et al.,1997), little is known about the role of this phosphatase in synaptic plasticity, partlybecause of the lack of appropriate pharmacological tools. One likely substrate isCaMKII, which is dephosphorylated by PP2C at its autophosphorylation site(Fukunaga et al., 1993; Strack et al., 1997).

1.2.3 PhosphoproteinsA number of mechanisms can coordinate the actions of kinases and

phosphatases. These include changes in their expression level, their subcellularlocalization, and phosphorylation of catalytic and regulatory subunits. The action ofprotein kinases and phosphatases can also be modulated via the intervention ofprotein phosphatase inhibitors, which are activated by second messenger-regulatedprotein kinases. Concomitant control of kinases and phosphatases by theseendogenous phosphatase inhibitors provides the cell with the capacity to rapidlyswitch proteins from their phosphorylated to their dephosphorylated state.

Inhibitor-1 was the first such protein to be identified, being found in skeletalmuscle (Nimmo and Cohen, 1978). Inhibitor-1 becomes inhibitory to PP1 afterphosphorylation on Thr35 by PKA (Endo et al., 1996; Foulkes et al., 1983) and canbe dephosphorylated by calcineurin and PP2A (Shenolikar, 1994). It is alsophosphorylated at serine-67 (Ser67) by a cyclin-dependent kinase family member(Cdk5) which makes it a less efficient substrate for PKA (Bibb et al., 2001).Inhibitor-1 shows a ubiquitous distribution in the brain (Hemmings et al., 1992;Gustafson et al., 1991; MacDougall et al., 1989).

Dopamine- and cAMP-regulated phosphoprotein with an apparent Mr 32000(DARPP-32) is a homologue of inhibitor-1 that inhibits PP1 with the same potencyupon phosphorylation at Thr34 by PKA. DARPP-32 is also an excellent substratefor phosphorylation by cyclic GMP-dependent protein kinase (PKG) in vitro(Hemmings et al., 1984). It differs from inhibitor-1 in that it can also bephosphorylated on Ser45 and 102 by casein kinase II which promotesphosphorylation at Thr34 by PKA (Girault et al., 1989). Furthermore, DARPP-32can also be phosphorylated by casein kinase I at Ser137 (Desdouits et al., 1995a).Phosphorylation at this site inhibits dephosphorylation of the regulatory Thr34 site(Desdouits et al., 1995b). DARPP-32 is dephosphorylated in vitro by calcineurin andPP2A at Thr34 (Hemmings et al., 1984; King et al., 1984; Nishi et al., 1997, 1999)whereas PP2C can dephosphorylate DARPP-32 at Ser137 (Desdouits et al., 1998).Interestingly, Cdk5 can also phosphorylate DARPP-32 at Thr75 (Bibb et al., 1999).Phosphorylation of DARPP-32 at Thr75 inhibits PKA by a competitive mechanismand, thus, decreases the efficacy of dopaminergic signaling. However, dopamine,through the activation of PKA, can also increase the activity of PP2A, leading to thedephosphorylation of Thr75. In this way, activated PKA attenuates its owninhibition. This positive feedback loop will result in amplification of signalingthrough the dopamine / PKA / Thr34-DARPP-32 / PP1 signal transduction cascade(Nishi et al., 2000). DARPP-32 shows a strong region-specific distribution. It is


7

highly enriched in brain regions with dopaminergic innervation and striatonigralneurons like the caudatoputamen, nucleus accumbens and olfactory tubercle(Ouimet et al., 1984; Perez and Lewis, 1992; Walaas and Greengard, 1984).

A specific substrate for PKG, termed G-substrate with a high homology toDARPP-32, is highly expressed in Purkinje cells of the cerebellum (Detre et al.,1984). Purified recombinant G-substrate has been shown to be an inhibitor of PP1.The inhibition was dependent upon phosphorylation by PKG (Hall et al., 1999).Despite its structural similarity to DARPP-32, which is highly selective for PP1,recombinant G-substrate inhibits PP2A more effectively than PP1, suggesting adistinct role as a protein phosphatase inhibitor (Endo et al., 1999).

Inhibitor-2, unlike inhibitor-1, G-substrate or DARPP-32, does not need tobe phosphorylated to inhibit PP1. It has no sequence identity to inhibitor-1 and theonly physiological role proposed for inhibitor-2 is in the control of sperm motility(Vijayaraghavan et al., 1996).

Heterogeneity in the regulation and distribution of these protein kinases,phosphatases and inhibitor proteins leads to the possibility that each protein mightbe activated in specific cell types and intracellular compartments and may be a partof distinct cellular pathways involved in synaptic plasticity. Brain regionaldifferences in the phosphorylation of target proteins can thus be expected. In thisthesis we investigate the impact of phosphorylation events on the NMDA receptor ina brain region-specific manner.

1.3 Glutamate receptorsGlutamate receptors are divided into two broad categories: (1) metabotropic,

G protein-coupled glutamate receptors (mGluR) that modify neuronal and glialexcitability through G protein subunits acting on membrane ion channels and secondmessengers such as DAG and cAMP and (2) ionotropic receptors, which are ligand-gated ion channels (Ozawa et al., 1998).

There are three families of ionotropic receptors named NMDA, AMPA andkainate based on electrophysiological and pharmacological properties (Hollmannand Heinemann, 1994). In this thesis, we will concentrate on the role of the NMDAreceptor in changes in synaptic strength.

1.3.1 NMDA receptor (Fig. 2)NMDA receptors are characterized by a voltage-dependent Mg2+ block

(Nowak et al., 1984). Upon binding glutamate, the NMDA receptor allows the influxof both Na+ and Ca2+ ions (Ascher and Nowak, 1988). The ability to flux Ca2+ isessential for the special role that NMDA receptors play in synaptic plasticity andneurotoxicity (Malenka and Nicoll, 1993; Perkel et al., 1993). Besides binding ofglutamate to the receptor, glycine is required as coagonist for the NMDA receptorchannel to enter the open state (Kleckner and Dingledine, 1988).

The subunits that make up NMDA receptors have been identified bymolecular cloning techniques and have been characterized as NR1, NR2A–NR2D.Functional NMDA receptors are heteromultimeric complexes of the NR1 subunit incombination with at least one of the four NR2 subunits. Coassembly between thevarious NR1 and NR2 subunits generates functionally distinct types of NMDA

Chapter 1

8

receptors. Incorporation of different NR1 splice variants into NMDA receptorcomplexes influences receptor properties such as modulation by zinc, polyaminesand PKC, as well as binding to intracellular proteins (Durand et al., 1993; Ehlers etal., 1996; Hollmann et al., 1993; Lin et al., 1998). The NR2 subunit compositiondetermines biophysical characteristics of the channel such as conductance, meanopen time, and sensitivity to Mg2+ block (Monyer et al., 1992, 1994; Stern et al.,1992).

Figure 2. Schematic drawing of the NMDA receptor complex

NMDA receptors are found throughout the brain but predominantly withinthe forebrain. The highest levels in the entire brain are found in the CA1 region ofthe hippocampus. NR1 mRNA is distributed ubiquitously but the four NR2 subunitsdisplay distinct regional patterns. NR2A and NR2B are mainly found in theforebrain, NR2C in the cerebellum and NR2D in thalamus, brain stem and olfactorybulb (Kutsuwada et al., 1992; Monyer et al., 1992, 1994). Since the functionalproperties of the NMDA receptor depend on coassembly of the four NR2, it isconceivable that spatial patterns of expression of the NR2 genes lead to regionspecific differences in receptor function.

Recently, an additional NMDA receptor subunit, NR3A, has been identifiedin mammalian brain (Ciabarra et al., 1995; Sucher et al., 1995). To date, only a fewstudies have reported evidence for a functional role of this subunit in the brain (Daset al., 1998; Pérez-Otaño et al., 2001).

Several studies have reported that phosphorylation of NMDA receptors canmodify their function. NMDA receptors have been shown to be modulated by PKCphosphorylation (Ben-Ari et al., 1992). Phorbol esters, which activate PKC,selectively enhance NMDA receptor currents in oocytes. Activation of G-proteincoupled receptors that activate PKC also enhances NMDA-mediated ion currents


9

(e.g. Blank et al., 1996; Chen and Huang, 1992; Urushihara et al., 1992).Unexpectedly, phorbol esters have also been reported to depress rather than enhanceNMDA responses in hippocampal neurons (Markram and Segal, 1992) but theconcentrations of phorbol ester used in these experiments were much higher thanthose required to stimulate PKC in vitro. Thus, the inhibition may be attributable toa direct block of the channels (Hockberger et al., 1989). Interestingly, Leonard andHell (1997) demonstrated that NR1, NR2A and NR2B can be directlyphosphorylated by PKA and various PKC isoforms in vitro and that the majorphosphorylation sites for PKA and PKC differ for all three subunits. However, itremains rather unclear whether PKA directly phosphorylates NMDA receptors invivo. Nevertheless, reagents that increase adenylate cyclase and PKA activity canenhance NMDA-receptor-mediated excitatory postsynaptic currents/potentials inspinal dorsal horn (Cerne et al. 1993), amygdala (Huang et al. 1993), neostriatum(Colwell and Levine 1995) and in hippocampal microcultures and slices (Raman etal., 1996; Yang, 2000).

The CaMKII regulatory site is conserved in NMDA receptors suggestingthat CaMKII activity influences these receptors (Omkumar et al., 1996). Consistentwith this suggestion, intracellular application of the α-subunit of CaMKII enhancesNMDA receptor-mediated ion currents (Kolaj et al., 1994). It was recently foundthat NMDA receptor subunits NR2A/B are a target for CaMKII in the postsynapticdensity (Gardoni et al., 1998, 1999; Strack and Colbran, 1998) but the exact natureof these complex interactions still requires further investigation.

Involvement of phosphatases in NMDA receptor regulation has beenreported (Lieberman and Mody, 1994; Yakel, 1997). Phosphatase inhibitors enhanceNMDA receptor-mediated ion currents, whereas injected PP1 or PP2A decreases theopen probability of these channels (Wang et al., 1994).

2. Synaptic plasticity and gene expression (Fig. 3)

Changes in synaptic plasticity resulting from the interplay of various signaltransduction pathways can be translated into long-term effects via the regulation ofgene expression.

Neurotransmitters released from the presynaptic cell regulate geneexpression in the postsynaptic cell by activation of specific postsynaptic receptors.The activation of these receptors leads to the stimulation of second messengerpathways, which transduce the signal to the nucleus where expression of specificgenes is regulated (Fig. 1). The expression of these genes is rapid and transient andthus, they are often referred to as immediate early genes (IEG). Neurotransmitterscan act via distinct signal transduction pathways to induce IEG expression in variousneuronal populations. Induction of IEGs occurs independently of de novo proteinsynthesis and is believed to be mediated by the post-translational modifications ofproteins already present in the cell. Many IEG encode transcription factors that areinvolved in the regulation or expression of so-called late response genes. Theseinduced genes may in turn be involved in changes of synaptic plasticity.

In addition to being a putative messenger involved in transduction of signalsto gene expression, activation of IEGs has proven to be a useful tool for the study of

Chapter 1

10

transsynaptic activation of certain neuronal pathways in the brain (Bullitt, 1990;Sagar et al., 1988). Especially the expression of c-fos is a useful mapping tool sinceits level of transcription is low under basal conditions (Hughes et al., 1992) and it isinduced upon a wide range of stimuli.

Signaling pathways involving the constitutive transcription factor CREB areparticularly important for synaptic plasticity and the conversion of short-term tolong-term memory (Bailey et al., 1996). Therefore, in this thesis we concentrate onchanges in the expression level of FOS protein and phosphorylated CREB followingthe activation or inhibition of NMDA receptors in vivo.

2.1 CREBCREB mediates gene expression through a Ca2+/cAMP responsive element

(CRE) on target genes. CREB is a member of a multigene family of structurallyrelated proteins that all bind to the CRE promotor as a dimer (Yamamoto et al.,1988). It is a continually expressed transcription factor, i.e. its expression is drivenby ‚stand alone‘ mechanisms within cells and is not dependent upon extracellularsignals (Gonzales and Montminy, 1989; Montminy et al., 1987; Yamamoto et al.,1988) but a phosphorylation event activates the ability of CREB to stimulatetranscription. There are several pathways leading to CREB phosphorylation.Activated PKA can rapidly phosphorylate CREB at Ser133 (Gonzales andMontminy, 1989), whereas a rise in intracellular Ca2+ levels by Ca2+ influx throughvoltage gated Ca2+ channels or NMDA receptors also activates CREB byphosphorylation at Ser133. Ca2+ may activate CREB via several distinct signalingpathways. The best characterized of the Ca2+ activated CREB kinases are the CaMkinases. Although phosphopeptide mapping indicated that CaMKI, CaMKII andCaMKIV all phosphorylate CREB in vitro (Dash et al., 1991; Sheng et al., 1991),CaMKIV in particular seems to mediate Ca2+-dependent CREB activation in vivo(Bito et al., 1996; Kasahara et al., 2001). Another pathway that may contribute toCa2+ stimulation of Ser133 phosphorylation is the mitogen-activated protein (MAP)kinase pathway. In certain neuronal subtypes, Ca2+ influx can also lead to theactivation of PKA, which then phosphorylates CREB. Furthermore, growth factorscan phosphorylate CREB via MAP kinases or members of a family of ribosomalkinases (RSKs) (Xing et al., 1996). Ser133 phosphorylation is considered to be thekey event that mediates the initiation of transcription since mutation of Ser133 toalanine abolishes transcription (Gonzalez and Montminy, 1989) but CREB activityand specificity can be further modulated by phosphorylation of additional sites onCREB or of proteins associated with CREB. CaMKII can phosphorylate CREB alsoat Ser142 which is inhibitory to CREB function (Sun et al., 1994) and PKC is able tophosphorylate CREB in vitro at multiple sites (Yamamoto et al., 1988). Uponphosphorylation of Ser133, CREB becomes bound to a transcription coactivatorprotein, CREB binding protein (CBP), or its cognate relative p300 (Kwok et al.,1994) which is critical for stimulus-induced transcription.

Subsequent attenuation of CREB activity can potentially occur by severaldifferent mechanisms. CREB itself is dephosphorylated by PP1 (Hagiwara et al.,1992; Bito et al., 1996) and CREB dephosphorylation in nuclear extracts iscatalyzed by a specific holoenzyme of PP2A (Wadzinski et al., 1993). However,CRE activity can also be antagonized by the binding of inhibitory transcription


11

factors such as inducible cAMP early repressor (ICER) (Foulkes et al., 1991; Lamasand Sassone-Corsi, 1997; Mellstrom et al., 1993).

2.2 FOSA very well characterized gene that contains the CRE element is the c-fos

gene. The c-fos gene codes for the FOS transcription factor which must dimerizewith Jun-family (c-Jun, Jun-B, Jun-D) transcription factors, using a leucine-zippermotif, for DNA binding. Upon dimerization a transcription factor referred to asactivator protein-1 (AP-1) is formed that stimulates or represses transcription ofother (late) genes (Morgan and Curran, 1991; Sheng and Greenberg, 1990). Theregulatory region of the c-fos gene contains several cis-acting elements besides CRE(e.g. serum responsive element (SRE), sis-inducible element (SIE) and AP-1) thatcan confer the signals mediated by the signal transduction pathways involving PKA,PKC, CaM kinase and MAP kinase cascades.

Figure 3. Schematic overview of the pathways and promotor elements leading to c-fostranscription. The major regulatory upstream elements are Ca2+/cAMP responseelement (CRE), serum response element (SRE), activator protein-1 binding site (AP-1) and sis-inducible element (SIE). The transacting serum response factor (SRF) andcAMP-response element binding protein (CREB) are targets of different signaltransduction pathways, including PKA, PKC, CaMK, MAP kinases and RSK. Otherabbreviations: P, phosphate (modified from Kovacs, 1998).

Increase in cAMP levels results in c-fos activation via binding ofphosphorylated CREB to CRE. However, Ca2+ can induce c-fos expression via twodistinct pathways. Increased Ca2+ levels can promote CREB phosphorylation viaCaM kinase pathways and thus induce c-fos via CRE, or Ca2+ influx leads toactivation of the MAP kinase pathways and targets SRE (Bading et al., 1993; Ghoshet al., 1994). Depending on its route of entry, calcium may influence transcriptionalresponses in neurons either via SRE or CRE (Bading et al., 1993).

In vivo, there may be crosstalk between the multiple elements in theregulation of c-fos expression. Robertson and colleagues (Robertson et al., 1995)

FOS/JUNSRFP

TATA c-fosSIE SRE AP-1 CRE

CREB

DAG

PKC MAP kinases

cAMP

PKACaMKIV

Ca2+

RSK

Chapter 1

12

showed by mutation of different regulatory elements that all regulatory elements arerequired collectively for regulation of the c-fos promotor.

3. Synaptic plasticity and learning and memory

Changing the strength of connections between neurons is widely assumed tobe the mechanism by which memory traces are encoded and stored in the centralnervous system. Memory is the retention or storage of knowledge whereas learningis the process of acquiring that knowledge. There are at least two forms of memory:(1) declarative memory also named explicit or relational memory, is the consciousrecall of facts and events and is particularly well-developed in the vertebrate brain.This is the kind of memory that is usually referred to when the term ‘memory’ or‘remembering’ is used in ordinary language. (2) Implicit or nondeclarative memoryis the nonconscious recall of motor skills and other tasks and includes simpleassociative forms and nonassociative forms, such as sensitization and habituation. Ithas turned out that different kinds of memory are supported by different brainsystems comprising numerous brain areas.

3.1 HippocampusFrom a large amount of literature published on lesions or pharmacological

inactivations in rodents and higher primates including humans, the hippocampalformation was concluded to play a critical role in the initial storage of certain formsof relational long-term memory (Zola-Morgan and Squire, 1993). The hippocampalformation consists of four architectonically distinct regions which include thedentate gyrus, the hippocampus proper, which is subdivided in three fields (CA1,CA2 and CA3), subicular complex (subiculum, presubiculum and parasubiculum)and the entorhinal cortex. The main reason for arranging these regions into thehippocampal formation is that they are linked onto the next by unique and largelyunidirectional projections, which are arranged in parallel, well-defined laminae. Theentorhinal cortex is interconnected to the dentate gyrus via the perforant pathway.The cells in the dentate gyrus project via their distinctive mossy fibers to the CA3field of the hippocampus proper. Afferent fibers from CA3 pyramidal neuronsproject via the Schaffer collateral system to the stratum radiatum, pyramidale, andoriens of CA1 pyramidal neurons. A similar patterns holds for the other intrinsicconnections (e.g. Amaral, 1991, 1993). Detailed knowledge of its anatomy andconnections together with its highly laminated pattern of inputs and outputs make itan area well suitable for electrophysiological recordings. Because of this and itswell-known role in learning and memory, experiments in the final part of this thesis,which concentrate on the relation between changes in synaptic strength and learningand memory, focus on this brain structure.

3.2 Long-term potentiation and depression as a model for learning and memoryThe brain uses short- and long-lasting modifications in synaptic strength in

critical neuronal circuits to process and store large amounts of information. Onesuch activity-dependent modification is long-term potentiation (LTP) in thehippocampus, a sustained increase in synaptic strength that is elicited by brief high


13

frequency or theta-burst stimulation of excitatory afferents. Mechanisms underlyingLTP have been proposed as possible mechanisms underlying memory formation inthe brain. This is in part due to the properties LTP displays (Bliss and Lomo, 1973;Nicoll et al., 1988; Bliss and Collingridge, 1993) and supported by the observationthat a LTP-like phenomenon can be seen in brains of animals successively learning abehavioral task (Berger, 1984; Moser et al., 1994; Sharp et al., 1985; Skelton et al.,1985). Like memories, LTP can be generated rapidly and is prolonged andstrengthened with repetition. It exhibits input specificity, that is LTP occurs only atsynapses stimulated by afferent activity not at adjacent synapses at the samepostsynaptic cell, and is associative. Temporally pairing activity in a weak inputwith activation of a strong input results in LTP of the weak input. Furthermore,pharmacologically blocking LTP interferes with the acquisition of behaviorallearning (e.g. Davis et al., 1992; Morris et al., 1986) and targeted gene knockouts ofvarious protein kinases which impair LTP generation also impair spatial learning(Abeliovich et al., 1993; Grant and Silva, 1994).

Several forms of LTP in the hippocampus have been described so far. Oneform of LTP features the activation of voltage-dependent calcium channels on thepostsynaptic cell (Grover and Teyler, 1990; Cavus and Teyler, 1996). The mostextensively studied form of LTP is mediated by activation of the NMDA receptor,which upon gating Ca2+ in the postsynaptic cell, induces the processes that lead toLTP expression. However, at certain synapses a robust form of LTP can begenerated that does not require NMDA receptor activation. In the hippocampus thisform of LTP occurs at mossy fiber synapses (Nicoll and Malenka, 1995).

With respect to mechanisms of action underlying LTP initiated via NMDAreceptors, considerable interest and evidence have accumulated regarding the role ofserine/threonine kinases in LTP expression. However, a distinction should be madebetween molecules that play a direct active role in the induction of LTP and thosethat play a modulatory or permissive role. Pharmacological inhibition of CaMKII, aswell as genetic knockout experiments, provides strong evidence for a key role ofCaMKII in LTP. The experimental evidence for other kinases like PKA and PKC isconsiderably weaker. The initial activation of PKA has been suggested to boost theactivity of CaMKII indirectly via decreased phosphatase activity by activation ofinhibitor-1 (Blitzer et al., 1998; Makhinson et al., 1999). PKC may play a roleanalogous to CaMKII as PKC inhibitors have been reported to block LTP, and PKCinjection into hippocampal cells has been shown to elicit features of LTP (Hu et al.,1987; Klann et al., 1991; Linden and Routtenberg, 1989; Sacktor et al., 1993).Activation of signal transduction pathways may ultimately lead to alterations in geneexpression.

No question concerning LTP has generated more debate and confusion overthe last two decades than the question whether the increase in synaptic strength isdue primarily to a pre- or postsynaptic modification. Most neurobiologists agree thatthe most obvious postsynaptic change that could cause LTP, would be amodification in AMPA receptor function and/or number because LTP increases theAMPA receptor mediated current to a greater extent than the NMDA receptorcurrent (Kauer et al., 1988). Much evidence has accumulated that phosphorylation ofthe AMPA receptor subunit GluR1 by CaMKII is critically important for the changein AMPA receptor responsiveness (Barria et al., 1997). Furthermore, there is now

Chapter 1

14

reasonably strong evidence that supports the model of a rapid and selectiveupregulation and trafficking of AMPA receptors after the induction of LTP whichcan be regulated by CaMKII (Hayashi et al., 2000; Lu et al., 2001; Luscher et al.,1999; Shi et al., 1999). However, it must be noted that these studies were performedin newborn rats (Hayashi et al., 2000; Shi et al., 1999). In mature animals LTP doesnot alter membrane association of AMPA receptors but LTP leads to a rapid surfaceexpression of NMDA receptors (Grosshans et al, 2002).

Besides long-lasting changes on the postsynaptic terminal, recent papersprovide additional evidence for a candidate role of enhanced transmitter release inLTP (Zakharenko et al., 2001; Tyler and Pozzo-Miller, 2001).

If memories are stored in spatial patterns of synaptic strength, it would beimportant to be able to decrease as well as increase synaptic strength to improve theflexibility and storage capacity of the system. Long-term depression (LTD) has beendescribed as a way to decrease synaptic strength. The mechanisms of hippocampalLTD are not so well understood as those for LTP but, although some commonprocesses exist, there is increasing evidence that LTD is not mechanistically thereverse of LTP (reviewed in: Braunewell and Manahan-Vaughan, 2001). In the CA1area NMDA receptors as well as mGluRs appear to be critically involved in the LTDinduction process (Mulkey and Malenka, 1992; Otani and Connor, 1998). For LTD,a moderate calcium influx preferentially leads to activation of calcium-dependentprotein phosphatases such as PP1 and PP2A (Mulkey et al., 1993), and to theinactivation of kinases such as CaMKII and PKC (Bear and Abraham, 1996).

Although LTP mechanisms have been shown to be an attractive candidate asa mechanisms for memory formation, it is important to note that there is littleempirical evidence that directly links LTP to the storage of memories (Barnes,1995;Goda and Stevens, 1996; Holscher, 1999; Martin et al., 2000). For example, Gu andcolleagues (2002) observed impaired conditioned fear but enhanced LTP in Fmr2knock-out mice. Instead of being a learning mechanism, LTP is also suggested torather serve as a neural equivalent to an arousal or attention device in the brain andaccordingly could act to facilitate and maintain learning indirectly by altering theorganism’s responsiveness to, or perception of, environmental stimuli (Shors andMatzel, 1997).

3.3 Influence of stress on synaptic plasticity and learning and memoryStress, which is defined as an event or events that are interpreted as

uncontrollable and threatening to an individual and which elicit physiological andbehavioral responses, has been shown to decrease LTP in the hippocampus as firstreported by Foy and colleagues (Foy et al., 1987). They found reduced LTP inhippocampal slices obtained from rats subjected to physical stress, i.e., restraint andelectric shock. Since then it was demonstrated in numerous studies that the inductionof hippocampal LTP by high frequency stimulation or by multiple bursts ofelectrical pulses delivered in a theta-related pattern is inhibited after social stress(Von Frijtag et al., 2001), restraint stress and inescapable tail shock (Foy et al.,1987; Garcia et al., 1997; Kim et al., 1996; Shors et al., 1989; Shors et al., 1990a,b;Shors and Dryver, 1994). LTD observed after low frequency stimulation is


15

facilitated after social stress (Von Frijtag et al., 2001), restraint stress andinescapable tail shock (Kim et al., 1996; Xu et al., 1997). Furthermore, Diamondand coworkers (Diamond et al., 1994; Mesches et al., 1999) found that exposing ratsto a predator or to a novel environment, both representing acute psychological stress,blocks primed burst potentiation, which is a long-term increase in CA1 populationspike amplitude produced by brief physiologically patterned electrical stimulation ofthe hippocampal commissure, but not conventional LTP.

In addition, it might be suggested from several experiments that stress alsoimpairs short-term potentiation in the CA1 area of hippocampal slices preparedimmediately after the stress episode in vitro (Garcia et al., 1997; Kim et al., 1996) aswell as in vivo (Diamond et al., 1994).

Several groups whose stress paradigm had a negative effect on hippocampalsynaptic plasticity also found impairment of spatial learning by this stressor(Bodnoff et al., 1995; de Quervain et al., 1998; Diamond and Rose, 1994; Diamondet al., 1996, 1999). Interestingly, various other studies show that stress can improvelearning (Maier, 1990; Radulovic et al., 1999; Servatius and Shors, 1994; Shors andMathew, 1998; Shors and Servatius, 1995; Shors et al., 1992; Wilson et al., 1975).For example, repeated restraint stress (Conrad et al., 1999) or inescapable footshock(Deak et al., 1999) enhance contextual fear conditioning which represents ahippocampus-dependent learning task (Holland and Bouton, 1999; Kim andFanselow, 1992; Phillips and LeDoux, 1992). Thus, stressful experiences can impair(Krugers et al., 1997; Luine et al., 1994) or enhance learning and memory dependingon the type and duration of the stressor, learning task, animal species and a numberof other experimental conditions

In this thesis we use acute immobilization stress or the neuropeptidecorticotropin-releasing factor (CRF), which represents an early stress signal, toinvestigate the relation between altered synaptic plasticity and changes in learning.

3.3.1 Corticotropin-releasing factor as mediator of the stress responseEndocrine responses to stress are mediated by release of CRF. CRF is part

of the body’s stress axis, the hypothalamo-pituitary adrenal axis (HPA-axis) where itstimulates hypophyseal adrenocorticotropic hormone (ACTH) secretion, whichsubsequently elicits adrenal glucocorticoid release (Spiess et al., 1981; Vale et al.,1981). Besides these neuroendocrine actions, CRF also acts in the brain to modulatebehavioral, autonomic and neuroendocrine responses to stress (reviewed in: Owensand Nemeroff, 1991). Two CRF receptors, encoded by distinct genes, CRFR1 andCRFR2, which can exist in two alternatively spliced forms, have been cloned inrodents (Chang et al., 1993; Lovenberg et al., 1995a). The type-1 CRF receptor isexpressed in many areas of the rodent brain and in several peripheral sites. Highestdensities of CRFR1 mRNA have been found in the pituitary, several areas of thecortex, amygdala, cerebellum, parts of the hippocampus and the olfactory bulbs(Chalmers et al., 1995, Van Pett et al., 2000). In the rodent brain, the CRFR2αreceptor variant is primarily found in neuronal brain structures such ashypothalamus, lateral septum and amygdala, whereas the CRF2β subtype isexpressed in non-neuronal structures and in peripheral organs (Chalmers et al., 1995;Grigoriadis et al., 1996; Lovenberg et al., 1995b; Van Pett et al., 2000). The CRFreceptors are members of a 7-transmembrane receptor family that includes GH-

Chapter 1

16

releasing hormone (GRF), calcitonin, vasoactive intestinal peptide (VIP), secretin,and PTH receptors (VIP/GRF/PTH receptor family), whose actions are mediatedthrough activation of adenylate cyclase. In addition to its interaction with membranereceptors, CRF binds with high affinity to a CRF binding protein (CRF-BP)(Cortright et al., 1995; Potter et al., 1991). The distribution of CRF-BP appears to bepredominant in the neocortex, the limbic system and the brain stem (Potter et al.,1992).

After the isolation of CRF, three structurally-related peptides were identifiedin vertebrate species, namely urocortin (Vaughan et al., 1995), urocortin II (alsoknown as stresscopin-related peptide) (Hsu and Hsueh, 2001; Reyes et al., 2001) andurocortin III (also known as stresscopin) (Hsu and Hsueh, 2001; Lewis et al., 2001).These peptides differ in localization and their binding affinity to CRFR1 and CRF2.CRF is relatively selective for both CRFR1 and CRFR2. Urocortin is bound to bothCRFR1 and CRFR2 with high affinity (Chen et al., 1993; Lovenberg et al., 1995).Urocortin II and III are selective ligands for CRFR2 (Hsu and Hsueh, 2001) Neitherurocortin II nor urocortin III binds to CRF-BP whereas CRF and urocortin do. Thediscovery of these new ligands and the development of selective receptor antagonistshas opened new avenues for investigating the role of the two CRFRs in synapticplasticity and behavior under normal and pathological conditions. In this thesis thefocus was put on CRF.

Electrophysiological studies provided evidence for an excitatory action ofCRF in several brain areas like locus coeruleus, solitary tract, hippocampus andsome regions of the hypothalamus (Conti and Foote, 1995; Hollrigel et al., 1998;Siggins et al., 1985; Smith and Dudek, 1994; Yamashita et al., 1991). Exogenousapplication of CRF to hippocampal slices was shown to make the pyramidal neuronsmore excitable i.e. to reduce the slow afterhyperpolarization and spike frequencyaccommodation (Aldenhoff et al., 1983; Haug and Storm, 2000; Smith and Dudek,1994), and to enhance the amplitude of CA1 population spikes evoked bystimulation of the Schaffer collateral pathway in rats (Hollrigel et al., 1998). CRFalso increases the amplitude of orthodromically evoked (stratum radiatumstimulation) population spikes in rat hippocampus (Smith and Dudek, 1994).

Like stress, CRF can modulate learning and memory, either enhancing orimpairing retention in a time, dose and site specific manner. Intracerebroventricularinjections of CRF or its displacement from CRF-BP levels before or immediatelyafter training improve memory in multiple learning tasks (Behan et al., 1995,Heinrichs et al., 1997; Koob and Bloom, 1985; Liang and Lee, 1988; Radulovic etal., 1999). CRF directly injected into the dentate gyrus consistently enhancedmemory retention in rats in a one-way passive avoidance task (Lee et al., 1993).Intrahippocampal infusion of antisense oligodeoxynucleotides (ODNs) directedagainst CRF mRNA has been reported to impair passive avoidance retention (Wu etal., 1997). However, intracerebroventricular CRF injection before the memory testor overexpression of CRF in mice seems to impair memory (Diamant and De Wied,1993; Heinrichs et al., 1996). Especially, the presence of CRFR1 in thehippocampus suggests a role for CRF in learning and memory processes (Radulovicet al., 1999). The role of CRF in the hippocampus is further supported byelectrophysiological data, which show that CRF induces a long-lasting enhancementof synaptic efficacy in the hippocampus (Wang et al., 1998) which appears to be


17

critically dependent on protein synthesis (Wang et al., 2000). Wang and colleaguesfound that application of CRF to the dentate gyrus increased the populationexcitatory postsynaptic potential as well as the population spike. Furthermore,pretreatment with a CRF receptor antagonist dose-dependently diminishedtetanization induced LTP, while CRF and LTP had an additive effect on cellexcitation of hippocampal neurons. On the basis of these findings, they suggest thatCRF-induced potentiation and LTP may share some similar mechanisms, and thatCRF is probably involved in the neuronal circuits underlying LTP.

Aim and outline of this thesis:

Synaptic plasticity, i.e. changes in synaptic strength, is critical for themaintenance and adaptation of brain function and is assumed to be the neuronalbasis of learning and memory processes in the brain.

A multidisciplinary approach is used to gain more understanding on themolecular mechanisms of synaptic plasticity in the central nervous system. Themajor aim of this thesis can be divided in two questions:

1. Are there brain regional differences in the modulation of NMDA receptorsignaling that could contribute to the regulation of synaptic plasticity?

2. To what extent can changes in hippocampal synaptic plasticity accountfor stress- or CRF-induced alterations in hippocampus-dependent learning andmemory?

As described in the introduction, the phosphorylation state of the NMDAreceptor is particularly important to mediate changes in synaptic strength. Thus,brain region-specific modulation of the NMDA receptor is expected to provide moreinsight into regional differences in the regulation of synaptic plasticity. In the firstpart of this thesis we investigate whether the NMDA receptor from various brainregions can be modulated by PKA and PKC and protein phosphatases. Using theoocyte expression system, it is demonstrated that stimulation of PKC potentiatedNMDA receptor function from all brain regions investigated, whereas the activationof PKA increases NMDA receptor function only in few defined regions.

In Chapter 2, the modulation of hypothalamic NMDA receptor function byPKA and phosphatases is examined.

Chapter 3 describes experiments performed in oocytes, which are injectedwith mRNA isolated from the striatum. NMDA receptor modulation by PKA andphosphatases and the involvement of an endogenous protein phosphatase inhibitorare investigated.

The region-specific differences in the ability of PKA to regulate NMDAreceptor function as described in Chapters 2 and 3 suggest that neurotransmittersystems which are linked to the cAMP system may control NMDA receptorphosphorylation and thereby the efficacy of synaptic transmission in a brain region-distinctive manner.

In many models of synaptic plasticity, increased activation of thetranscription factor CREB and c-fos gene expression is for the induction and

Chapter 1

18

maintenance of long-lasting changes of synaptic efficacy. Thus, changes inimmediate early gene expression may provide significant understanding ofintracellular mechanisms underlying synaptic plasticity in the brain areas examined.Since dopamine receptors are thought to modulate PKA activity, activation of thedopaminergic system is used to investigate the modulation of NMDA receptorfunction in vivo by monitoring immediate early gene expression.

In Chapter 4 we investigate whether FOS protein expression in the limbicsystem and basal ganglia is altered by intraperitoneal injection of dopamine orNMDA receptor agonists and antagonists or both.

The promotor region of the c-fos gene contains a binding site forphosphorylated CREB, and CREB can be phosphorylated by PKA suggesting CREBphosphorylation by PKA as a possible pathway for FOS production. Thus,experiments are performed to gain more information on whether FOS productioncan be correlated with changes in the levels of phosphorylated CREB (Chapter 5).

In the last part of this thesis, we focus on mechanisms of synaptic plasticityin the hippocampus. We examine the relation between stress- or CRF-inducedalterations in synaptic strength and changes in learning and memory. In theintroduction, stress is mentioned as a useful tool to elicit changes in synapticplasticity, and it has been shown to affect learning and memory. The influence ofone-hour immobilization on context-dependent fear conditioning and long- andshort-term hippocampal plasticity is described in Chapters 6 and 7.

Chapter 6 focuses on changes in learning and synaptic plasticityimmediately after the end of the stress session, and the role of CRF receptors isassessed by intraperitoneal injection of the non-peptidic CRF receptor antagonistCP-154,526.

These experiments are continued in Chapter 7 where changes in synapticefficacy and learning are determined 1, 2 and 3 hours after the end of the stresssession.

The involvement of CaMKII (Chapters 6 and 7) and PKC activity (Chapter7) is a main research target.

After establishing the role of CRF in stress-mediated changes in synapticplasticity and learning in Chapter 6 and 7, the effect of CRF on neuronal excitabilityis examined in more detail in Chapter 8. Possible underlying signal transductionpathways of CRF stimulation in mouse hippocampus are investigated in twodifferent mouse strains, Balb/c and C57BL/6N.

In the final chapter (Chapter 9), a summary of the data from the precedingchapters is provided and the data from the separate chapters are discussed in theircontribution to synaptic plasticity.

REFERENCES

Abeliovich A, Chen C, Goda Y, Silva AJ, Stevens CF and Tonegawa S (1993) Modified hippocampallong-term potentiation in PKC gamma-mutant mice. Cell 75: 1253-1262.


19

Aldenhoff JB, Gruol DL, Rivier J, Vale W and Siggins GR (1983) Corticotropin releasing factordecreases postburst hyperpolarizations and excites hippocampal neurons. Science 221: 875-877.

Amaral DG (1991) Is there ‘channeling’ of information through the intrinsic circuit of the rathippocampus? Brain Res. Rev. 16: 193-220.

Amaral DG (1993) Emerging principles of intrinsic hippocampal organization. Curr. Opin. Neurobiol.3: 225-229.

Ascher P and Nowak L (1988) The role of divalent cations in the N-methyl-D-aspartate responses ofmouse central neurones in culture. J. Physiol. 399: 247-266.

Bading H, Ginty DD and Greenberg ME (1993) Regulation of gene expression in hippocampal neuronsby distinct calcium signaling pathways. Science 260: 181-186.

Bailey CH, Bartsch D and Kandel ER (1996) Toward a molecular definition of long-term memorystorage. Proc. Natl. Acad. Sci. USA 93: 13445-13452.

Barnes CA (1995) Involvement of LTP in memory: are we "searching under the street light"? Neuron15: 751-754.

Barnes GN, Slevin JT and Vanaman TC (1995) Rat brain protein phosphatase 2A: an enzyme that mayregulate autophosphorylated protein kinases. J. Neurochem. 64: 340-353.

Barria A, Muller D, Derkach V, Griffith LC and Soderling TR (1997) Regulatory phosphorylation ofAMPA-type glutamate receptors by CaM-KII during long-term potentiation. Science 276: 2042-2045.

Bear MF and Abraham WC (1996) Long-term depression in hippocampus. Annu. Rev. Neurosci. 19:437-462.

Behan DP, Heinrichs SC, Troncoso JC, Liu XJ, Kawas CH, Ling N and De Souza EB (1995)Displacement of corticotropin releasing factor from its binding protein as a possible treatment forAlzheimer's disease. Nature 378: 284-287.

Ben-Ari Y, Aniksztejn L and Bregestovski P (1992) Protein kinase C modulation of NMDA currents:an important link for LTP induction. Trends Neurosci. 15: 333-339.

Bennett MK, Erondu NE and Kennedy MB (1983) Purification and characterization of a calmodulin-dependent protein kinase that is highly concentrated in brain. J. Biol. Chem. 258: 12735-12744.

Berger T (1984) Long-term potentiation of hippocampal synaptic transmission affects rate ofbehavioral learning. Science 224: 627-630.

Bibb JA, Nishi A, O'Callaghan JP, Ule J, Lan M, Snyder GL, Horiuchi A, Saito T, Hisanaga S, CzernikAJ, Nairn AC and Greengard P (2001) Phosphorylation of protein phosphatase inhibitor-1 by Cdk5. J.Biol. Chem. 276: 14490-14497.

Bibb JA, Snyder GL, Nishi A, Yan Z, Meijer L, Fienberg AA, Tsai LH, Kwon YT, Girault JA, CzernikAJ, Huganir RL, Hemmings HC Jr, Nairn AC and Greengard P (1999) Phosphorylation of DARPP-32by Cdk5 modulates dopamine signalling in neurons. Nature 402: 669-671.

Bito H, Deisseroth K and Tsien RW (1996) CREB phosphorylation and dephosphorylation: a Ca(2+)-and stimulus duration-dependent switch for hippocampal gene expression. Cell 87: 1203-1214.

Blank T, Zwart R, Nijholt I and Spiess J (1996) Serotonin 5-HT2 receptor activation potentiates N-methyl-D-aspartate receptor-mediated ion currents by a protein kinase C-dependent mechanism. J.Neurosci. Res. 45: 153-160.

Chapter 1

20

Bliss TV and Collingridge GL (1993) A synaptic model of memory: long-term potentiation in thehippocampus. Nature 361: 31-39.

Bliss TV and Lomo T (1973) Long-lasting potentiation of synaptic transmission in the dentate area ofthe anaesthetized rabbit following stimulation of the perforant path. J. Physiol. 232: 331-356.

Blitzer RD, Connor JH, Brown GP, Wong T, Shenolikar S, Iyengar R and Landau EM (1998) Gating ofCaMKII by cAMP-regulated protein phosphatase activity during LTP. Science 280: 1940-1942.

Bodnoff SR, Humphreys AG, Lehman JC, Diamond DM, Rose GM and Meaney MJ (1995) Enduringeffects of chronic corticosterone treatment on spatial learning, synaptic plasticity, hippocampalneuropathology in young and mid-aged rats. J. Neurosci. 15: 61-69.

Braunewell KH and Manahan-Vaughan D (2001) Long-term depression: a cellular basis for learning?Rev. Neurosci. 12: 121-140.

Bullitt E (1990) Expression of c-fos-like protein as a marker for neuronal activity following noxiousstimulation in the rat. J. Comp. Neurol. 296: 517-530.

Cadd G and McKnight GS (1989) Distinct patterns of cAMP-dependent protein kinase gene expressionin mouse brain. Neuron 3: 71-79.

Cavus I and Teyler T (1996) Two forms of long-term potentiation in area CA1 activate different signaltransduction cascades. J. Neurophysiol. 76: 3038-3047.

Cerne R, Rusin KI and Randic M (1993) Enhancement of the N-methyl-D-aspartate response in spinaldorsal horn neurons by cAMP-dependent protein kinase. Neurosci. Lett. 161: 124-128.

Chalmers DT, Lovenberg TW and De Souza EB (1995) Localization of novel corticotropin-releasingfactor receptor (CRF2) mRNA expression to specific subcortical nuclei in rat brain: comparison withCRF1 receptor mRNA expression. J. Neurosci. 15: 6340-6350.

Chang CP, Pearse RV 2nd, O'Connell S and Rosenfeld MG (1993) Identification of a seventransmembrane helix receptor for corticotropin-releasing factor and sauvagine in mammalian brain.Neuron 11: 1187-1195.

Chen L and Huang LY (1992) Protein kinase C reduces Mg2+ block of NMDA-receptor channels as amechanism of modulation. Nature 356: 521-523.

Chen TC, Law B, Kondratyuk T and Rossie S (1995) Identification of soluble protein phosphatases thatdephosphorylate voltage-sensitive sodium channels in rat brain. J. Biol. Chem. 270: 7750-7756.

Chen R, Lewis KA, Perrin MH and Vale WW (1993) Expression cloning of a human corticotropin-releasing-factor receptor. Proc. Natl. Acad. Sci. USA 90: 8967-8971.

Ciabarra AM, Sullivan JM, Gahn LG, Pecht G, Heinemann S and Sevarino KA (1995) Cloning andcharacterization of chi-1: a developmentally regulated member of a novel class of the ionotropicglutamate receptor family. J. Neurosci. 15: 6498-6508.

Cohen PTW (1997) Novel protein serine/threonine phosphatases: variety is the spice of life. TrendsBiochem. Sci. 22: 245-251.

Colwell CS and Levine MS (1995) Excitatory synaptic transmission in neostriatal neurons: regulationby cyclic AMP-dependent mechanisms. J. Neurosci. 15: 1704-1713.


21

Conrad CD, LeDoux JE, Magarinos AM and McEwen BS (1999) Repeated restraint stress facilitatesfear conditioning independently of causing hippocampal CA3 dendritic atrophy. Behav. Neurosci. 113:902-913.

Conti LH and Foote SL (1995) Effects of pretreatment with corticotropin-releasing factor on theelectrophysiological responsivity of the locus coeruleus to subsequent corticotropin-releasing factorchallenge. Neurosci. 69: 209-219.

Cortright DN, Nicoletti A and Seasholtz AF (1995) Molecular and biochemical characterization of themouse brain corticotropin-releasing hormone-binding protein. Mol. Cell. Endocrinol. 111: 147-157.

Da Cruz e Silva EF, Fox CA, Ouimet CC, Gustafson E, Watson SJ and Greengard PJ (1995)Differential expression of protein phosphatase 1 isoforms in mammalian brain. J. Neurosci. 15: 3375-3389.

Das S, Sasaki YF, Rothe T, Premkumar LS, Takasu M, Crandall E, Dikkes P, Conner DA, Rayudu PV,Cheung W, Chen HSV, Sucher NJ, Lipton SA and Nakanishi N (1998) Increased NMDA current andspine density in mice lacking the NMDA receptor subunit NR3A. Nature 393: 377-381.

Dash PK, Karl KA, Colicos MA, Prywes R and Kandel ER (1991) cAMP response element-bindingprotein is activated by Ca2+/calmodulin- as well as cAMP-dependent protein kinase. Proc. Natl. Acad.Sci. USA 88: 5061-5065.

Davis S, Butcher SP and Morris RG (1992) The NMDA receptor antagonist D-2-amino-5-phosphonopentanoate (D-AP5) impairs spatial learning and LTP in vivo at intracerebral concentrationscomparable to those that block LTP in vitro. J. Neurosci. 12: 21-34.

Deak T, Nguyen KT, Ehrlich AL, Watkins LR, Spencer RL, Maier SF, Licinio J, Wong ML, ChrousosGP, Webster E and Gold PW (1999) The impact of the nonpeptide corticotropin-releasing hormoneantagonist antalarmin on behavioral and endocrine responses to stress. Endocrinology 140: 79-86.

Dekker LV and Parker PJ (1994) Protein kinase C - a question of specificity. Trends Biochem. Sci. 19:73-77.

Desdouits F, Cohen D, Nairn AC, Greengard P and Girault JA (1995a) Phosphorylation of DARPP-32,a dopamine- and cAMP-regulated phosphoprotein, by casein kinase I in vitro and in vivo. J. Biol.Chem. 270: 8772-8877.

Desdouits F, Siciliano JC, Greengard P and Girault JA (1995b) Dopamine- and cAMP-regulatedphosphoprotein DARPP-32: phosphorylation of Ser-137 by casein kinase I inhibits dephosphorylationof Thr-34 by calcineurin. Proc. Natl. Acad. Sci. USA 92: 2682-2685.

Desdouits F, Siciliano JC, Nairn AC, Greengard P and Girault JA (1998) Dephosphorylation of Ser-137in DARPP-32 by protein phosphatases 2A and 2C: different roles in vitro and in striatonigral neurons.Biochem. J. 330: 211-216.

Detre JA, Nairn AC, Aswad DW and Greengard P (1984) Localization in mammalian brain of G-substrate, a specific substrate for guanosine 3',5'-cyclic monophosphate-dependent protein kinase. J.Neurosci. 4: 2843-2849.

De Quervain D, Roozendaal B and McGaugh JL (1998) Stress and glucocorticoids impair retrieval oflong-term spatial memory. Nature 394: 787-790.

Diamant M and de Wied D (1993) Structure-related effects of CRF and CRF-derived peptides:dissociation of behavioral, endocrine and autonomic activity. Neuroendocrinology 57: 1071-1081.

Chapter 1

22

Diamond DM, Fleshner M, Ingersoll N and Rose GM (1996) Psychological stress impairs spatialworking memory: relevance to electrophysiological studies of hippocampal function. Behav. Neurosci.110: 661-672.

Diamond DM, Fleshner M and Rose GM (1994) Psychological stress repeatedly blocks hippocampalprimed burst potentiation in behaving rats. Behav. Brain Res. 62: 1-9.

Diamond DM, Park CR, Heman KL and Rose GM (1999) Exposing rats to a predator impairs spatialworking memory in the radial arm water maze. Hippocampus 9: 542-552.

Diamond DM and Rose GM (1994) Stress impairs LTP and hippocampal-dependent memory. Ann. N.Y. Acad. Sci. 746: 411-414.

Durand GM, Bennett MV and Zukin RS (1993) Splice variants of the N-methyl-D-aspartate receptorNR1 identify domains involved in regulation by polyamines and protein kinase C. Proc. Natl. Acad.Sci. USA 90: 6731-6735.

Ehlers MD, Zhang S, Bernhadt JP and Huganir RL (1996) Inactivation of NMDA receptors by directinteraction of calmodulin with the NR1 subunit. Cell 84: 745-755.

Endo S, Suzuki M, Sumi M, Nairn AC, Morita R, Yamakawa K, Greengard P and Ito M (1999)Molecular identification of human G-substrate, a possible downstream component of the cGMP-dependent protein kinase cascade in cerebellar Purkinje cells. Proc. Natl. Acad. Sci. USA 96: 2467-2472.

Endo S, Zhou X, Connor J, Wang B and Shenolikar S (1996) Multiple structural elements define thespecificity of recombinant human inhibitor-1 as a protein phosphatase-1 inhibitor. Biochemistry 35:5220-5228.

Erondu NE and Kennedy MB (1985) Regional distribution of type II Ca2+/calmodulin-dependentprotein kinase in rat brain. J. Neurosci. 5: 3270-3277.

Foulkes NS, Laoide BM, Schlotter F and Sassone-Corsi P (1991) Transcriptional antagonist cAMP-responsive element modulator (CREM) down-regulates c-fos cAMP-induced expression. Proc. Natl.Acad. Sci. USA 88: 5448-5452.

Foulkes JG, Strada SJ, Henderson PJ and Cohen P (1983) A kinetic analysis of the effects of inhibitor-1and inhibitor-2 on the activity of protein phosphatase-1. Eur. J. Biochem. 132: 309-313.

Foy MR, Stanton ME, Levine S and Thompson RF (1987) Behavioral stress impairs long-termpotentiation in rodent hippocampus. Behav. Neural. Biol. 48: 138-149.

Fukunaga K, Kobayashi T, Tamura S and Miyamoto E (1993) Dephosphorylation ofautophosphorylated Ca2+/calmodulin-dependent protein kinase II by protein phosphatase 2C. J. Biol.Chem. 268: 133-137.

Garcia R, Musleh W, Tocco G, Thompson RF and Baudry M (1997) Time-dependent blockade of STPand LTP in hippocampal slices following acute stress. Neurosci. Lett. 233: 41-44.

Gardoni F, Caputi A, Cimino M, Pastorino L, Cattabeni F and Di Luca M (1998) Calcium/calmodulin-dependent protein kinase II is associated with NR2A/B subunits of NMDA receptor in postsynapticdensities. J. Neurochem. 71: 1733-1741.

Gardoni F, Schrama LH, van Dalen JJ, Gispen WH, Cattabeni F and Di Luca M (1999) AlphaCaMKIIbinding to the C-terminal tail of NMDA receptor subunit NR2A and its modulation byautophosphorylation. FEBS Lett. 456: 394-398.


23

Gonzalez GA and Montminy MR (1989) Cyclic AMP stimulates somatostatin gene transcription byphosphorylation of CREB at serine 133. Cell 59: 675-680.

Ghosh A, Ginty DD, Bading H and Greenberg ME (1994) Calcium regulation of gene expression inneuronal cells. J. Neurobiol. 25: 294-303.

Girault JA, Hemmings HC Jr, Williams KR, Nairn AC and Greengard P (1989) Phosphorylation ofDARPP-32, a dopamine- and cAMP-regulated phosphoprotein, by casein kinase II. J. Biol. Chem. 264:21748-21759.

Goda Y and Stevens CF (1996) Synaptic plasticity: the basis of particular types of learning. Curr. Biol.6: 375-378.

Grant SG and Silva AJ (1994) Targeting learning. Trends Neurosci. 17: 71-75.

Greengard P, Allen PB and Nairn AC (1999) Beyond the dopamine receptor: the DARPP-32/proteinphosphatase-1 cascade. Neuron 23: 435-447.

Grigoriadis DE, Lovenberg TW, Chalmers DT, Liaw C and De Souza EB (1996) Characterization ofcorticotropin-releasing factor receptor subtypes. Ann. N. Y. Acad. Sci. 780: 60-80.

Grosshans DR, Clayton DA, Coultrap SJ and Browning MD (2002) LTP leads to rapid surfaceexpression of NMDA but not AMPA receptors in adult rat CA1. Nat. Neurosci. 5: 27-33.

Grover LM and Teyler TJ (1990) Two components of long-term potentiation induced by differentpatterns of afferent activation. Nature 347: 477-479.

Gu Y, McIlwain KL, Weeber EJ, Yamagata T, Xu B, Antalffy BA, Reyes C, Yuva-Paylor L,Armstrong D, Zoghbi H, Sweatt JD, Paylor R and Nelson DL (2002) Impaired conditioned fear andenhanced long-term potentiation in Fmr2 knock-out mice. J. Neurosci. 22: 2753-2763.

Gustafson EL, Girault JA, Hemmings HC Jr, Nairn AC and Greengard P (1991) Immunocytochemicallocalization of phosphatase inhibitor-1 in rat brain. J. Comp. Neurol. 310: 170-188.

Hagiwara M, Alberts A, Brindle P, Meinkoth J, Feramisco J, Deng T, Karin M, Shenolikar S andMontminy M (1992) Transcriptional attenuation following cAMP induction requires PP-1-mediateddephosphorylation of CREB. Cell 70: 105-113.

Hall KU, Collins SP, Gamm DM, Massa E, DePaoli-Roach AA and Uhler MD (1999)Phosphorylation-dependent inhibition of protein phosphatase-1 by G-substrate. A Purkinje cellsubstrate of the cyclic GMP-dependent protein kinase. J. Biol. Chem. 274: 3485-3495.

Haug T and Storm JF (2000) Protein kinase A mediates the modulation of the slow Ca(2+)-dependentK(+) current, I(sAHP), by the neuropeptides CRF, VIP, and CGRP in hippocampal pyramidal neurons.J. Neurophysiol. 83: 2071-2079.

Hayashi Y, Shi SH, Esteban JA, Piccini A, Poncer JC and Malinow R (2000) Driving AMPA receptorsinto synapses by LTP and CaMKII: requirement for GluR1 and PDZ domain interaction. Science 287:2262-2267.

Hebb DO (1949) The organization of behavior. New York: John Wiley and Sons.

Heinrichs SC, Stenzel-Poore MP, Gold LH, Battenberg E, Bloom FE, Koob GF, Vale WW and PichEM (1996) Learning impairment in transgenic mice with central overexpression of corticotropin-releasing factor. Neurosci. 74: 303-311.

Chapter 1

24

Heinrichs SC, Vale EA, Lapsansky J, Behan DP, McClure LV, Ling N, De Souza EB and Schulteis G(1997) Enhancement of performance in multiple learning tasks by corticotropin-releasing factor-binding protein ligand inhibitors. Peptides 18: 711-716.

Hemmings HC Jr, Girault JA, Nairn AC, Bertuzzi G and Greengard P (1992) Distribution of proteinphosphatase inhibitor-1 in brain and peripheral tissues of various species: comparison with DARPP-32.J. Neurochem. 59: 1053-1061.

Hemmings HC Jr, Nairn AC and Greengard P (1984) DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated neuronal phosphoprotein. II. Comparison of the kinetics of phosphorylationof DARPP-32 and phosphatase inhibitor 1. J. Biol. Chem. 259: 14491-14497.

Hockberger P, Toselli M, Swandulla D and Lux HD (1989) A diacylglycerol analogue reducesneuronal calcium currents independently of protein kinase C activation. Nature 338: 340-342.

Holland PC and Bouton ME (1999) Hippocampus and context in classical conditioning. Curr. Opin.Neurobiol. 9:195-202.

Hollmann M, Boulter J, Maron C, Beasley L, Sullivan J, Pecht G and Heinemann S (1993) Zincpotentiates agonist-induced currents at certain splice variants of the NMDA receptor. Neuron 10: 943-954.

Hollmann M and Heinemann S. (1994) Cloned glutamate receptors. Annu. Rev. Neurosci. 17: 31-108.

Hollrigel GS, Chen K, Baram TZ and Soltesz I (1998) The pro-convulsant actions of corticotropin-releasing hormone in the hippocampus of infant rats. Neurosci. 84: 71-79.

Holscher C (1999) Synaptic plasticity and learning and memory: LTP and beyond. J. Neurosci. Res. 58:62-75.

Hsu SY and Hsueh AJ (2001) Human stresscopin and stresscopin-related peptide are selective ligandsfor the type 2 corticotropin-releasing hormone receptor. Nature Med. 7: 605-611.

Hu GY, Hvalby O, Walaas SI, Albert KA, Skjeflo P, Andersen P and Greengard P (1987) Proteinkinase C injection into hippocampal pyramidal cells elicits features of long term potentiation. Nature328: 426-429.

Huang CC, Tsai JJ and Gean PW. (1993) Enhancement of NMDA receptor-mediated synaptic potentialby isoproterenol is blocked by Rp-adenosine 3,5-cyclic monophosphothioate. Neurosci. Lett. 161: 207-210.

Hughes P, Lawlor P and Dragunow M (1992) Basal expression of Fos, Fos-related, Jun, and Krox 24proteins in rat hippocampus. Mol. Brain Res. 13: 355-357.

Kanaseki T, Ikeuchi Y, Sugiura H and Yamauchi T (1991) Structural features of Ca2+/calmodulin-dependent protein kinase II revealed by electron microscopy. J. Cell Biol. 115: 1049-1060.

Kasahara J, Fukunaga K and Miyamoto E (1999) Differential effects of a calcineurin inhibitor onglutamate-induced phosphorylation of Ca2+/calmodulin-dependent protein kinases in cultured rathippocampal neurons. J. Biol. Chem. 274: 9061-9067.

Kasahara J, Fukunaga K, Miyamoto E (2001) Activation of calcium/calmodulin-dependent proteinkinase IV in long term potentiation in the rat hippocampal CA1 region. J. Biol. Chem. 276: 24044-24050.

Kauer JA, Malenka RC and Nicoll RA (1988) A persistent postsynaptic modification mediates long-term potentiation in the hippocampus. Neuron 1: 911-917.


25

Kennedy MB (1989) Regulation of synaptic transmission in the central nervous system: long-termpotentiation. Cell 59: 777-787.

Kim JJ and Fanselow MS (1992) Modality-specific retrograde amnesia of fear. Science 256: 675-677.

Kim JJ, Foy MR and Thompson RF (1996) Behavioral stress modifies hippocampal plasticity throughN-methyl-D-aspartate receptor activation. Proc. Natl. Acad. Sci. USA 93: 4750-4753.

King MM, Huang CY, Chock PB, Nairn AC, Hemmings HC Jr, Chan KF and Greengard P (1984)Mammalian brain phosphoproteins as substrates for calcineurin. J. Biol. Chem. 259: 8080-8083.

Klann E, Chen SJ and Sweatt JD (1991) Persistent protein kinase activation in the maintenance phaseof long-term potentiation. J. Biol. Chem. 266: 24253-24256.

Kleckner NW and Dingledine R (1988) Requirement for glycine in activation of NMDA-receptorsexpressed in Xenopus oocytes. Science 241: 835-837.

Klee CB, Draetta GF and Hubbard MJ (1988) Calcineurin. Adv. Enzymol. Relat. Areas Mol. Biol. 61:149-200.

Klee CB, Ren H and Wang X (1998) Regulation of the calmodulin-stimulated protein phosphatase,calcineurin. J. Biol. Chem. 273: 13367-13370.

Kolaj M, Cerne R, Cheng G, Brickey DA and Randic M (1994) Alpha subunit of calcium/calmodulin-dependent protein kinase enhances excitatory amino acid and synaptic responses of rat spinal dorsalhorn neurons. J. Neurophysiol. 72: 2525-2531.

Koob GF and Bloom FE (1985) Corticotropin-releasing factor and behavior. Fed. Proc. 44: 259-263.

Kovacs KJ (1998) c-Fos as a transcription factor: a stressful (re)view from a functional map.Neurochem. Int. 33: 287-297.

Krugers HJ, Douma BR, Andringa G, Bohus B, Korf J and Luiten PG (1997) Exposure to chronicpsychosocial stress and corticosterone in the rat: effects on spatial discrimination learning andhippocampal protein kinase Cgamma immunoreactivity. Hippocampus 7: 427-436.

Kutsuwada T, Kashiwabuchi N, Mori H, Sakimura K, Kushiya E, Araki K, Meguro H, Masaki H,Kumanishi T, Arakawa M and Mishina M (1992) Molecular diversity of the NMDA receptor channel.Nature 358: 36-41.

Kwok RP, Lundblad JR, Chrivia JC, Richards JP, Bachinger HP, Brennan RG, Roberts SG, GreenMR and Goodman RH (1994) Nuclear protein CBP is a coactivator for the transcription factor CREB.Nature 370: 223-226.

Lamas M and Sassone-Corsi P (1997) The dynamics of the transcriptional response to cyclic adenosine3',5'-monophosphate: recurrent inducibility and refractory phase. Mol. Endocrinol. 11: 1415-1424.

Lee EH, Lee CP, Wang HI and Lin WR (1993) Hippocampal CRF, NE, and NMDA systeminteractions in memory processing in the rat. Synapse 14: 144-153.

Leonard AS and Hell JW (1997) Cyclic AMP-dependent protein kinase and protein kinase Cphosphorylate N-methyl-D-aspartate receptors at different sites. J. Biol. Chem. 272: 12107-12115.

Lewis K, Li C, Perrin MH, Blount A, Kunitake K, Donaldson C, Vaughan J, Reyes TM, Gulyas J,Fischer W, Bilezikjian L, Rivier J, Sawchenko PE and Vale WW (2001) Identification of urocortin III,an additional member of the corticotropin-releasing factor (CRF) family with high affinity for theCRF2 receptor. Proc. Natl. Acad. Sci. USA 98: 7570-7575.

Chapter 1

26

Liang KC and Lee EH (1988) Intra-amygdala injections of corticotropin releasing factor facilitateinhibitory avoidance learning and reduce exploratory behavior in rats. Psychopharmacology (Berl.) 96:232-236.

Lieberman DN and Mody I (1994) Regulation of NMDA channel function by endogenous Ca(2+)-dependent phosphatase. Nature 369: 235-239.

Lin JW, Wyszynski M, Madhavan R, Sealock R, Kim JU and Sheng M (1998) Yotiao, a novel proteinof neuromuscular junction and brain that interacts with specific splice variants of NMDA receptorsubunit NR1. J. Neurosci. 18: 2017-2027.

Linden DJ and Routtenberg A (1989) The role of protein kinase C in long-term potentiation: a testablemodel. Brain Res. Rev. 14: 279-296.

Lovenberg TW, Liaw CW, Grigoriadis DE, Clevenger W, Chalmers DT, De Souza EB and OltersdorfT (1995a) Cloning and characterization of a functionally distinct corticotropin-releasing factor receptorsubtype from rat brain. Proc. Natl. Acad. Sci. USA 92: 836-840.

Lovenberg TW, Chalmers DT, Liu C and De Souza EB (1995b) CRF2 alpha and CRF2 beta receptormRNAs are differentially distributed between the rat central nervous system and peripheral tissues.Endocrinol. 136: 4139-4142.

Lu W, Man H, Ju W, Trimble WS, MacDonald JF, and Wang YT (2001) Activation of synapticNMDA receptors induces membrane insertion of new AMPA receptors and LTP in culturedhippocampal neurons. Neuron 29: 243-254.

Luine V, Villegas M, Martinez C and McEwen BS (1994) Repeated stress causes reversibleimpairments of spatial memory performance. Brain Res. 639: 167-170.

Luscher C, Xia H, Beattie EC, Carroll RC, von Zastrow M, Malenka RC and Nicoll RA (1999) Role ofAMPA receptor cycling in synaptic transmission and plasticity. Neuron 24: 649-658.

MacDougall LK, Campbell DG, Hubbard MJ and Cohen P (1989) Partial structure and hormonalregulation of rabbit liver inhibitor-1; distribution of inhibitor-1 and inhibitor-2 in rabbit and rat tissues.Biochim. Biophys. Acta. 1010: 218-226.

Maier SF (1990) Role of fear in mediating shuttle escape learning deficit produced by inescapableshock. J. Expt. Psychol. Anim. Behav. Processes 16: 137-149.

Makhinson M, Chotiner JK, Watson JB and O'Dell TJ (1999) Adenylyl cyclase activation modulatesactivity-dependent changes in synaptic strength and Ca2+/calmodulin-dependent kinase IIautophosphorylation. J. Neurosci. 19: 2500-2510.

Malenka RC and Nicoll RA (1993) NMDA-receptor-dependent synaptic plasticity: multiple forms andmechanisms. Trends Neurosci. 16: 521-527.

Markram H and Segal M (1992) Activation of protein kinase C suppresses responses to NMDA in ratCA1 hippocampal neurones. J. Physiol. 457: 491-501.

Martin SJ, Grimwood PD and Morris RG (2000) Synaptic plasticity and memory: an evaluation of thehypothesis. Annu. Rev. Neurosci. 23: 649-711.

Mellstrom B, Naranjo JR, Foulkes NS, Lafarga M and Sassone-Corsi P (1993) Transcriptionalresponse to cAMP in brain: specific distribution and induction of CREM antagonists. Neuron 10: 655-665.


27

Mesches MH, Fleshner M, Heman KL, Rose GM and Diamond DM (1999) Exposing rats to a predatorblocks primed burst potentiation in the hippocampus in vitro. J. Neurosci. 19: RC18: 1-5.

Meyer T, Hanson PI, Stryer L and Schulman H (1992). Calmodulin trapping by calcium-calmodulin-dependentprotein kinase. Science 256: 1199-1202.

Millward TA, Zolnierowicz S and Hemmings BA (1999) Regulation of protein kinase cascades C byprotein phosphatase 2A. Trends Biochem. Sci. 24: 186-191.

Montminy MR and Bilezikjian LM (1987)Binding of a nuclear protein to the cyclic-AMP responseelement of the somatostatin gene. Nature 328: 175-178.

Monyer H, Burnashev N, Laurie DJ, Sakmann B and Seeburg PH (1994) Developmental and regionalexpression in the rat brain and functional properties of four NMDA receptors. Neuron 12: 529-540.

Monyer H, Sprengel R, Schoepfer R, Herb A, Higuchi M, Lomeli H, Burnashev N, Sakmann B andSeeburg PH (1992) Heteromeric NMDA receptors: molecular and functional distinction of subtypes.Science 256: 1217-1221.

Morgan JI and Curran T (1991) Stimulus-transcription coupling in the nervous system: involvement ofthe inducible proto-oncogenes fos and jun. Annu. Rev. Neurosci. 14: 4421-451.

Morris RGM, Anderson E, Lynch GS and Baudry M (1986) Selective impairment of learing andblockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist, AP. Nature 319:774-776.

Moser EI, Moser MB and Andersen P (1994) Potentiation of dentate synapses initiated by exploratorylearning in rats: dissociation from brain temperature, motor activity, and arousal. Learn. Mem. 1: 55-73.

Mulkey RM, Herron CE and Malenka RC (1993) An essential role for protein phosphatases inhippocampal long-term depression. Science 261: 1051-1055.

Mulkey RM and Malenka RC (1992) Mechanisms underlying induction of homosynaptic long-termdepression in area CA1 of the hippocampus. Neuron 9: 967-975.

Nicoll RA, Kauer JA and Malenka RC (1988) The current excitement in long-term potentiation.Neuron 1: 97-103.

Nicoll RA and Malenka RC (1995) Contrasting properties of two forms of long-term potentiation in thehippocampus. Nature 377: 115-118.

Nimmo GA and Cohen P (1978) The regulation of glycogen metabolism. Purification andcharacterisation of protein phosphatase inhibitor-1 from rabbit skeletal muscle. Eur. J. Biochem. 87:341-351.

Nishi A, Bibb JA, Snyder GL, Higashi H, Nairn AC and Greengard P (2000) Amplification ofdopaminergic signaling by a positive feedback loop. Proc. Natl. Acad. Sci. USA 97: 12840-12845.

Nishi A, Snyder GL and Greengard P (1997) Bidirectional regulation of DARPP-32 phosphorylationby dopamine. J. Neurosci. 17: 8147-8155.

Nishi A, Snyder GL, Nairn AC and Greengard P (1999) Role of calcineurin and protein phosphatase-2A in the regulation of DARPP-32 dephosphorylation in neostriatal neurons. J. Neurochem. 72: 2015-2021.

Nishizuka Y (1988) The molecular heterogeneity of protein kinase C and its implications for cellularregulation. Nature 334: 661-665.

Chapter 1

28

Nowak L, Bregestovski P, Ascher P, Herbet A and Prochiantz A (1984) Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307: 462-465.

Omkumar RV, Kiely MJ, Rosenstein AJ, Min KT and Kennedy MB (1996) Identification of aphosphorylation site for calcium/calmodulindependent protein kinase II in the NR2B subunit of the N-methyl-D-aspartate receptor. J. Biol. Chem. 271: 31670-31678.

Otani S and Connor JA (1998) Requirement of rapid Ca2+ entry and synaptic activation ofmetabotropic glutamate receptors for the induction of long-term depression in adult rat hippocampus. J.Physiol. 511: 761-770.

Ouimet CC, da Cruz e Silva EF and Greengard P (1995) The alpha and gamma 1 isoforms of proteinphosphatase 1 are highly and specifically concentrated in dendritic spines. Proc. Natl. Acad. Sci. USA92: 3396-3400.

Ouimet CC, Miller PE, Hemmings HC Jr, Walaas SI and Greengard P (1984) DARPP-32, a dopamine-and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brainregions. III. Immunocytochemical localization. J. Neurosci. 4: 111-124.

Owens MJ and Nemeroff CB (1991) Physiology and pharmacology of corticotropin-releasing factor.Pharmacol. Rev. 43: 425-473.

Ozawa S, Kamiya H and Tsuzuki K (1998) Glutamate receptors in the mammalian central nervoussystem. Prog. Neurobiol. 54: 581-618.

Perez RG and Lewis RM (1992) Regional distribution of DARPP-32 (dopamine- and adenosine 3',5'-monophosphate-regulated phosphoprotein of Mr = 32,000) mRNA in mouse brain. J. Comp. Neurol.318: 304-315.

Pérez-Otaño I, Schulteis CT, Contractor A, Lipton SA, Trimmer JS, Sucher NJ and Heinemann SF(2001) Assembly with the NR1 Subunit Is Required for Surface Expression of NR3A-ContainingNMDA Receptors. J. Neurosci. 21: 1228-1237.

Perkel DJ, Petrozzino JJ, Nicoll RA and Connor JA (1993) The role of Ca2+ entry via synapticallyactivated NMDA receptors in the induction of long-term potentiation. Neuron 11: 817-823.

Phillips RG and LeDoux JE (1992) Differential contribution of amygdala and hippocampus to cued andcontextual fear conditioning. Behav. Neurosci. 106: 274-285.

Potter E, Behan DP, Fischer WH, Linton EA, Lowry PJ and Vale WW (1991) Cloning andcharacterization of the cDNAs for human and rat corticotropin releasing factor-binding proteins. Nature349: 423-426.

Potter E, Behan DP, Linton EA, Lowry PJ, Sawchenko PE and Vale WW (1992) The centraldistribution of a corticotropin-releasing factor (CRF)-binding protein predicts multiple sites and modesof interaction with CRF. Proc. Natl. Acad. Sci. USA 89: 4192-4196.

Radulovic J, Ruhmann A, Liepold T and Spiess J (1999) Modulation of learning and anxiety bycorticotropin-releasing factor (CRF) and stress: differential roles of CRF receptors 1 and 2. J. Neurosci.19: 5016-5025.

Raman IM, Tong G and Jahr CE (1996) Beta-adrenergic regulation of synaptic NMDA receptors bycAMP-dependent protein kinase. Neuron 16: 415-421.

Ramon y Cayal S (1911) Histologie du système nervuex de l’homme et des vertébrés. Paris: A.Maloine.


29

Reyes TM, Lewis K, Perrin MH, Kunitake KS, Vaughan J, Arias CA, Hogenesch JB, Gulyas J, RivierJ, Vale WW and Sawchenko PE (2001) Urocortin II: a member of the corticotropin-releasing factor(CRF) neuropeptide family that is selectively bound by type 2 CRF receptors. Proc. Natl. Acad. Sci.USA 98: 2843-2848.

Robertson LM, Kerppola TK, Vendrell M, Luk D, Smeyne RJ, Bocchiaro C, Morgan JI and Curran T(1995) Regulation of c-fos expression in transgenic mice requires multiple interdependent transcriptioncontrol elements. Neuron 14: 241-252.

Roche KW, Tingley WG and Huganir RL (1994) Glutamate receptor phosphorylation and synapticplasticity. Curr. Opin. Neurobiol. 4: 383-388.

Sacktor TC, Osten P, Valsamis H, Jiang X, Naik MU and Sublette E (1993) Persistent activation of thezeta isoform of protein kinase C in the maintenance of long-termpotentiation. Proc. Natl. Acad. Sci.USA 90: 8342-8346.

Sagar SM, Sharp FR and Curran T (1988) Expression of c-fos protein in brain: metabolic mapping atthe cellular level. Science 240: 1328-1331.

Servatius RJ and Shors TJ (1994) Exposure to inescapable stress persistently facilitates associative andnonassociative learning in rats. Behav. Neurosci. 108: 1101-1106.

Sharp PE, Mc Naughton BL and Barnes CA (1985) Enhancement of hippocampal field potentials inrats exposed to a novel complex enviroment. Brain Res. 339: 361-365.

Sheng M and Greenberg ME (1990) The regulation and function of c-fos and other immediate earlygenes in the nervous system. Neuron 4: 477-485.

Sheng M, Thompson MA and Greenberg ME (1991) CREB: a Ca(2+)-regulated transcription factorphosphorylated by calmodulin-dependent kinases. Science 252: 1427-1430.

Shenolikar S (1994) Protein serine/threonine phosphatases--new avenues for cell regulation. Annu.Rev. Cell. Biol. 10: 55-86.

Shi SH, Hayashi Y, Petralia RS, Zaman SH, Wenthold RJ, Svoboda K and Malinow R (1999) Rapidspine delivery and redistribution of AMPA receptors after synaptic NMDA receptor activation. Science284: 1811-1816.

Shors TJ and Dryver E (1994) Effects of stress and long-term potentiation (LTP) on subsequent LTPand the theta burst response in the dentate gyrus. Brain Res. 666: 232-238.

Shors TJ, Foy MR, Levine S and Thompson RF (1990a) Unpredictable and uncontrollable stressimpairs neuronal plasticity in the rat hippocampus. Brain Res. Bull. 24: 663-667.

Shors TJ, Levine S and Thompson RF (1990b) Effect of adrenalectomy and demedullation on thestress-induced impairment of long-term potentiation. Neuroendocrinology 51: 70-75.

Shors TJ and Mathew PR (1998) NMDA receptor antagonism in the lateral/basolateral but not centralnucleus of the amygdala prevents the induction of facilitated learning in response to stress. Learn.Mem. 5: 220-230.

Shors TJ and Matzel LD(1997) Long-term potentiation: what's learning got to do with it? Behav. BrainSci. 20: 597-614

Shors TJ, Seib TB, Levine S and Thompson RF (1989) Inescapable versus escapable shock modulateslong-term potentiation in the rat hippocampus. Science 244: 224-226.

Chapter 1

30

Shors TJ and Servatius RJ (1995) Stress-induced sensitization and facilitation of learning are dependenton NMDA-receptor activation. Neuro Report 6: 677-680.

Shors TJ, Weiss C and Thompson RF (1992) Stress-induced facilitation of classical conditioning.Science 257: 537-539.

Siggins GR, Gruol D, Aldenhoff J and Pittman Q (1985) Electrophysiological actions of corticotropin-releasing factor in the central nervous system. Fed. Proc. 44: 237-242.

Skelton RW, Miller JJ and Phillips AG (1985) Long-term potentiation facilitates behavioral respondingto single-pulse stimulation of the perforant path. Behav. Neurosci. 99: 603-620.

Smith BN and Dudek FE (1994) Age-related epileptogenic effects of corticotropin-releasing hormonein the isolated CA1 region of rat hippocampal slices. J. Neurophysiol. 72: 2328-2333.

Snyder GL, Fienberg AA, Huganir RL and Greengard P (1998) A dopamine/D1 receptor/protein kinaseA/dopamine- and cAMP-regulated phosphoprotein (Mr 32 kDa)/protein phosphatase-1 pathwayregulates dephosphorylation of the NMDA receptor. J. Neurosci. 18: 10297-10303.

Soderling TR (2000) CaM-kinases: modulators of synaptic plasticity. Curr. Opin. Neurobiol. 10: 375-380.

Spiess J, Rivier J, Rivier C and Vale W (1981) Primary structure of corticotropin-releasing factor fromovine hypothalamus. Proc. Natl. Acad. Sci. USA 78: 6517-6521.

Stern P, Behe P, Schoepfer R and Colquhoun D (1992) Single-channel conductances of NMDAreceptors expressed from cloned cDNAs: comparison with native receptors. Proc. R. Soc. Lond. B.Biol. Sci. 250: 271-277.

Strack S, Barban MA, Wadzinski BE and Colbran RJ (1997) Differential inactivation of postsynapticdensity-associated and soluble Ca2+/calmodulin-dependent protein kinase II by protein phosphatases 1and 2A. J. Neurochem. 68: 2119-2128.

Strack S and Colbran RJ (1998) Autophosphorylation-dependent targeting of calcium/ calmodulin-dependent protein kinase II by the NR2B subunit of the N-methyl- D-aspartate receptor. J. Biol. Chem.273: 20689-20692.

Sucher NJ, Akbarian S, Chi CL, Leclerc CL, Awobuluyi M, Deitcher DL, Wu MK, Yuan JP, Jones EGand Lipton SA (1995) Development and regional expression pattern of a novel NMDA receptor-likesubunit (NMDAR-L) in the rodent brain. J. Neurosci. 15: 6509-6520.

Sun P, Enslen H, Myung PS and Maurer RA (1994) Differential activation of CREB byCa2+/calmodulin-dependent protein kinases type II and type IV involves phosphorylation of a site thatnegatively regulates activity. Genes Dev. 8: 2527-2539.

Tanaka C and Saito N (1992) Localization of subspecies of protein kinase C in the mammalian centralnervous system. Neurochem. Int. 21: 499-512.

Tanzi E (1893) I fatti e le induzioni nell’odierna istologia del sistema nervoso. Riv. sperim. freniatriamedic. leg. 19: 419-472.

Taylor SS, Buechler JA and Yonemoto W (1990) cAMP-dependent protein kinase: framework for adiverse family of regulatory enzymes. Annu. Rev. Biochem. 59:971-1005.

Tobimatsu T and Fujisawa H (1989) Tissue-specific expression of four types of rat calmodulin-dependent protein kinase II mRNAs. J. Biol. Chem. 264: 17907-17912.


31

Tyler WJ and Pozzo-Miller LD (2001) BDNF enhances quantal neurotransmitter release and increasesthe number of docked vesicles at the active zones of hippocampal excitatory synapses. J. Neurosci. 21:4249-4258.

Urushihara H, Tohda M and Nomura Y (1992) Selective potentiation of N-methyl-D-aspartate-inducedcurrent by protein kinase C in Xenopus oocytes injected with rat brain RNA. J. Biol. Chem. 267:11697-11700.

Vale W, Spiess J, Rivier C and Rivier J (1981) Characterization of a 41-residue ovine hypothalamicpeptide that stimulates secretion of corticotropin and beta-endorphin. Science 213: 1394-1397.

Van Pett K, Viau V, Bittencourt JC, Chan RK, Li HY, Arias C, Prins GS, Perrin M, Vale W andSawchenko PE (2000) Distribution of mRNAs encoding CRF receptors in brain and pituitary of rat andmouse. J. Comp. Neurol. 428: 191-212.

Vaughan J, Donaldson C, Bittencourt J, Perrin MH, Lewis K, Sutton S, Chan R, Turnbull AV, LovejoyD, Rivier C, Rivier J, Sawchenko PE and Vale WW (1995) Urocortin, a mammalian neuropeptiderelated to fish urotensin I and to corticotropin-releasing factor. Nature 378: 287-292.

Vijayaraghavan S, Stephens DT, Trautman K, Smith GD, Khatra B, da Cruz e Silva EF and GreengardP (1996) Sperm motility development in the epididymis is associated with decreased glycogen synthasekinase-3 and protein phosphatase 1 activity. Biol. Reprod. 54: 709-718.

Von Frijtag JC, Kamal A, Reijmers LG, Schrama LH, van den Bos R and Spruijt BM (2001) Chronicimipramine treatment partially reverses the long-term changes of hippocampal synaptic plasticity insocially stressed rats. Neurosci. Lett. 309: 153-156.

Wadzinski BE, Wheat WH, Jaspers S, Peruski LF Jr, Lickteig RL, Johnson GL and Klemm DJ (1993)Nuclear protein phosphatase 2A dephosphorylates protein kinase A-phosphorylated CREB andregulates CREB transcriptional stimulation. Mol. Cell. Biol. 13: 2822-2834.

Walaas SI and Greengard P (1984) DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. I. Regional and cellulardistribution in the rat brain. J. Neurosci. 4: 84-98.

Wang LY, Orser BA, Brautigan DL and MacDonald JF (1994) Regulation of NMDA receptors incultured hippocampal neurons by protein phosphatases 1 and 2A. Nature 369: 230-232.

Wang HL, Tsai LY and Lee EH (2000) Corticotropin-releasing factor produces a protein synthesis--dependent long-lastingpotentiation in dentate gyrus neurons. J. Neurophysiol. 83: 343-349.

Wang HL, Wayner MJ, Chai CY and Lee EHY (1998) Corticotrophin-releasing factor produces a long-lasting enhancement of synaptic efficacy in the hippocampus. Eur. J. Neurosci. 10: 3428-3437.

Wilson LM, Wilson JR and Dicara LV (1975) Facilitation of Pavlovian conditioned cardiodecelerationsfollowing preshock in immobilized rats. Physiol. Behav. 15: 653-658.

Wu HC, Chen KY, Lee WY and Lee EH (1997) Antisense oligonucleotides to corticotropin-releasingfactor impair memory retention and increase exploration in rats. Neurosci. 78: 147-153.

Xing J, Ginty DD and Greenberg ME (1996) Coupling of the RAS-MAPK pathway to gene activationby RSK2, a growth factor-regulated CREB kinase. Science 273: 959-963.

Xu L, Anwyl R and Rowan MJ (1997) Behavioural stress facilitates the induction of long-termdepression in the hippocampus. Nature 387: 497-500.

Chapter 1

32

Yakel JL (1997) Calcineurin regulation of synaptic function: from ion channels to transmitter releaseand gene transcription. Trends Pharmacol. Sci. 18: 124-134.

Yamamoto KK, Gonzalez GA, Biggs WH and Montminy MR (1988) Phosphorylation-induced bindingand transcriptional efficacy of nuclear factor CREB. Nature 334: 494-498.

Yamashita H, Kasai M and Inenaga K (1991) Effects of corticotropin-releasing factor on neurons in thehypothalamic paraventricular nucleus in vitro. Brain Res. Bull. 27: 321-325.

Yan Z, Hsieh-Wilson L, Feng J, Tomizawa K, Allen PB, Fienberg AA, Nairn AC and Greengard P(1999) Protein phosphatase 1 modulation of neostriatal AMPA channels: regulation by DARPP-32 andspinophilin. Nat. Neurosci. 2: 13-17.

Yang SN (2000) Sustained enhancement of AMPA receptor- and NMDA receptor-mediated currentsinduced by dopamine D1/D5 receptor activation in the hippocampus: an essential role of postsynapticCa2+. Hippocampus 10: 57-63.

Zakharenko SS, Zablow L and Siegelbaum SA (2001 Visualization of changes in presynaptic functionduring long-term synaptic plasticity. Nat. Neurosci. 4: 711-717.

Zola-Morgan S and Squire LR (1993) Neuroanatomy of memory. Annu. Rev. Neurosci. 16: 547-563.

CHAPTER 2

Modulation of hypothalamic NMDA receptor function

by cyclic AMP-dependent protein kinase and

phosphatases

Journal of Neurochemistry (2000) 75: 749-754

ABSTRACT

In the present study we investigated the modulation of hypothalamic NMDAreceptor-mediated currents by cAMP-dependent protein kinase (PKA) using the twoelectrode voltage clamp technique in Xenopus oocytes injected with rathypothalamic mRNA. Application of forskolin, which activates PKA by means ofcAMP stimulation, caused a transient increase of NMDA-induced currents, whereasthe inactive forskolin analogue 1,9-dideoxyforskolin had no effect. Incubation ofoocytes with a membrane-permeable analogue of cyclic AMP, 8-bromoadenosine3’,5’-cyclic monophosphate, potentiated NMDA responses even more prominentlythan with forskolin. NMDA-induced currents recorded from Xenopus oocytesinjected with cRNA encoding the NMDA receptor subunits NR1, NR2A, and/orNR2B, mainly found in rat hypothalamus, were not affected by PKA activation butwere increased by PKC stimulation. It is interesting that inhibition of endogenousprotein phosphatase 1 and/or 2A by calyculin A resulted in a similar enhancement ofhypothalamic NMDA-induced currents. Preinjection of oocytes with calyculin Aimpeded the PKA- but not the PKC-mediated potentiation of hypothalamic NMDA-induced currents. We propose the involvement of an additional third messenger inthe PKA effect, which acts most likely via the inhibition of tonically active proteinphosphatase 1 and/or 2A.

Chapter 2

34

INTRODUCTION

The hypothalamus acts as a major regulatory center for autonomic andendocrine homeostasis. Structurally, it consists of a group of nuclei that regulate abroad array of physiological and behavioral activities, including feeding andmetabolism but also emotional behavior and reproduction (Brann and Mahesh,1994).

Glutamate, the main excitatory neurotransmitter in the central nervoussystem, is found in large concentrations in various important hypothalamic nuclei(Van den Pol et al., 1990; Van den Pol, 1991; Goldsmith et al., 1994) and is able toexcite virtually all neurons tested in the hypothalamus, including the arcuatenucleus, the median eminence, and other medial hypothalamic regions (Arnauld etal., 1983; Van den Pol et al., 1990). Studies examining the effect of glutamateantagonists on spontaneous excitatory postsynaptic potentials in hypothalamic nucleihave clearly demonstrated that glutamatergic neurotransmission is an importantcomponent in the regulation of neuroendocrine function in the hypothalamus(Gribkoff and Dudek, 1990; Van den Pol et al., 1990; Waurin and Dudek, 1991;Inenaga et al., 1998).

The majority of studies investigating the NMDA receptor, an importantionotropic glutamate receptor subtype distributed throughout the hypothalamus (Vanden Pol et al., 1994; Kus et al., 1995), have focused on its role in inducing release ofhormones such as growth hormone and gonadotropin-releasing hormone, but it hasalso been demonstrated to play a pivotal role in the induction and regulation of thecircadian rhythm (reviewed in: Brann and Mahesh, 1994).

Although the importance of the NMDA receptor in regulating thehypothalamic neuroendocrine system is now widely acknowledged, little is knownabout the modulation of hypothalamic NMDA receptor function itself. In general, acommon mechanism for the modulation of the NMDA receptor-channel complex isphosphorylation by second messenger-dependent protein kinases.

In the present study, we investigated the modulation of hypothalamicNMDA receptor function by cyclic AMP (cAMP)-dependent protein kinase (PKA)and its possible underlying mechanism by using the two electrode voltage clamptechnique in Xenopus oocytes.

MATERIALS AND METHODS

MaterialsMature female specimens of Xenopus laevis were obtained from Kähler (Hamburg,

Germany). Collagenase (type II) was from Worthington Biochemicals (Freehold, NJ, U.S.A.). ThemRNA Purification Kit was from Pharmacia LKB Biotechnology (Uppsala, Sweden). Calyculin A (CalA) was from Boehringer (Mannheim, Germany). 1,9-Dideoxyforskolin, 8-bromoadenosine 3`5`-cyclicmonophosphate (8-Br-cAMP; sodium salt) and H-89 dihydrochloride were from Calbiochem (La Jolla,CA, U.S.A.). Phorbol 12-myristate 13-acetate (PMA) and 4α-PMA were from Research Biochemicals(Natick, MA, U.S.A.). Penicillin and streptomycin were from Gibco (Paisley, U.K.). All other drugsand salts were purchased from Sigma (St. Louis, MO, U.S.A.).

PKA modulation of hypothalamic NMDA responses

35

We confirm that all animal protocols used have been approved by the animal experimentationcommittee of the State Government of Lower Saxony.

Handling,Injection, and Maintenance of OocytesOocytes were surgically obtained from adult X. laevis frogs that were anesthetized by

immersion into 3-aminobenzoic acid ethyl ester (1 g/L) for 45 min. Follicular cell layers were removedmanually after incubation in Ca2+-free modified Barth's solution [88 mM NaCl, 1 mM KCl, 2.4 mMNaHCO3, 15 mM HEPES, 0.8 mM MgSO4, 50 U/ml penicillin and 50 µg/ml streptomycin (adjusted topH 7.6 with NaOH)] containing 3.0 mg/ml collagenase (type II) for 2 h at room temperature (20-24°C).Stage V and stage VI oocytes were pressure-injected with 50 ng hypothalamic poly (A)+ mRNA or 10-15 ng NR1b cRNA within 24 h after harvesting. In some experiments the NR1b cRNA was coinjectedwith NR2A and/or NR2B cRNA (25-30 ng) in a 1:2 ratio. After injection, the oocytes were maintainedat 18°C in modified Barth's solution. Modified Barth's solution contained 88 mM NaCl, 1 mM KCl, 2.4mM NaH CO3, 15 mM HEPES, 0.8 mM Mg SO4, 0.3 mM Ca(NO3)2, 0.6 mm CaCl2, 50 U/ml penicillinand 50 µg/ml streptomycin (adjusted to pH 7.6 with NaOH).

RNA PreparationTotal RNA was isolated from the hypothalamus of adult Wistar rats by extraction of fresh

tissue with guanidine thiocyanate and precipitation with LiCl (Cathala et al., 1983). Poly(A)+ mRNAwas purified by oligo(dT) cellulose chromatography (Pharmacia mRNA Purification Kit) and dissolvedin RNase-free water at a concentration of 0.5 µg/µl.

Plasmids containing the cDNA clones encoding the NR1b, NR2A, and NR2B subunits werelinearized by digestion with PvuI (NR1b) or NotI (NR2A and NR2B) and used as transcriptiontemplates. In vitro transcription was carried out by T7 or T3 RNA polymerase in the presence of thecapping reagent m7G(5')ppp(5')G. RNA transcripts were precipitated by ethanol, and the precipitatewas dissolved in diethyl pyrocarbonate-treated water.

Electrophysiological RecordingsAt 7-8 days after injection oocytes were placed in a recording chamber (volume ≈ 20 µl) and

continuously superfused (1.0 ml/min) with recording solution at room temperature. The recordingsolution contained (115 mM NaCl, 2.5 mM KCl, 1.8 mM BaCl2 and 10 mM HEPES (pH adjusted to7.2 with NaOH). The oocytes were voltage-clamped by a conventional two-microelectrode voltage-clamp technique (Stühmer, 1992). The membrane potential of the oocytes was held at -80 mV using aTurbo Clamp Tec 01C amplifier (N.P.I. Electronic, Tamm, Germany). The two microelectrodes werefilled with 3 M KCl and had resistances between 1-3 MΩ. NMDA currents were elicited by switchingthe flow from recording solution to recording solution containing NMDA (100 µM) and glycine (10µM) without changing the perfusion rate.

In some experiments oocytes were preincubated with 8-Br-cAMP (10 mM) or Cal A (100nM). During this incubation oocytes were continuously superfused with recording solution containingthe drug to be applied. In some experiments, oocytes were incubated for 3 h with H-89, which wasdiluted to a final concentration of 10 µM in modified Barth's solution. The H-89 stock solution (10mM) was prepared in dimethyl sulfoxide (DMSO). Forskolin (50 µM), 1,9-dideoxyforskolin (50 µM),PMA (10 nM), and 4α-PMA (10 nM) were steadily applied for 1 min with a micropipette (200 µl) afterstopping the solution flow. Cal A was prepared in 100% ethanol as 1 mM stock solution, which wasdiluted with recording solution to the appropriate concentration shortly before the electrophysiologicalexperiments. Cal A was preinjected 50 - 70 min before voltage-clamping following the RNA injectionprotocol. Aliquots of concentrated stock solutions of forskolin, 1,9-dideoxyforskolin, PMA and 4α-PMA in DMSO and stock solutions of NMDA and glycine in distilled water were added to therecording solution immediately before the experiments. The final DMSO concentration never exceeded0.1% (vol/vol), which, upon application, induced no change in membrane current by itself and whichhad no effect on NMDA-induced currents in control experiments. 8-Br-cAMP was directly dissolved tothe final concentration in the recording solution. Effects of the various compounds were tested after atleast two or three identical responses to NMDA could be elicited (control responses). Oocytes were notused for experiments if reproducible responses to NMDA could not be recorded within 30 min.

Current signals were low-pass-filtered at 30 Hz using a four-pole Bessel filter and digitizedby an ITC 16-MAC interface (Instrutech Corp., Great Neck, NY). Data were sampled at 100 Hz andstored on a Macintosh (7100/66) computer using data acquisition software (Pulse 7.40; HEKA

Chapter 2

36

Electronic, Lambrecht/Pfalz, Germany). To compare responses between different oocytes, theamplitude of the NMDA-induced current at any given time was normalized with respect to theamplitude of control responses. The n value represents the number of oocytes in one experiment. Atleast two different batches of oocytes were used per experiment.

StatisticsStatistical comparisons were made using Student's t test. Results were expressed as mean ±

SEM values. Values of p ≥ 0.05 were regarded as not significant.

RESULTS

Stimulation of cAMP potentiates hypothalamic NMDA responsesThe effect of cAMP stimulation on NMDA-induced responses was

examined in oocytes injected with hypothalamic mRNA isolated from adult ratbrain. Application of NMDA together with glycine at a holding potential of -80 mVelicited responses with the typical characteristics previously described for NMDA-induced responses (Leonard and Kelso, 1990; Kelso et al., 1992). The averageresponse amplitude was 25 ± 6 nA (n = 34).

Figure 1. Enhancement of hypothalamic NMDA-induced responses by cAMPstimulation. A. Representative NMDA responses of hypothalamic mRNA-injectedoocytes before (control) and after 1-min application of 50 µM forskolin or 10-minapplication of 10 mM 8-Br-cAMP. B and C. Time course of NMDA-induced currentamplitudes after application of 50 µM forskolin or 50 µM 1,9-dideoxyforskolin (B) or(C) 10 mM 8-Br-cAMP. Horizontal bars intervals of drug superfusion. Externalsolution was applied in control experiments. Data are the mean ± SEM (bars) valuesof three to five oocytes.

Control8-Br-cAMP (10 mM)

0.5

1

1.5

2

2.5

-5 0 5 10 15 20 25 30 35Time (min)

ControlForskolin (50 µM)1,9-Dideoxy. (50 µM)

0.5

1

1.5

2

-5 0 5 10 15 20 25Time (min)

Rel

ativ

e cu

rren

t am

plitu

de

B C

A

2 s20 nA

NMDA

Control

8-Br-cAMP (10 mM)

2 s20 nA

Control

Forskolin (50 µM)

NMDA -80 mV


37

When oocytes were incubated with forskolin (50 µM), which activates PKAby increasing cAMP levels (reviewed by Seamon and Daly, 1986), an enhancementof the NMDA response amplitude was observed with a maximum reached after 7min (150 ± 9%; n = 5; Fig. 1A and B). This forskolin concentration appeared to bethe minimal concentration required to exhibit maximal forskolin effects on NMDA-induced currents (data not shown). Because it is known that forskolin can sometimesact directly on ion channels without the involvement of cAMP (Laurenza et al.,1989), oocytes were incubated with 1,9-dideoxyforskolin (50 µM), an inactiveanalogue of forskolin that does not activate adenylate cyclase. 1,9-Dideoxyforskolindid not affect NMDA responses (100 ± 0%; n = 3; Fig. 1B), indicating that forskolincaused the potentiation of NMDA-induced responses via elevated cAMP levels.Bath application of 8-Br-cAMP (10 mM, 10 min), a membrane-permeable analogueof cAMP, also caused a large and transient increase of hypothalamic NMDAresponses, which was more prominent than the forskolin effect (225 ± 19%; n = 3;Fig. 1A and C). Repeated application of NMDA and glycine with 2-min intervalsshowed no change in NMDA response amplitude for the time course of theexperiments (indicated as control; Fig. 1).

Stimulation of cAMP does not affect responses of recombinant NMDAreceptors

To test whether PKA potentiates NMDA-induced currents via a direct actionon the NMDA receptor itself, we investigated the effect of forskolin (50 µM)incubation on NMDA-induced currents recorded from oocytes expressingrecombinant NMDA receptors.

Table 1. Effect of protein kinase modulators on recombinant NMDA receptors

NR subunits*

Compound 1b 1b/2A 1b/2B 1b/2A2B

Control 1.02 ± 0.03 (3) 0.98 ± 0.07 (3) 0.87 ± 0.09 (3) 1.04 ± 0.08 (3)

Forskolin (50µM) 1.11 ± 0.05 (5) 1.01 ± 0.07 (3) 0.95 ± 0.02 (3) 1.01 ± 0.06 (3)

PMA (10 nM)‡ 2.08 ± 0.28 (6) 2.13 ± 0.38 (3) 1.68 ± 0.16 (5) 1.69 ± 0.30 (3)

4-α-PMA (10 nM)‡ 0.99 ± 0.03 (3) 0.85 ± 0.19 (3) 0.89 ± 0.11 (3) 0.87 ± 0.08 (3)

* Data are mean ± SEM values of the relative current amplitude as determined by theI/Ic ratio of steady-state responses to 100 µM NMDA and 10 µM glycine after (I) andbefore (Ic) 1-min application of recording solution (control), 50 µM forskolin, 10 nMPMA, or 10 nM 4α-PMA. The number of oocytes in an experiment is given inparentheses. Experiments were performed at a holding potential of -80 mV.

Chapter 2

38

We recorded NMDA-induced currents from oocytes injected with cRNAencoding NR1b, NR1b/NR2A, NR1b/NR2B, and NR1b/NR2A/NR2B subunits, themain subunits found in the adult rat hypothalamus (Laurie et al., 1997). Averageresponse amplitudes were 36.8 ± 3.0 nA (n = 20) for oocytes expressing NR1b, 476± 207 nA (n = 15) for oocytes expressing NR1b/NR2A, 13,390 ± 2,389 nA (n = 17)for oocytes expressing NR1b/NR2B, and 21,847 ± 7,354 nA (n = 15) for oocytesexpressing NR1b/NR2A/NR2B. In contrast to the potentiation of NMDA-inducedcurrent amplitudes recorded from oocytes injected with hypothalamic mRNA,incubation with forskolin did not affect NMDA-induced currents from recombinantNMDA receptors (Table 1). However, a 1-min incubation with PMA (10 nM), apotent protein kinase C (PKC) activator, resulted in a transient potentiation of theNMDA-induced current responses. This potentiation reached a maximum value after17 min (Table 1). The inactive PMA analogue 4α-PMA (10 nM) had no effect(Table 1).

Inhibitors of PKA or protein phosphatases prevent forskolin-mediatedpotentiation of hypothalamic NMDA-induced currents

Because NMDA-induced responses can not only be modulated by proteinkinases but also by endogenous protein phosphatases, we studied the effect ofprotein phosphatase inhibition on hypothalamic NMDA-induced currents. Whenoocytes injected with hypothalamic mRNA were incubated with Cal A (100 nM, 5min), a potent inhibitor of protein phosphatase 1 (PP1) and/or 2A (PP2A) (Ishiharaand Martin, 1989; Suganuma et al., 1990), the NMDA-induced current amplitudeincreased to 159 ± 4% (n = 3) and reached a maximal potentiation after 50 - 70 min(Fig. 2). The potentiation of hypothalamic NMDA responses observed afterapplication of Cal A did not significantly differ from the value obtained afterforskolin application.

Figure 2. Enhancement of hypothalamic NMDA-induced responses by Cal A. NMDA-induced currents were measured before (control) and 50-70 min after Cal A (100 nM,5 min) application.

Next, we were interested in whether forskolin could still enhance NMDA responsesafter Cal A treatment. To circumvent run-down problems duringelectrophysiological recordings over several hours, oocytes were preinjected insteadof incubated with Cal A to a final intracellular concentration of 500 nM 50 - 70 minbefore voltage voltage-clamping. Subsequent incubation with forskolin could no

2 s20 nAControl

Calyculin A (100 nM)

NMDA NMDA

-80 mV


39

longer enhance NMDA-induced currents (99 ± 1%; n = 5; Fig. 3A). The forskolin-mediated potentiation could also be blocked (99 ± 0.7%; n = 5; Fig. 3A) bypreincubation of oocytes with a cell-permeable, selective, potent inhibitor of PKA,H-89 (10 µM, 3 h). A 1-min incubation with PMA (10 nM) also increased NMDA-induced current amplitudes from oocytes injected with hypothalamic mRNA. Thispotentiation reached a maximum 17 min after the start of the incubation.Preinjection of oocytes with Cal A (500 nM) had no significant effect on the PMA-mediated potentiation (Fig. 3B).

Figure 3. Cal A antagonizes PKA- but not PKC-mediated potentiation ofhypothalamic NMDA responses. The time course of NMDA-induced currentamplitudes was recorded from oocytes injected with hypothalamic mRNA with orwithout preinjection of Cal A to a final intracellular concentration of 500 nM 50-70min before voltage clamping. The intracellular concentration of Cal A was calculatedby assuming standard oocytes with a volume of 0.5 µl. The horizontal bar indicates a1-min drug application of either (A) control solution or forskolin (50 µM) or (B)control solution or PMA (10 nM). Some oocytes were incubated for 3 h in 10 µM H-89 before forskolin application (A). Data are mean ± SEM (bars) values of three tofive oocytes.

DISCUSSION

In the present study we investigated the modulation of hypothalamic NMDAreceptors by PKA and PP1 and/or PP2A. NMDA-induced currents recorded fromoocytes injected with rat hypothalamic mRNA were increased after stimulation ofPKA, whereas PKA stimulation had no significant effect on NMDA currents fromrecombinant NMDA receptors. Recently, Leonard and Hell (1997) demonstratedthat PKA and members of the PKC family phosphorylate all three NMDA receptorsubunits in vitro and in vivo. The main phosphorylation sites of PKA and PKC aresubstantially different for NR1, NR2A, and NR2B in agreement with the finding that

ControlForskolin (50 µM)Forskolin (CalA preinj.)

0.5

1

1.5

2

-5 0 5 10 15 20Time (min)

Rel

ativ

e cu

rren

t am

plitu

de

Forskolin (H-89 preincub.)

PMA (10 nM)Control

PMA (10 nM; CalA preinj.)

0.51

1.5

3

-5 0 5 10 15 20

22.5

25 30

3.5

Time (min)35

BA

Chapter 2

40

[32P]phosphate incorporation by PKA and PKC is additive. However, both kinasesare equally efficient in phosphorylating NR1 in vitro (Leonard and Hell, 1997).Thus, the question arises why NMDA currents in oocytes expressing NMDAreceptor subunits were not sensitive to forskolin. It is highly unlikely that forskolindid not enhance NMDA responses of recombinant NMDA receptors simply becausethe receptors were already fully phosphorylated and could therefore not produce anyfurther enhancement because PKC stimulation could still potentiate NMDA-inducedcurrents from recombinant receptors in our study, Moreover, Xenopus oocytes havebeen shown to express endogenously all necessary components to activate PKAeffectively (Keller et al., 1992). Therefore, it does not seem probable that forskolinwas not efficient in stimulating PKA when cRNA coding for receptor subunits wasinjected instead of hypothalamic mRNA.

Although the NMDA receptor was shown to have distinct PKA and PKCphosphorylation sites, only the forskolin-mediated enhancement of hypothalamicNMDA responses was prevented by pretreatment with the general PP1 and PP2Ainhibitor Cal A. PMA was still able to enhance NMDA responses after Cal Atreatment. In agreement with our data obtained using the Xenopus oocytesexpression system, Westphal et al. (1999) found that responses recorded from HEK293 cells expressing heteromeric NMDA receptors consisting of NR1a and NR2Aare also barely susceptible to activation of the PKA system. However, thisenhancement is intensified if yotiao, an NMDA receptor-associated protein that canbind PP1 and the PKA holoenzyme, is coexpressed with NMDA receptor subunits.It is interesting that anchored PP1 is active and thereby tonically inhibits nearbyNMDA receptors (Westphal et al., 1999). This finding is in support of ourobservation that inhibition of protein phosphatase activity by Cal A enhancedhypothalamic NMDA responses. Westphal et al. (1999) further demonstrated that, inthe presence of yotiao, PKA activation overcomes the tonic PP1 activity, causingrapid enhancement of NMDA receptor currents. We might not have observed anyfurther forskolin effect after Cal A treatment because the Cal A application alreadycompletely shifted the equilibrium from dephosphorylation to phosphorylation bythe inhibition of PP1.

So far, all electrophysiological experiments investigating the modulation ofNMDA responses by PKA activation documented a clear brain region-specificinteraction. In previous studies, we demonstrated that NMDA responses recordedfrom oocytes injected with total rat brain, hippocampal, or cortical mRNA were notaffected by incubation with forskolin (Blank et al., 1996; 1997; T.B. and I.N.,unpublished data). Similarly, whole-cell recordings with the standard patch-clampmethod showed that forskolin had no effect on current responses to NMDA incultured hippocampal pyramidal neurons (Greengard et al., 1991). On the otherhand, stimulation of PKA enhances NMDA responses, for example, in rat spinaldorsal horn neurons (Cerne et al., 1993) and rat neostriatal neurons (Colwell andLevine, 1995).

However, these studies did not clearly identify the target substrate of PKAaction. Previously, we showed the involvement of another subcellular signalingmolecule, the 32-kDa dopamine- and cAMP-regulated phosphoprotein (DARPP-32),in cAMP-dependent modulation of striatal NMDA responses (Blank et al., 1997).Like yotiao, DARPP-32 shifts the equilibrium in favor of phosphorylation following


41

PKA stimulation. Future studies have to determine which signaling molecule mightbe involved in the PKA-mediated potentiation of hypothalamic NMDA responses. Apossible candidate is the phosphoprotein inhibitor-1, which has a high sequencehomology with DARPP-32 and was shown to be abundant in several hypothalamicnuclei (Gustafson et al., 1991). Like DARPP-32, inhibitor-1 is converted from aninactive form to a potent and specific inhibitor of PP1 by PKA phosphorylation(Hemmings et al., 1992).

Our finding that Cal A did not significantly lower the effectiveness of PMAto enhance NMDA currents points to a different underlying mechanism of PKC-mediated potentiation than overcoming endogenous protein phosphatase activity.Recent studies have indicated that calmodulin binds directly to the NR1 subunit ofthe NMDA receptor complex at two distinct sites in the COOH-terminal region,causing a decrease in NMDA channel activity by reduction of the probability ofchannel opening (Ehlers et al., 1996). PKC-mediated phosphorylation on serineresidues of the NR1 subunit decreases its affinity for calmodulin and thereby inhibitsthe inactivation of NMDA receptors (Hisatsune et al., 1997). Because calmodulin isan ubiquitous molecule that is also enriched in Xenopus oocytes (Lan et al., 1996;Peracchia et al., 1996), it is conceivable that PKC, in contrast to PKA, potentiatescurrents evoked from both recombinant NMDA receptors and hypothalamic NMDAreceptors by reducing the affinity of calmodulin to NR1.

This study provides evidence that the function of hypothalamic NMDAreceptors can be regulated by PKA and suggests the involvement of an additionalNMDA receptor-associated molecule that acts most likely via the inhibition oftonically active protein phosphatases. Thus, several PKA-coupled neurotransmittersystems might use this cascade in the rat hypothalamus to interfere with thesignaling of NMDA receptors.

ACKNOWLEDGMENTS

We would like to thank Dr. Michael Hollmann for kindly providing thecDNAs encoding the NR1b, NR2A and NR2B subunits.

REFERENCES

Arnauld E, Layton B, Padjen A and Renaud LP (1983) Actions of acidic amino acids on the excitabilityof medial hypothalamic neurons in the rat. Neuroendocrinol. 37: 184-192.

Blank T, Zwart R, Nijholt I and Spiess J (1996) Serotonin 5-HT2 receptor activation potentiates N-

methyl-D-aspartate receptor-mediated ion currents by a protein kinase C-dependent mechanism. J.Neurosci. Res. 45: 153-160.

Blank T, Nijholt I, Teichert U, Kügler H, Behrsing H, Fienberg A, Greengard P and Spiess J (1997)The phosphoprotein DARPP-32 mediates cAMP-dependent potentiation of striatal N-methyl-D-aspartate responses. Proc. Natl. Acad. Sci. USA 94: 14859-14864.

Chapter 2

42

Brann DW and Mahesh VB (1994) Excitatory amino acids: function and significance in reproductionand neuroendocrine regulation. Frontiers in Neuroendocrinol. 15: 3-49.

Cathala G, Savouret JF, Mendez M, West BL, Karin M, Martial JA and Baxter JD (1983) A method forisolation of intact, translationally active ribonucleic acid. DNA 2: 329-335.

Cerne R, Rusin KI and Randic M (1993) Enhancement of the N-methyl-D-aspartate response in spinaldorsal horn neurons by cAMP-dependent protein kinase. Neurosci. Lett. 161: 124-128.


Ehlers MD, Zhang S, Bernhardt JP and Huganir RL (1996) Inactivation of NMDA receptors by directinteraction of calmodulin with the NR1 subunit. Cell 84: 745-755.

Goldsmith PC, Thind KK, Perera AD and Plant T (1994) Glutamate-immunoreactive neurons and theirgonadotropin-releasing hormone - neuronal interactions in the monkey hypothalamus. Endocrinol. 134:858-868.

Greengard P, Jen J, Nairn AC and Stevens CF (1991) Enhancement of the glutamate response bycAMP-dependent protein kinase in hippocampal neurons. Science 253: 1135-1138.

Gribkoff VK and Dudek FE (1990) Effects of excitatory amino acid antagonists on synaptic responsesof supraoptic neurons in slices of rat hypothalamus. J. Neurophysiol. 63: 60-71.

Gustafson EL, Girault JA, Hemmings HC, Nairn AC and Greengard P (1991) Immunocytochemicallocalization of phosphatase inhibitor-1 in rat brain. J. Comp. Neurol. 310: 170-188.

Hemmings HC, Girault J-A, Nairn AC, Bertuzzi G and Greengard P (1992) Distribution of proteinphosphatase inhibitor-1 in brain and peripheral tissues of various species: comparison with DARPP-32.J. Neurochem. 59: 1053-1061.

Hisatsune C, Umemori H, Inoue T, Michikawa T, Kohda K, Mikoshiba K and Yamamoto T (1997)Phosphorylation-dependent regulation of N-methyl-D-aspartate receptors by calmodulin. J. Biol. Chem.272: 20805-20810.

Inenaga K, Honda E, Hirakawa T, Nakamura S and Yamashita H (1998) Glutamatergic synaptic inputsto mouse supraoptic neurons in calcium-free medium in vitro. J. Neuroendocrinol. 10: 1-7.

Ishihara H and Martin BL (1989) Calyculin A and okadaic acid: inhibitors of protein phosphataseactivity. Biochem. Biophys. Res. Comm. 159: 871-877.

Keller BU, Hollmann M, Heinemann S and Konnerth A (1992) Calcium influx through subunitsGluR1/GluR3 of kainate/AMPA receptor channels is regulated by cAMP dependent protein kinase.EMBO J. 11: 891-896.

Kelso SR, Nelson TE and Leonard JP (1992) Protein Kinase C mediated enhancement of NMDAcurrents by metabotropic glutamate receptors in Xenopus oocytes. J. Physiol. (Lond.) 449: 705-718.

Kus L, Handa RJ, Sanderson JJ, Kerr JE and Beitz AJ (1995) Distribution of NMDAR1 subunit mRNAand [125I]MK-801 binding in the hypothalamus of intact, castrate and castrate-DHTP treated male rats.Mol. Brain Res. 28: 55-60.

Lan L, Bawden MJ, Auld AM and Barritt GJ (1996) Expression of Drosophila trpl cRNA in Xenopuslaevis oocytes leads to the appearance of a Ca2+ channel activated by Ca2+ and calmodulin, and byguanosine 5’(gamma-thio)triphosphate. Biochem. J. 316: 793-803.


43

Laurenza A, Sutkowski EM and Seamon KB (1989) Forskolin: a specific stimulator of adenylyl cyclaseor a diterpene with multiple sites of action? Trends Pharmacol. Sci. 10: 442-447.

Laurie DJ, Bartke I, Schoepfer R, Naujoks K and Seeburg PH (1997) Regional, developmental andinterspecies expression of the four NMDAR2 subunits, examined using monoclonal antibodies. Mol.Brain Res. 51: 23-32.


Leonard JP and Kelso SR (1990) Apparent desensitization of NMDA responses in Xenopus oocytesinvolves calcium dependent chloride current. Neuron 2: 53-60.

Peracchia C, Wang X, Li L and Peracchia LL (1996) Inhinition of calkmodulin expression preventslow-pH-induced gap junction uncoupling in Xenopus oocytes. Pflueg. Arch. Eur. J. Physiol. 431: 379-387.

Seamon KB and Daly JW (1986) Forskolin: its biological and chemical properties. Adv. CyclicNucleotide Protein Phosphorylation Res. 20: 1-150.

Stühmer W (1992) Electrophysiological recording from Xenopus oocytes. Methods Enzymol. 207: 319-339.

Suganuma M, Fujiki H, Furuya-Suguri H, Yoshizawa S, Yasumoto S, Kato Y, Fusetani N andSugimura T (1990) Calyculin A, an inhibitor of protein phosphatases, a potent tumor promoter on CD-1mouse skin. Cancer Res. 50: 3521-3525.

Van den Pol A. (1991) Glutamate and aspartate immunoreactivity in hypothalamic presynaptic axons.J. Neurosci. 11: 2087-2101.

Van den Pol A, Hermans-Borgmeyer I, Hofer M, Ghosh P and Heinemann S (1994) Ionotropicglutamate-receptor gene expression in hypothalamus: localization of AMPA, kainate, and NMDAreceptor RNA with in situ hybridization. J. Comp. Neurol. 343: 428-444.

Van den Pol A, Waurin J and Dudek F (1990) Glutamate, the dominant excitatory transmitter inneuroendocrine regulation. Science 250: 1276-1278.

Waurin J and Dudek FE (1991) Excitatory amino acid antagonists inhibit synaptic responses in theguinea pig hypothalamic paraventricular nucleus. J. Neurophysiol. 65: 946-951.

Westphal RS, Tavalin SJ, Lin JW, Alto NM, Fraser IDC, Langeberg LK, Sheng M and Scott JD (1999)Regulation of NMDA receptors by an associated phosphatase-kinase signaling complex. Science 285:93-96.

CHAPTER 3

The phosphoprotein DARPP-32 mediates cAMP-

dependent potentiation of striatal N-methyl-D-

aspartate responses

Proc. Natl. Acad. Sci. USA (1997) 75: 749-754

ABSTRACT

The signal transduction pathway underlying the cAMP-dependentmodulation of rat striatal N-methyl-D-aspartate (NMDA) responses was investigatedby using the two-electrode voltage-clamp technique. In oocytes injected with ratstriatal poly (A)+ mRNA, activation of cAMP-dependent protein kinase (PKA) byforskolin potentiated NMDA responses. Inhibition of protein phosphatase 1 (PP1)and/or protein phosphatase 2A (PP2A) by the specific inhibitor calyculin Aoccluded the PKA-mediated potentiation of striatal NMDA responses, suggestingthat the PKA effect was mediated by inhibition of a protein phosphatase.Coinjection of oocytes with striatal mRNA and antisense oligodeoxynucleotidesdirected against the protein phosphatase inhibitor DARPP-32 dramatically reducedthe PKA enhancement of NMDA responses. NMDA responses recorded fromoocytes injected with rat hippocampal poly (A)+ mRNA were not affected bystimulation of PKA. When oocytes were coinjected with rat hippocampal poly (A)+

mRNA plus complementary RNA coding for DARPP-32, NMDA responses werepotentiated after stimulation of PKA. The results provide evidence that DARPP-32,which is enriched in the striatum, may participate in the signaling between the twomajor afferent striatal pathways, the glutamatergic and the dopaminergic projections,by the cAMP-dependent regulation of striatal NMDA currents.

Chapter 3

46

INTRODUCTION

The striatum receives dense dopaminergic afferents from midbrain areas(Doucet et al., 1986; Voorn et al., 1986) and glutamatergic projections from all areasof the cerebral cortex (Fonnum et al., 1981; McGeer et al., 1977) and thalamus(Dube et al., 1988). Ionotropic glutamate receptors have been traditionally classifiedinto three families on the basis of pharmacological and electrophysiological data: N-methyl-D aspartate (NMDA), amino-3-hydroxy-5-methyl-4-isoxazole propionate(AMPA), and kainate (KA) receptors. In the striatum excessive activation ofNMDA receptors has been associated with neurological disorders such asHuntington`s (Choi, 1988; DiFiglia, 1990) and Parkinson`s disease (Greenamyreand O’Brien, 1991).

Phosphorylation of neurotransmitter receptors plays an important role in themodulation of their function. NMDA receptors contain multiple consensus sites forphosphorylation by several protein kinases (Hollmann and Heinemann, 1994).Protein kinase C (PKC) and tyrosine kinases have been shown to phosphorylateNMDA receptor subunits directly (Lau and Huganir, 1995; Moon et al., 1994;Tingley et al., 1993). Moreover, forskolin, dopamine and a D1 agonist increased thephosphorylation of the NMDAR1 receptor apparently through activation of cAMP-dependent protein kinase (PKA) (Snyder et al., 1996). In rat neostriatal brain slicesNMDA-induced excitatory synaptic transmission was potentiated by the PKAactivators forskolin and Sp-cAMPS (Colwell and Levine, 1995). Furthermore,activation of D1 receptors potentiated NMDA-induced responses (Cepeda et al.,1993), whereas stimulation of metabotropic glutamate receptors had inhibitoryeffects on NMDA receptor function (Colwell and Levine, 1994). In both cases, itwas suggested that a cAMP/PKA-dependent mechanism was involved. However, itcould not be concluded from those experiments whether the NMDA receptorcomplex and/or an additional intracellular regulatory protein was the target of PKAphosphorylation.

In the present study, we used the Xenopus oocyte expression system tofurther unravel the molecular mechanisms underlying the modulation of striatalNMDA receptor function by PKA. In initial experiments, we recorded NMDA-induced currents from Xenopus oocytes expressing NR1b, NR1b/NR2A,NR1b/NR2B, or NR1b/NR2A/NR2B NMDA receptor subunits, the main subunitsfound in rat striatum (Petralia et al., 1994; Portera-Cailliau et al., 1996). Thesecurrents were not affected by PKA stimulation (T.B. and I.N., unpublished data)suggesting the involvement of an indirect mechanism underlying the PKA-dependent modulation of striatal NMDA receptor function.

DARPP-32 mediates PKA potentiation of striatal NMDA responses

47


RNA Preparation and Expression in Xenopus OocytesTotal RNA was isolated from striatum and hippocampus of adult Wistar rats by extraction of

fresh tissue with guanidine thiocyanate and precipitation with LiCl (Cathala et al., 1983). Poly((A)+

mRNA was purified by oligo(dT)-cellulose chromatography (Pharmacia mRNA Purification Kit) anddissolved in RNAse-free water at a concentration of 0.5 µg/µl.

Plasmid pET3A-DARPP-32 containing the cDNA coding for the rat Mr 32,000 dopamine andadenosine 3', 5'-monophosphate-regulated phosphoprotein (DARPP-32) was linearized by digestionwith NheI and used as transcription template. In vitro transcription was carried out by T7 RNApolymerase in the presence of a cap analog. RNA transcripts were precipitated by ethanol and theprecipitate was dissolved in water treated with diethyl pyrocarbonate.

Oocytes were collected from anesthetized specimens of Xenopus laevis as described (Kushneret al., 1988). Follicular cell layers were removed manually after incubation for 2 h at room temperaturein Ca2+-free modified Barth's saline solution containing 2.2 mg/ml collagenase, type II. Stage V andstage VI oocytes were pressure-injected with 50 ng of poly (A)+ mRNA per oocyte. In some

experiments, hippocampal poly (A)+ mRNA was coinjected with DARPP-32 cRNA (20 or 80 ng peroocyte). Oocytes were kept at 18°C in modified Barth's solution.

RNA (Northern) Blot AnalysisSamples were size-fractionated on 1.5% agarose-formaldehyde gels and were transferred to

nylon membranes (Hybond) according to standard protocols (Amersham). The full-length cDNAencoding DARPP-32 was labeled with [α-32P]dATP by random priming using a Prime-It II-Kit(Stratagene). Blots were hybridized at 68°C for 18 h in Church Buffer (Church and Gilbert, 1984).Blots were washed in 2x SSC/1% SDS for 5 min, twice in 1x SSC/1% SDS at 65°C for 15 min, andfinally twice in 0.1x SSC/1% SDS at 65°C for 15 min. They were exposed to x-ray film with anintensifying screen at -70°C.

Antisense Oligodeoxynucleotides (ODNs)Antisense 14-mer (ODNs) complementary to part of the cRNA coding for the phosphoprotein

DARPP-32, and random and reversed control ODNs, were synthesized by Biognostik (Goettingen,Germany). ODNs were shipped as lyophilized DNA Na salt. The lyophilized ODNs were dissolved in1x TE buffer (10 mM Tris.HCl/1 mM EDTA, pH 7.2) to a final concentration of 250 µM and stored at -

20°C. The ODN suspension (100 nl, 2.5 µM) was injected according to the poly (A)+ mRNA injectionprotocol. The ODNs were injected 5 days after the oocytes were injected with striatal mRNA and thentested for NMDA/glycine-induced responses on day 7. The nucleotide sequences of the syntheticoligomers corresponding to base numbers 995-1009 of the rat DARPP-32 cDNA were as follows:antisense-DARPP-32, 5'-CTGAGCTTATGTGC-3'; reversed control-DARPP-32, 5'-CGTGTATTCGAGTC-3';random control-DARPP-32, 5'-AGTCGCTGTGATCT-3'.

Electrophysiological RecordingsSeven to eight days after injection, oocytes were placed in a recording chamber (volume ≈ 40

µl) and continuously superfused (1.5 ml/min) with recording solution at room temperature (20-24°C).The oocytes were voltage-clamped by a conventional two-microelectrode voltage-clamp technique(Stühmer, 1992). The membrane potential of the oocytes was held at -80 mV (unless otherwiseindicated), using a Turbo Clamp Tec 01C amplifier (N.P.I. Electronic, Tamm, Germany). The twomicroelectrodes were filled with 3 M KCl and had resistances between 1 and 3 MΩ. Drugs wereapplied with a micropipette (200 µl) after stopping solution flow. If not indicated otherwise, NMDAand glycine were co-applied at concentrations of 100 µM and 10 µM, respectively. In someexperiments, oocytes were preincubated with a drug-containing solution. During incubation, oocyteswere continuously superfused with solutions containing the drugs to be applied. Calyculin A wasprepared in 100% ethanol as a 1 mM stock solution. Immediately after dilution in distilled water to theappropriate concentration, calyculin A was microinjected into the oocytes according to the RNAinjection protocol. Oocytes were preinjected 1-1.5 h before voltage-clamping. Aliquots of concentratedstock solutions of phorbol 12-myristate 13-acetate (PMA), 4-α-phorbol 12-myristate 13-acetate (4-

Chapter 3

48

α−PΜΑ), forskolin, and 1,9-dideoxyforskolin in dimethyl sulfoxide (DMSO) and NMDA and glycinein distilled water were added to the recording solution immediately before the experiments. 8-Bromo-adenosine 3`,5`monophosphate (8-Br-cAMP, sodium salt) was directly dissolved to the finalconcentration in the external solution. The final DMSO as well as the final ethanol concentration neverexceeded 0.1% (v/v), which, upon application, induced no change in membrane current on its own andwhich had no effect on NMDA-induced ion currents in control experiments. Effects of the variouscompounds tested were studied after at least two or three identical responses to NMDA could beelicited. Oocytes were not used for experiments when reproducible responses to NMDA could not berecorded within 60 min. Current signals were low-pass filtered at 30 Hz using a four-pole Bessel filterand digitized by an ITC 16-MAC interface (Instrutech , Great Neck, NY). Data were sampled at 100Hz and stored on a Macintosh (7100/66) computer using data acquisition software (Pulse 7.40, HEKAElectronic, Lambrecht/Pfalz, Germany). The relative current amplitude of 100 µM NMDA- and 10 µMglycine-induced inward currents is defined as the ratio (I/Ic) of the maximum current amplitude after (I)and before (Ic) treatment with drugs.

StatisticsStatistical comparisons were made by using Student's t test. Results were expressed as mean

± standard error of the mean (SEM). p-values ≥ 0.05 were regarded as not significant.

SolutionsModified Barth's solution contained (in mM): 88 NaCl, 1 KCl, 2.4 NaH CO3, 15 Hepes, 0.8

MgSO4, 0.3 Ca(NO3)2 and 0.6 CaCl2, and 50 units/ml penicillin and 50 µg/ml streptomycin (adjusted topH 7.6 with NaOH). Ca2+-free modified Barth's solution contained (in mM): 88 NaCl, 1 KCl, 2.4NaHCO3, 15 Hepes, 0.8 MgSO4 (adjusted to pH 7.6 with NaOH). Electrophysiological recordings werecarried out in saline solution containing (in mM): 115 NaCl, 2.5 KCl, 1.8 BaCl2, 10 Hepes (pH adjustedto 7.2 with NaOH).

MaterialsMature female specimens of Xenopus laevis were obtained from Kähler (Hamburg,

Germany). Collagenase, type II, was from Worthington, and mRNA Purification Kit was fromPharmacia LKB Biotechnology. PMA and 4-α-PMA were obtained from Research Biochemicals(Natick, MA). Calyculin A was from Boehringer Mannheim. 1,9-Dideoxyforskolin and 8-Br-cAMP(sodium salt) were from Calbiochem (. Penicillin and streptomycin were from Gibco. All other drugsand salts were purchased from Sigma. Molecular biology reagents and restriction enzymes werepurchased from Pharmacia, Promega and New England Biolabs.

RESULTS

PKC and PKA modulate striatal NMDA responses.The effects of protein kinases on NMDA responses were studied by using

Xenopus oocytes injected with mRNA isolated from rat striatum or hippocampus.Coapplication of NMDA and glycine at a holding potential of -80 mV elicitedresponses with the conventional properties of NMDA receptors expressed in oocytes(e.g., Kelso et al., 1992; Leonard and Kelso, 1990). When NMDA was appliedwithout glycine, the amplitude of the NMDA responses was markedly reduced. TheNMDA-induced ion currents were nearly completely inhibited by 20 µM of thecompetitive antagonist 2-amino-5-phosphonovaleric acid (APV). NMDA responseswere almost completely blocked by coapplication of 1 mM Mg2+ or 1 µM MK-801,a noncompetitive NMDA receptor antagonist (data not shown). Repeatedapplications of NMDA and glycine with 4-min washing intervals between


49

subsequent applications resulted in a stable NMDA-induced inward current(indicated as control values; Fig. 1; Table 1).

Figure 1. Striatal NMDA-induced currents are potentiated by forskolin and PMA.The bar indicates drug superfusion. External solution was applied as control (x). (A)

Forskolin (50 µM;•), PMA (10 nM; ), 1,9-dideoxyforskolin (50 µM; o) and 4-α-

PMA (10 nM; ) were applied for 1 min. (B) After a 10 min incubation with 8-Br-

cAMP (10 mM; •), a 1 minute incubation with PMA (10 nM; ) followed. Each

point represents the mean ± SEM of three to seven oocytes.

After incubation with PMA, a potent PKC activator (Stea et al., 1995), arapid potentiation of striatal NMDA-induced responses occurred, and a continuousrise was observed until a maximum was reached after 13 min (Fig. 1A; Table 1).Subsequently, the current amplitude decreased again (Fig. 1A). The inactive isomer4-α-PMA had no significant effect (Fig. 1A; Table 1). Incubation with forskolin,which activates PKA by means of cAMP stimulation (reviewed in Seamon andDaly, 1986), enhanced NMDA-induced currents in a time-dependent manner with amaximum at 9 min (Fig. 1A; Table 1). Because forskolin has been reported to actdirectly on ion channels in addition to its action on the cAMP cascade (e.g. Laurenzaet al., 1989), the inactive forskolin analog 1,9-dideoxyforskolin was applied to testfor this possible mode of action. This compound, which does not stimulate adenylylcyclase, but mimics other actions of forskolin, did not exhibit a significantpotentiating effect on NMDA-induced currents (Fig.1A; Table 1).

Oocytes contain endogenous chloride channels that can be activated by anincrease in the intracellular calcium concentration (Barish, 1983). Thus, influx ofcalcium through NMDA receptor channels (MacDermott et al., 1986) may lead to asecondary activation of endogenous chloride channels. To exclude the involvementof these chloride channels in the observed potentiation of NMDA-induced currents,all experiments were performed by replacing extracellular Ca2+- with Ba2+-ionswhich are known to be less effective than Ca2+-ions in activating the calcium-dependent chloride currents in oocytes (Barish, 1983; Miledi and Parker, 1984). The

Forskolin PMA Control1,9 Dideoxyforskolin 4 α-PMA

0.5

1

1.5

2

2.5

3

-5 0 5 10 15 20 25 30

A 8-Br-cAMPPMAControl

0.5

1.5

2.5

3.5

4.5

5.5

-10 0 10 20 30 40Time (min)

B

I/Ic

Time (min)

Chapter 3

50

influence of endogenous chloride channels has further been excluded by repeatingthe experiments at a holding potential of -25 mV, which is the reported reversalpotential for chloride in oocytes (Barish, 1983; Miledi, 1992). At this holdingpotential, the potentiation induced by PMA and forskolin was not significantlydifferent from the one observed at -80 mV (P>0.05; Table 1). It was concluded thatthe observed potentiation was not due to the modulation of endogenous chloridechannels.

Table 1. Effect of protein kinase modulators on striatal NMDA-induced currents

CompoundI/Ic*

mean ± SEM n P

Control (9 min) 0.92 ± 0.02 3 -

Control (13 min) 0.92 ± 0,04 3 -

Control (14 min) 0.91 ± 0,01 4 -

PMA (10 nM)‡ 2.64 ± 0,33 6 ≤ 0.01

PMA (10 nM; -25 mV)‡ 2.40 ± 0.124 4 ≤ 0.001

4-α-PMA (10 nM)‡ 1.07 ± 0.04 4 N.S.

Forskolin (50 µM)† 1.45 ± 0.03 4 ≤ 0.001

Forskolin (50 µM; -25 mV)† 1.50 ± 0.10 4 ≤ 0.01

1,9-Dideoxyforskolin (50 µM)† 0.93 ± 0.043 3 N.S.

8-Br- cAMP (10 mM)§ 1.86 ± 0.137 7 ≤ 0.001

Experiments were performed at a holding potential of -80 mV unless indicatedotherwise.*The relative current amplitude is determined by the ratio (I/Ic) of steady-stateresponses to 100 µM NMDA and 10 µM glycine after (I) and before (Ic) drugtreatment. With the exception of 8-Br- cAMP, which was applied for 10 minutes, allother drugs were applied for 1 min. Results are mean ± SEM.p: p-values of Student's t test obtained by comparing drug-treated oocytes with non-

treated control oocytes. For comparisons controls recorded at †9 min, ‡13 min and§14 min were taken. NS, not significant.

Incubation with 8-Br-cAMP, the membrane-permeant analog of cAMP,increased striatal NMDA receptor-induced responses to 186% ± 13% (n = 7) ofcontrol. Interestingly, incubation with PMA produced further potentiation. Theadditional increase did not deviate significantly from the potentiation found withPMA alone (Fig. 1B).

PMA incubation of oocytes injected with hippocampal mRNA resulted in atransient potentiation of the NMDA-induced current responses. This potentiationreached a maximum value within approximately 13 min (Table 2). Similarincubation with the biologically inactive analog 4-α-PMA did not induce any


51

change in responses to the agonists (93% ± 7%, n = 3; Table 2). At a holdingpotential of -25 mV the maximal increase of NMDA-induced ion currents afterincubation with PMA was 227% ± 19% (n = 6). This degree of potentiation was notstatistically different from the potentiation (241% ± 32%, n = 9) found at a holdingpotential of -80 mV (P>0.05;Table 2). In contrast to the observed enhancement ofthe striatal NMDA response to PKA activation, incubation with forskolin did notsignificantly alter hippocampal NMDA responses (Fig. 4A; Table 2).

Table 2. Effect of protein kinase modulators on hippocampal NMDA-induced currents

CompoundI/Ic*

mean ± SEM n P

Control (9 min) 1.01 ± 0.03 5 -

Control (13 min) 1.02 ± 0.03 5 -

PMA (10 nM)‡ 2.41 ± 0.32 9 ≤ 0.01

PMA (10 nM; -25 mV)‡ 2.27 ± 0.19 6 ≤ 0.001

4-α-PMA (10 nM)‡ 0.93 ± 0.07 3 N.S.

Forskolin (50 µM)† 1.01 ± 0.03 5 N.S.

*Data presented as for table 1.

Striatal and hippocampal NMDA responses are regulated by endogenousprotein phosphatases.

The possibility that NMDA responses could be regulated not only by proteinkinases but also by endogenous protein phosphatases was investigated by 5 minapplication of calyculin A (100 nM), a potent inhibitor of PP1 and PP2A (Ishihira etal., 1989; Suganuma et al., 1990). Following this treatment, NMDA responsesincreased and reached maximal potentiation after 70-90 min. Striatal NMDA-induced currents were potentiated to 157% ± 12% (n = 5) and hippocampal NMDA-induced currents were enhanced to 137% ± 8% (n = 4) of the corresponding controlvalues. The enhancement of striatal NMDA responses observed after application ofcalyculin A did not differ significantly from the value obtained after PKAstimulation.

Calyculin A antagonizes PKA- but not PKC-mediated potentiation of striatalNMDA responses.

In oocytes injected with striatal mRNA, the potentiation of NMDA-inducedcurrents observed after forskolin application (Fig. 2A; Table 1) could be preventedby preinjection of calyculin A 60-90 min before voltage-clamping (Fig. 2B). Theinhibitory effect of calyculin A on the PKA-mediated potentiation of striatal NMDAresponses was dose-dependent. A 50% reduction (IC50) of the potentiation wasobserved at 0.56 nM calyculin A (final intracellular concentration). The PMA-

Chapter 3

52

induced potentiation of striatal NMDA responses (Fig. 2C; Table 1) was notsignificantly altered when oocytes were preinjected with calyculin A (Fig. 2D).

Figure 2. Calyculin A inhibitsforskolin- but not PMA-mediatedpotentiation of striatal NMDAresponses. The NMDA-inducedcurrents of striatal mRNA injectedoocytes were measured before and8 min after 1-minute application of50 µM forskolin (A) or before and12 min after 1-minute applicationof 10 nM PMA (C). (B and D) Thesame protocol was followed afterinjecting oocytes with calyculin Ato a f inal intracel lularconcentration of 500nM 60-90 minbefore voltage-clamping. Theintracellular concentration ofcalyculin A was calculated byassuming standard oocytes with avolume of 0.5 µl.

Expression of DARPP-32 renders hippocampal NMDA receptors sensitive toPKA.

The observation that calyculin A was able to prevent the PKA-mediatedpotentiation of striatal NMDA responses suggested the involvement of a third-messenger, which would inhibit protein phosphatases after stimulation of PKA. Onepossible molecule to be considered was DARPP-32, which is highly enriched in ratstriatum (Hemmings and Greengard, 1986; Walaas and Greengard, 1984). A varietyof studies demonstrated that DARPP-32 is converted into a potent inhibitor of PP1by PKA-mediated phosphorylation (Hemmings et al., 1992; Hemmings et al., 1984).

Figure 3. Northern blot analysis of poly(A)+mRNAisolated from rat striatum and hippocampus. Ineach lane 3 µg RNA was loaded. The full-lengthcDNA encoding DARPP-32 was radioactivelylabeled and used as a probe. The lower bands showthe hybridization of the RNA probes with mouse β-actin.

4 s50 nA

Forskolin (50 µM)

NMDA

Control

B

Control

Forskolin (50 µM)

NMDA

4 s100 nA

A

PMA (10 nM)

NMDA

Control

4 s100 nA

C

-80 mVPMA (10 nM)

NMDAControl

4 s100 nA

D

-80 mV

Calyculin A-preinjected oocytes

2.4 kb

1.4 kb

β-Actin

striat

al mRNA

hippo

c. mRNA


53

The amount of DARPP-32 mRNA in mRNA aliquots used for injecting theoocytes was assessed by Northern blot analysis. Hybridization with radioactivelylabeled DARPP-32 cDNA revealed an RNA doublet of approximately 1.8 and 2 kbin striatum. By contrast, only extremely small amounts of DARPP-32 mRNA weredetectable in hippocampal mRNA (Fig. 3). Hybridization with β-actin demonstratedthat equal amounts of mRNA were used for each sample.

To investigate whether activation of DARPP-32 could cause anenhancement of NMDA-induced currents after stimulation of PKA, hippocampalmRNA was coinjected with DARPP-32 cRNA. Under these conditions, incubationwith forskolin potentiated NMDA-induced currents within 9 min to 123% ± 6% (n =3; 20 ng DARPP-32 cRNA per oocyte) or 128% ± 4% of control values (n = 8; 80ng DARPP-32 cRNA per oocyte; Fig. 4B). This enhancement of hippocampalNMDA responses was not significantly different from the one observed aftercalyculin A treatment. The inactive forskolin analog 1,9-dideoxyforskolin had noeffect (Fig. 4B). When oocytes were preinjected with calyculin A 60-90 min beforevoltage-clamping, forskolin could not alter NMDA-induced currents. These dataresembled the results observed with oocytes injected with striatal mRNA.

Figure 4. Forskolin-mediatedpotentiation of hippocampalNMDA responses is observedin the presence of DARPP-32.

Forskolin (50 µM,• ), 1,9-

dideoxyforskolin (50 µM; )and external solution (control,x) were applied for 1 minute asindicated by the bar. NMDA-induced current responses wererecorded from oocytes injectedwith hippocampal mRNA (A)or oocytes injected withhippocampal mRNA andDARPP-32 cRNA (80 ng peroocyte) (B ). Each pointrepresents the mean ± SEM ofthree to nine oocytes.

Inhibition of DARPP-32 expression reduces the PKA-mediated Potentiation ofstriatal NMDA responses.

Oocytes injected with striatal mRNA and antisense ODNs directed againstDARPP-32 were tested for PKA-mediated potentiation of NMDA-induced currents.After injection of the antisense ODNs in striatal mRNA-injected oocytes, PKAstimulation caused a markedly reduced potentiation of NMDA-induced currents(112% ± 4%, n = 7; Fig. 5) compared to control NMDA-induced currents recordedfrom oocytes injected with striatal mRNA and TE buffer (160% ± 12%, n = 4; Fig.

1.5

1.0

0.5

1.5

1.0

0.5

-5 0 5 10 15 20 25 30

I/Ic

I/Ic

Time (min)-5 0 5 10 15 20 25 30

Forskolin1,9-DideoxyforskolinControl

ForskolinControl

A

B

Chapter 3

54

5). The specificity of the antisense ODNs was assessed by injecting striatal mRNA-injected oocytes with either random or reversed control ODNs. Neither random norreversed control ODNs abolished the PKA-mediated potentiation of NMDAresponses (163% ± 16%, n = 4; 158% ± 6%, n = 4; Fig. 5, respectively).

Figure 5. Forskolin-mediated potentiation of striatal NMDA responses requiresDARPP-32. ( A) Representative NMDA responses of oocytes injected with striatalmRNA and ODNs or TE buffer before (controls) and 8 min after 1-minute applicationof forskolin (50 µM). (B ) Average relative current amplitudes (I/Ic) of theexperiments presented in A. Each bar represents the mean ± SEM of 4 or 7 oocytes asindicated.

DISCUSSION

PMA enhanced the NMDA receptor function in Xenopus oocytes injectedwith rat mRNA isolated from either striatum or hippocampus. In contrast, forskolinenhanced the NMDA-induced currents recorded from oocytes injected with striatalmRNA but not hippocampal mRNA. These findings were in general agreement with

ControlNMDA

Forskolin (50 µM)

ControlNMDA

Forskolin (50 µM)

ControlNMDA

TEAntisense

Forskolin (50 µM)

Control

NMDA

Random Reversed

Forskolin (50 µM)

4 s20 nA

-80 mV

-80 mV

A

B

I/Ic

1.25

1.5

1.75

(7)

(4)(4)

1

2

Antisense

Random

Reversed

TE

(4)


55

previous experiments performed in rat hippocampal and striatal neurons (Aniksztejnet al., 1992; Colwell and Levine, 1995; Greengard et al., 1991; Harvey andCollingridge, 1993; O’Connor et al., 1995; Wang et al., 1991). PKC and PKA actedadditively, indicating that the two enzymes exerted their modulatory action on theNMDA receptor complex by phosphorylation at separate sites.

The finding that PKA-induced potentiation of striatal NMDA responses wasprevented by preinjection of oocytes with the protein phosphatase inhibitor calyculinA suggested the involvement of PP1 and/or PP2A. The basal activity of endogenousprotein phosphatases was indicated by the fact that application of calyculin Aenhanced striatal and hippocampal NMDA currents. These results were in agreementwith the observation by Colwell and Levine (1995) that the phosphatase inhibitorokadaic acid enhanced excitatory synaptic transmission in the neostriatum.

After application of calyculin A, striatal NMDA currents were enhanced tothe same extent as observed after the activation of PKA. In contrast, calyculin A didnot prevent the PKC-mediated potentiation of striatal NMDA responses.

DARPP-32 is phosphorylated by PKA, and acts in its phosphorylated formas a potent and specific inhibitor of PP1 (Hemmings et al., 1984; Walaas et al.,1983; Walaas and Greengard, 1984). On the basis of our experimental data,DARPP-32 seemed to play a pivotal role in the PKA-mediated enhancement ofNMDA-induced currents. This hypothesis was strongly supported by the observationthat the PKA-mediated enhancement of striatal NMDA responses was significantlyreduced by the injection of antisense ODNs directed against mRNA coding forDARPP-32. It is also in agreement with this hypothesis that hippocampal NMDAresponses could be enhanced by forskolin when cRNA coding for DARPP-32 wascoinjected. Interestingly, the PKA-mediated enhancement of NMDA responsesrecorded from oocytes injected with hippocampal mRNA plus DARPP-32 cRNAwas similar to the enhancement of hippocampal NMDA responses following theinhibition of protein phosphatases by calyculin A. The PKA-mediated potentiationof NMDA responses recorded from striatal mRNA-injected oocytes or hippocampalmRNA-injected oocytes coexpressing DARPP-32 was prevented by preinjectingcalyculin A. Thus the effects of forskolin and of calyculin A were not additive ineither system. These results suggested that PKA, acting bymeans of phosphorylationof DARPP-32, as well as calyculin A modulated NMDA responses by inhibitingendogenous PP1.

Cepeda and colleagues (1993) reported the enhancement of NMDA-evokedexcitations in cat and rat neostriatal slices after application of D1 receptor agonists.Stimulation of D2 receptors attenuated responses evoked by NMDA. On the basis ofour findings, one can speculate that dopamine modulates striatal NMDA responsesby indirect regulation of PP1 activity via DARPP-32. In agreement with thisspeculation, previous studies have shown that D1 receptor agonists increased theamount of both cAMP (Azzaro et al., 1987) and phosphorylated DARPP-32(Szmigielski and Zalewska-Kaszubska, 1991; Walaas et al., 1983). The increasingphosphorylation of DARPP-32 was mediated through the activation of PKA(Hemmings et al., 1990; Hemmings and Nairn et al., 1984). Conversely, stimulationof D2 receptors in rat striatum reduced the amount of both cAMP (Cooper et al.,1986; Weiss et al., 1985) and phosphorylated DARPP-32 (Nishi et al., 1997). Thiseffect of D2 receptor activation on DARPP-32 appeared to be due both to decreased

Chapter 3

56

phosphorylation of DARPP-32 through the cAMP/PKA pathway and to increaseddephosphorylation of DARPP-32 through the Ca2+/calcineurin pathway (Nishi etal., 1997).

The regional distribution of DARP-32 within the rat brain is very similar tothe pattern of dopamine-innervated brain regions. Immunocytochemical studies havelocalized DARP-32 at particularly high levels in the substantia nigra, olfactorytubercle, bed nucleus of the stria terminalis, globus pallidus, caudatoputamen,nucleus accumbens, portions of the amygdaloid complex, and corticothalamicneurons (Ouimet et al., 1984). Therefore, the signal transduction cascades describedabove may contribute to synaptic plasticity in certain brain regions.Neurotransmitters (e.g. dopamine) coupled either positively or negatively to theadenylyl cyclase system could differentially modulate the intracellular cAMP leveland thereby the amount of phosphorylated DARPP-32. The concentration ofphosphorylated DARPP-32 in turn would be responsible for the degree of PP1-inhibition, thus modulating NMDA receptor function by decreasing or increasing thedephosphorylation of the NMDA receptor itself or, alternatively, a regulatoryprotein (Fig. 6). On the other hand, stimulation of NMDA receptors in rat striatalslices has been shown to induce dephosphorylation of phosphorylated DARPP-32,presumably by the calcium-dependent phosphatase calcineurin (Halpain et al.,1990). This mechanism might reflect a negative feedback-loop protecting striatalcells from NMDA receptor-mediated toxicity in the presence of high amounts ofphosphorylated DARPP-32 (Colwell et al., 1996).

Figure 6. Schematic diagram illustrating the possible role of DARPP-32 and PP1 inmediating dopamine-induced increase in conductance through NMDA receptors inrat striatum. Glu, glutamate; DA, dopamine; P, phosphate group; PK, protein kinase.

PKA

cAMP

DA

Y

D1 receptor

DARPP-32

Y

PP-1

PK

NMDA receptor

Glu

Y

Ca2+

inhibits

DARPP-32P

Ca2+

Glu

P

P

P


57

The present study provides electrophysiological evidence that PKA mayregulate NMDA receptor function by means of DARPP-32 and endogenously activeprotein phosphatases. In the DARPP-32-enriched striatum this third messengerpathway may participate in the control of the phosphorylation-state of NMDAreceptors and thereby control the excitability of striatal neurons and the efficacy ofsynaptic transmission. The complex control system of NMDA receptorphosphorylation may facilitate the integration of multiple neuronal signalsconverging on DARP-32 containing neurons.

ACKNOWLEDGMENTS

We thank Dr. Frank Dautzenberg and Dr. Sabine Sydow for discussions andGudrun Fricke-Bode for technical assistance.

REFERENCES

Aniksztejn L, Otani S and Ben-Ari Y (1992) Quisqualate metabotropic receptors modulate NMDAcurrents and facilitate induction of long-term potentiation through protein kinase C. Eur. J. Neurosci. 4:500-505.

Azzaro AJ, Liccione J and Lucci J (1987) Opposing actions of D-1 and D-2 dopamine receptor-mediated alterations of adenosine-3',5'-cyclic monophosphate (cyclic AMP) formation during theamphetamine-induced release of endogenous dopamine in vitro. Naunyn-Schmiedebergs Arch.Pharmac. 336: 133-138.

Barish ME (1983) A transient calcium-dependent chloride current in the immature Xenopus oocyte. J.Physiol. (Lond.) 342: 309-325.

Cathala G, Savouret JF, Mendez B, West BL, Karin M, Martial JA and Baxter JD (1983) A method forisolation of intact, translationally active ribonucleic acid. DNA 2: 329-335.

Cepeda C, Buchwald NA and Levine MS (1993) Neuromodulatory actions of dopamine in theneostriatum are dependent upon the excitatory amino acid receptor subtypes activated. Proc. Natl.Acad. Sci. USA 90: 9576-9580.

Choi DW (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1: 623-634.

Church GM and Gilbert W (1984) Genomic sequencing. Proc. Natl. Acad. Sci. USA 81: 1991-1995.

Colwell CS, Altemus KL, Cepeda C and Levine MS (1996) Regulation of N-methyl-D-aspartate-induced toxicity in the neostriatum: a role for metabotropic glutamate receptors? Proc. Natl. Acad. Sci.USA 93: 1200-1204.


Colwell CS and Levine MS (1994) Metabotropic glutamate receptors modulate N-methyl-D-aspartatereceptor function in neostriatal neurons. Neuroscience 61: 497-507.

Chapter 3

58

Cooper DMF, Bier-Laning CM, Halford MK, Ahlijanian MK and Zahniser NR (1986) Dopamine,acting through D-2 receptors, inhibits rat striatal adenylate cyclase by a GTP-dependent process.Molec. Pharmacol. 29: 113-119.

DiFiglia M (1990) Excitotoxic injury of the neostriatum: a model for Huntington's disease. TrendsNeurosci. 13: 286-289.

Doucet G, Descarries L and Garcia S (1986) Quantification of the dopamine innervation in adult ratneostriatum. Neuroscience 19: 427-445.

Dube´ L, Smith AD and Bolam JP (1988) Identification of synaptic terminals of thalamic or corticalorigin in contact with distinct medium-size spiny neurons in the rat neostriatum. J. Comp. Neurol. 267:455-471.

Fonnum F, Storm-Mathisen J and Divac I (1981) Biochemical evidence for glutamate asneurotransmitter in corticostriatal and corticothalamic fibres in rat brain. Neuroscience 6: 863-873.

Greenamyre JT and O`Brien CF (1991) N-methyl-D-aspartate antagonists in the treatment ofParkinson's disease. Arch. Neurol. 48: 977-981.

Greengard P, Jen J, Nairn AC and Stevens CF (1991) Enhancement of the glutamate response bycAMP-dependent protein kinase in hippocampal neurons. Science 253: 1135-1138.

Halpain S, Girault JA and Greengard P (1990) Activation of NMDA receptors inducesdephosphorylation of DARPP-32 in rat striatal slices. Nature 343: 369-372.

Harvey J and Collingridge GH (1993) Signal transduction pathways involved in the acute potentiationof NMDA responses by 1S,3R-ACPD in rat hippocampal slices. Brain J. Pharmacol. 109: 1085-1090.


Hemmings HC Jr and Greengard P (1986) DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein: regional, tissue, and phylogenetic distribution. J. Neurosci.6: 1469-1481.

Hemmings HC Jr, Greengard P, Tung HYL and Cohen P (1984) DARPP-32, a dopamine-regulatedneuronal phosphoprotein, is a potent inhibitor of protein phosphatase-1. Nature 310: 503-505.

Hemmings HC Jr, Nairn AC, Elliott JI and Greengard P (1990) Synthetic peptide analogs of DARPP-32 (Mr 32,000 dopamine- and cAMP-regulated phosphoprotein), an inhibitor of protein phosphatase-1.Phosphorylation, dephosphorylation, and inhibitory activity. J. Biol. Chem. 265: 20369-20376.

Hemmings HC Jr, Nairn AC and Greengard P (1984) DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated neuronal phosphoprotein. II. Comparison of the kinetics of phosphorylationof DARPP-32 and phosphatase inhibitor 1. J. Biol. Chem. 259: 14491-14497.

Hollmann M and Heinemann S (1994) Cloned glutamate receptors. Annu. Rev. Neurosci. 17: 31-108.

Ishihara H, Martin BL, Brautigan DL, Karaki H, Ozaki H, Kato Y, Fusetani N, Watabe S, HashimotoK, Uemura D et al. (1989) Calyculin A and okadaic acid: inhibitors of protein phosphatase activity.Biochem. Biophys. Res. Commun. 159: 871-877.

Kelso SR, Nelson TE and Leonard JP (1992) Protein kinase C-mediated enhancement of NMDAcurrents by metabotropic glutamate receptors in Xenopus oocytes. J. Physiol. (Lond.) 449: 705-718.


59

Kushner L, Lerma J, Zukin RS and Bennett MVL (1988) Coexpression of N-methyl-D-aspartate andphencyclidine receptors in Xenopus oocytes injected with rat brain mRNA. Proc. Natl. Acad. Sci. USA85: 3250-3254.

Lau LF and Huganir RL (1995) Differential tyrosine phosphorylation of N-methyl-D-aspartate receptorsubunits. J. Biol. Chem. 270: 20036-20041.

Laurenza A, Sutkowski EM and Seamon KB (1989) Forskolin: a specific stimulator of adenylyl cyclaseor a diterpene with multiple sites of action? Trends Pharmacol. Sci. 10: 442-447.

Leonard JP and Kelso SR (1990) Apparent desensitization of NMDA responses in Xenopus oocytesinvolves calcium-dependent chloride current. Neuron 2: 53-60.

MacDermott AB, Mayer ML, Westbrook GL, Smith SJ and Barker JL (1986) NMDA-receptoractivation increases cytoplasmic calcium concentration in cultured spinal cord neurones. Nature 321:519-522.

McGeer PL, McGeer EG, Scherer U and Singh K (1977) A glutamatergic corticostriatal path? BrainRes. 128: 369-373.

Miledi R (1982) A calcium-dependent transient outward current in Xenopus laevis oocytes. Proc. R.Soc. Lond. [Biol.] 215: 491-497.

Miledi R and Parker I (1984) Chloride current induced by injection of calcium into Xenopus oocytes. J.Physiol. (Lond.) 357: 173-183.

Moon IS, Apperson ML and Kennedy MB (1994) The major tyrosine-phosphorylated protein in thepostsynaptic density fraction is N-methyl-D-aspartate receptor subunit 2B. Proc. Natl. Acad. Sci. USA91: 3954-3958.

Nishi A, Snyder GL and Greengard P (1997) Bidirectional regulation of DARPP-32 phosphorylationby dopamine. J. Neurosci. 17: 8147-8155.

O'Connor JJ, Rowan MJ and Anwyl R (1995) Tetanically induced LTP involves a similar increase inthe AMPA and NMDA receptor components of the excitatory postsynaptic current: investigations ofthe involvement of mGlu receptors. J. Neurosci. 15: 2013-2020.


Petralia RS, Wang YX and Wenthold RJ (1994) The NMDA receptor subunits NR2A and NR2B showhistological and ultrastructural localization patterns similar to those of NR1. J. Neurosci. 14: 6102-6120.

Portera-Cailliau C, Price DL and Martin LJ (1996) N-methyl-D-aspartate receptor proteins NR2A andNR2B are differentially distributed in the developing rat central nervous system as revealed by subunit-specific antibodies. J. Neurochem. 66: 692-700.

Seamon KB and Daly JW (1986) Forskolin: its biological and chemical properties. Adv. CyclicNucleotide Protein Phosphoryl. Res. 20: 1-150.

Snyder G, Fienberg A, Dulubova I, Nairn AC and Greengard P (1996) Dopamine-mediatedphosphorylation of NMDA-R1 in the rat nucleus accumbens. Soc. Neurosci. Abstr. 22: 153.11

Stea A, Soong TW and Snutch TP (1995) Determinants of PKC-dependent modulation of a family ofneuronal calcium channels. Neuron 15: 929-940.

Chapter 3

60

Stühmer W (1992) Electrophysiological recording from Xenopus oocytes. Meth. Enzymol. 207: 319-339.

Suganuma M, Fujiki H, Furuya-Suguri H, Yoshizawa S, Yasumoto S, Kato Y, Fusetani N andSugimura T (1990) Calyculin A, an inhibitor of protein phosphatases, a potent tumor promoter on CD-1mouse skin. Cancer Res. 50: 3521-3525.

Szmigielski A and Zalewska-Kaszubska L (1991) Cooperation between D1- and D2-dopaminereceptors in the nucleus accumbens. Neuropharmacol. 30: 259-266.

Tingley WG, Roche KW, Thompson AK and Huganir RL (1993) Regulation of NMDA receptorphosphorylation by alternative splicing of the C-terminal domain. Nature 364: 70-73.

Voorn P, Jorritsma-Byham B, Van Dijk B and Buijs RM (1986) The dopaminergic innervation of theventral striatum in the rat: a light- and electron-microscopical study with antibodies against dopamine.J. Comp. Neurol. 251: 84-99.

Walaas SI, Aswad DW and Greengard P (1983) A dopamine- and cyclic AMP-regulatedphosphoprotein enriched in dopamine-innervated brain regions. Nature 301: 69-71.

Walaas SI and Greengard P (1984) DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine-innervated brain regions. I. Regional and cellulardistribution in the rat brain. J. Neurosci. 4: 84-98.

Wang LY, Salter MW and MacDonald JF (1991) Regulation of kainate receptors by cAMP-dependentprotein kinase and phosphatases. Science 253: 1132-1135.

Weiss S, Sebben M, Garcia-Sainz JA and Bockaert J (1985) D2-dopamine receptor-mediated inhibitionof cyclic AMP formation in striatal neurons in primary culture. Molec. Pharmacol. 27: 595-599.

CHAPTER 4

In vivo NMDA/dopamine interaction resulting in FOS

production in the limbic system and basal ganglia of

the mouse brain

Molecular Brain Research (2000) 75: 271–280

ABSTRACT

Glutamatergic and dopaminergic effects on molecular processes have beenextensively investigated in the basal ganglia. It has been demonstrated that NMDAand dopamine D1 and D2 receptors interact in the regulation of signal transductionand induction of transcription factors. In the present experiments, NMDA/dopamineinteractions were investigated in the normosensitive caudate nucleus, hippocampusand amygdala by monitoring FOS production. We demonstrated that NMDA and theD1 receptor agonist SKF 38393 triggered FOS levels in a distinct, non-overlappingand region-specific pattern. NMDA injected intraperitoneally (i.p.) elevated FOSlevels in all hippocampal subfields and the central amygdala, whereas SKF 38393triggered FOS production in basomedial, cortical, medial amygdala and caudatenucleus. The NMDA receptor antagonist CGS 19755 prevented NMDA- and SKF38393-triggered FOS production in all investigated brain areas. Similarly, the D1receptor antagonist SCH 23390 inhibited the effects produced by SKF 38393 orNMDA. The D2 receptor antagonist sulpiride exerted synergistic and antagonisticeffects on NMDA- and SKF 38393-triggered FOS production, in a region specificmanner. These data suggest that NMDA and dopamine receptors regulate FOSproduction within the limbic system and basal ganglia through regionallydifferentiated but interdependent actions.

Chapter 4

62

INTRODUCTION

Monitoring of the production pattern of the FOS protein has beenextensively employed to investigate the role of brain neurotransmitter systems inrapid gene responses (Morgan and Curran, 1991). The role of dopamine receptors inthe regulation of FOS production has been investigated in detail in the basal gangliaafter systemic injections of dopaminergic agonists in vivo (Moine et al., 1997;Nakazato et al., 1998; Wang and Ginty, 1996; Wirtshafter and Asin, 1994) and theirapplication to striatal cultures (Das et al., 1997; Konradi et al., 1996), on the basis oftheir implication in Parkinson’s disease, schizophrenia, compulsive disorders andaddiction (Bergmann et al., 1998; Blandini and Greenamyre, 1998; Lange et al.,1997; Starr, 1995; Wichmann and Delong, 1996). On the other hand, the effect ofNMDA receptors on FOS production has been primarily investigated inhippocampus-dependent experimental models to study convulsions, stress, noveltyand learning (Bozas et al., 1997; Radulovic et al., 1998; Shin et al., 1998). For thebasal ganglia it was consistently demonstrated that the effect of dopamine D1receptor agonists requires the activity of NMDA receptors, as demonstrated by theability of NMDA receptor antagonists to fully prevent D1 receptor-stimulated FOSproduction (Keefe and Ganguly, 1998; Konradi et al., 1996; Nakazato et al., 1998;Tomitaka et al., 1995). The modulation of NMDA receptor-currents by thedopaminergic system has also been demonstrated (Blank et al., 1997; Cepeda et al.,1993; Cepeda et al., 1998; Colwell and Levine, 1995; Fraser and Mac Vicar, 1994),suggesting an interdependence of both systems. However, the impact of dopaminereceptors on NMDA receptor-mediated gene responses has not been investigated sofar.

In addition, little is known on dopamine/NMDA interactions in brainregions other than the basal ganglia, although the joint activity of NMDA anddopamine receptors has been implicated in a variety of behaviors, not only restrictedto basal ganglia function. In particular, neuronal processes underlying learning andmemory of behavioral tasks mediated by limbic structures are sensitive todopaminergic and NMDA-dependent modulation (Mele et al., 1995; Roullet andMele, 1996; White et al., 1995). Numerous areas within the limbic system, includingthe hippocampus and amygdala, contain NMDA and dopamine receptors (Huang etal., 1992; Huang and Kandel, 1996; Laurie et al., 1995) which could modulate short-and long-term responses to environmental stimuli.

The objective of the present study was to evaluate the pattern of FOSproduction in the basal ganglia and limbic system in response to both dopamine andNMDA receptor stimulation and investigate the role of NMDA receptors indopamine-induced FOS production as well as the role of dopamine receptors inNMDA-induced FOS production.

NMDA/dopamine interactions in the mouse brain

63


AnimalsEight-week-old male inbred C57BL/6J mice, obtained from Harlan-Winkelmann (Borchen,

Germany) were used in the experiments. The mice were housed according to the recommendations ofthe Society for Laboratory Animal Science (Germany). Standard pelleted diet and water were offeredad libitum. Mice were kept individually in standard Macrolon cages in mouse "hotels" (consisting ofshelves located within a wooden enclosure providing olfactory, acoustic, and optic isolation from theexperimental room) 5 days before the beginning of the experiments. In the mouse hotels, thetemperature was maintained at 22 ± 1°C the humidity at 55 ± 10% and the light-dark cycle at 12 hr (7a.m. - 7 p.m.).

TreatmentsIn the first series of experiments, mice were intraperitoneally (i.p.) injected with different

doses of NMDA (15, 30 and 60 mg/kg) and the dopamine D1 receptor agonist (± ) SKF 38393hydrobromide (SKF 38393, 3.75 and 7.5 mg/kg) in order to establish a minimal dose of each drugrequired to trigger FOS production. In subsequent experiments, mice were injected i.p. with 30 mg/kgof NMDA or 7.5 mg/kg of SKF 38393. Combined treatment with agonists was performed by i.p.injection of mice with 30 mg/kg of NMDA followed 5 min later by i.p. injection of the D1 receptoragonist SKF 38393 or a reversed sequence of both agonists. Antagonist studies were performed by i.p.injection of the NMDA receptor antagonist CGS 19755 (10 mg/kg), D1 receptor antagonist SCH 23390(1 mg/kg) and D2 receptor antagonist sulpiride (50 mg/kg) alone, or 30 min before i.p. injection ofNMDA or SKF 38393. All agonists and antagonists were dissolved in physiological NaCl (saline)solution and the pH of the solutions was adjusted to 7.4. Each mouse received 100 µl of drug solution.Control mice were injected i.p. with 100 µl of saline. NMDA was purchased from Sigma (Deisenhofen,Germany) and all other chemicals from RBI (Köln, Germany). All efforts were taken to minimizeanimal suffering during the experiments.

FOS ImmunohistochemistryOne hour after injection of agonists or one and a half hours after injection of antagonists, four

to seven mice of each group were anesthetized with ketamine (130 mg/g b.w.) and xylasine (13 mg/gb.w.) in saline (0.01ml/g, i.p) and then transcardially perfused with ice cold PBS, followed by 4%paraformaldehyde in phosphate buffer (pH 7.4, 150 ml/mouse). Brains were postfixed for 48 hr in thesame fixative and then immersed for 24 hr each in 10%, 20% and 30% sucrose in PBS. After the tissuewas frozen with liquid nitrogen, 50-µm thick coronal sections were cut on the cryostat.Immunohistochemical staining of free-floating sections was performed as described previously(Radulovic et al., 1998). Briefly, FOS was detected with a rabbit anti- FOS antibody (OncogeneScience, 1:20,000 dilution) for 48 hr at +4°C. Subsequently, the sections were washed and incubated atroom temperature with biotinylated goat anti-rabbit antibody followed by the ABC complex (VectorABC kit). For visualization, DAB was used as chromogen (Sigma fast tablet set). The sections weremounted, dehydrated and coverslipped with Eukitt. The specificity of immunostaining was confirmedon sections that were incubated with FOS antibody preabsorbed overnight at +4°C with appropriatesynthetic antigenic peptide in 10-fold excess over the amount of antibody (Oncogene Science).

Quantification and Data AnalysisFOS-positive cells were counted in the hippocampus, amygdala and caudate nucleus with a

Macintosh-based image analysis system (NIH Image), as described previously (Radulovic et al., 1998).Counting was performed at a magnification x 100, in one field per area encompassing the entire brainregion included in quantification. An area of the same shape and size per brain region was used foreach mouse. The same light and threshold conditions were employed for all sections except for thedentate gyrus, where a higher threshold was used due to the high density of stained cells in this area.Nuclei were counted individually and expressed as number of FOS-positive nuclei per 0.1 mm2. Theanteroposterior (AP) coordinates relative to bregma of the areas (Franklin and Paxinos, 1997) includedfor detailed analysis were: AP -1.22, medial nucleus of the amygdala; AP -1.34 central, cortical,basomedial and basolateral nucleus of the amygdala, hippocampal CA1, CA2, CA3 regions and dentategyrus; AP + 0.98, dorsomedial, dorsolateral, ventromedial and ventrolateral parts of the caudatenucleus. Statistical analysis of behavioral and immunohistochemical data were performed by ANOVA

Chapter 4

64

followed by the Bonferonni-Dunn test for post-hoc comparisons. The results are presented as mean ±SEM.

Figure 1. FOS levels in the hippocampus (a), amygdala (b) and caudate nucleus (c)of mice injected i.p. with saline (A). Schematic representation of the analyzed brainareas (B). Scale bar = 200 µm.


65

RESULTS

In all brain areas subjected to quantitative analysis of FOS-positive nuclei,saline injection resulted in low to undetectable FOS levels (Fig. 1). Dose-responsestudies with NMDA and SKF 38393 were carried out with the aim to establish thelowest dose of each drug producing significant FOS production in the hippocampus,amygdala and caudate nucleus. This effect was achieved with 30 mg/kg of NMDAand 7.5 mg/kg of SKF 38393 (Figs. 2 and 3).

Figure 2. Effects of NMDA (A), SKF 38393 and sulpiride (B) on FOS production inthe hippocampus, amygdala and caudate nucleus. No. of mice per group was four toseven. Statistically significant differences: *p < 0.05 or greater vs. saline, ap < 0.05or greater vs. NMDA 15 mg/kg, bp < 0.05 or greater vs. SKF 38393 3.75 mg/kg.Abbreviations: CA1, CA2, CA3, hippocampal subfields; DG, dentate gyrus; BLA,basolateral amygdala; BMA, basomedial amygdala; CeA, central amygdala; CoA,cortical amygdala; MeA, medial amygdala; DLC, dorsolateral caudate nucleus;DMC, dorsomedial caudate nucleus; VLC, ventrolateral caudate nucleus; VMC,ventromedial caudate nucleus.

salineSKF 38393 3.75 mg/kgSKF 38393 7.5 mg/kg

CA1 CA2 CA3 DG050

100150200250300

CeA BLA BMA CoA MeA0

255075

100125150175200

0102030405060

DMC DLC VMC VLC DMC DLC VMC VLC

0255075

100125150175200

CA1 CA2 CA3 DG050

100150200250300

salineNMDA 15 mg/kgNMDA 30 mg/kg

No.

of

FOS+

nuc

lei/1

05 µ

m2

0102030405060

*,a

*

*,a

*,a*,a

**,b

*,b

*,b*,b

*,b

*,b*,b

*,b

A B

*

*

Sulpiride 50 mg/kg

CeA BLA BMA CoA MeA

Chapter 4

66

Figure 3. Effect of combined treatment of mice with NMDA and SKF 38393 on FOSproduction in the hippocampus, amygdala and caudate nucleus. No. of mice pergroup was four to six. Statistically significant differences: *p < 0.05 or greater vs.SKF 38393.

CA1 CA2 CA3 DG0

50

100

150

200

250

300

saline

NMDA 30 mg/kg

SKF 38393 7.5 mg/kg

NMDA 30 mg/kg + SKF 38393 7.5 mg/kg

0

25

50

75

100

125

150

175

200

DMC DLC VMC VLC0

10

20

30

40

50

No.

of

FOS+

nuc

lei/1

05 µ

m2

60

* *

**

*

**

* *

* *

SKF 38393 7.5 mg/kg+ NMDA 30 mg/kg

**

CeA BLA BMA CoA MeA


67

The dose of 60 mg/kg of NMDA was excluded on the basis of ourobservations that it produced convulsions. The dose of 15 mg/kg of NMDA was notsufficient to trigger FOS production, although it produced similar behavioralchanges (decreased activity) as the dose 30 mg/kg (data not shown). Interestingly,significant regional differences were observed in response to NMDA and SKF.Thus, i.p. injection of 30 mg/kg of NMDA strongly increased FOS levels in allhippocampal subfields, F (2,14) = 20.1, p < 0.001, for CA1, F (2,14) = 6.4 , p <0.05 for CA2, F (2,14) = 8.4, p < 0.01, for CA3 and F (2,14) = 54.8, p < 0.001, thedentate gyrus, and the central amygdala, F (2,14) = 10.8, p < 0.01, but not in theother nuclei of the amygdala and the striatum (Figs. 2 and 4). In contrast, i.p.injection of 7.5 mg/kg of SKF 38393 resulted in significant FOS productionthroughout the caudate nucleus, F (2,13) = 31.5, p < 0.001, as well as in thebasomedial, F (2,13) = 4.3, p < 0.05, cortical, F (2,13) = 11.05, p < 0.01, and medialamygdala, F (2,13) = 5.6, p < 0.05, but not in the CA1, CA2 and CA3 subfields ofthe hippocampus (Figs. 2 and 5). Both drugs induced FOS production only in thedentate gyrus, however, the effect of NMDA was significantly stronger than the oneof SKF 38393, t (1,8) = 4.3, p < 0.01. In the basolateral amygdala, neither NMDAnor SKF 38393 elicited FOS production that was different from the FOS productionafter saline injection.

In order to determine the receptor types involved in the NMDA- and SKF38393-mediated FOS production, NMDA, D1 and D2 receptor antagonists wereused in additional experiments. Injection of the NMDA receptor antagonist CGS19755 and the D1 receptor antagonist SCH 23390 did not produce significant FOSlevels in any of the tested forebrain areas (data not shown). The D2 receptor

Chapter 4

68

antagonist sulpiride increased the number of FOS-positive nuclei in the dorsomedial,F (6.587, p < 0.01, and ventromedial, F (3,15) = 5.91, p < 0.01, areas of the caudatenucleus (Fig. 2), but not in the hippocampus and amygdala (Fig. 2).

Combined treatment of mice with 30 mg/kg of NMDA followed by 7.5mg/kg of SKF 38393 was performed to investigate whether the drugs exhibitadditive effects on FOS production in the hippocampus, amygdala and caudatenucleus. The results obtained by FOS immunohistochemical analysis did not provideany evidence for additive interactions between NMDA and SKF 38393 (Fig. 3).FOS levels in the hippocampus and central amygdala, elicited by coadministrationof NMDA and SKF 38393 were similar to the ones induced by NMDA alone.Moreover, in the medial, F (3, 19) = 4.7, p < 0.05, and cortical, F (3, 19) = 7.3, p <0.01, nuclei of the amygdala and in the caudate nucleus, F (3, 19) = 11.3, p < 0.01,FOS levels were significantly lower after coadministration of NMDA and SKF38393 from the levels induced by SKF 38393 alone. A reversed sequence of agonistadministration (SKF 38393 preceding NMDA) also failed to show additive effectson FOS production (Fig. 3), in fact, a tendency for a lower FOS response wasobserved than after SKF 38393 injection, but this difference was not statisticallysignificant. A possibility for an additive effect was also excluded bycoadministration of NMDA (30 mg/kg) and a lower dose (3.75 mg/kg) of SKF38393 (data not shown).

NMDA-triggered FOS production in the hippocampus and central amygdalawas completely prevented by systemic pretreatment of mice with the NMDAreceptor antagonist CGS 19755 (Fig. 6a and b). In the hippocampus and centralamygdala the number of FOS-positive nuclei found after NMDA injection was


69

significantly reduced by pretreatment with the D1 receptor antagonist SCH 23390(Fig. 6a and b). In contrast, the D2 receptor antagonist sulpiride did not affect FOSproduction in the hippocampus and central amygdala.. In the dorsolateral andventrolateral areas of the caudate nucleus (Fig. 6c), FOS levels induced by sulpirideand NMDA were significantly higher than the ones induced by sulpiride alone [t(1,8) = 2.982, p < 0.05, for dorsolateral striatum and t (1,8) = 2.688, p < 0.05, forventrolateral striatum].

Figure 6. Effect of NMDA and dopamine receptor antagonists on NMDA-triggeredFOS production in the hippocampus (A), amygdala (B) and striatum (C). Mice werepretreated with saline, CGS 19755, SCH 23390, or sulpiride. Scale bar = 200 µm.No. of mice per group was four to six. Statistically significant differences: *p < 0.05or greater vs. saline, ap < 0.05 or greater vs. NMDA.

A

CA1 CA2 CA3 DG0

50

100

150

200

250

300

** * * *

*

*

*

a

aa

a

aaaa

No.

of

FOS+

nuc

lei/1

05 µ

m2

B

No.

of

FOS+

nuc

lei/1

05 µ

m2

0255075

100125150175200

*,a*

*

a

CeA BLA BMA CoA MeAC

No.

of

FOS+

nuc

lei/1

05 µ

m2

DMC DLC VMC VLC0

102030

4050

60 *,a

*,a

*,a

*,a

salineNMDA 30 mg/kgCGS-19755 10 mg/kg +NMDA 30 mg/kg

SCH 23390 1 mg/kg +NMDA 30 mg/kgSulpiride 50 mg/kg +NMDA 30 mg/kg

Chapter 4

70

SKF 38393-triggered FOS production was fully antagonized by both D1 andNMDA receptor antagonists in all examined brain areas (Fig. 7a-c). The effect of theD2 receptor antagonist sulpiride, however, showed marked regional differences.Sulpiride completely prevented SKF 38393-induced FOS production in the dentategyrus and amygdala (Fig. 7a and b). In contrast, co-administration of sulpiride andSKF 38393 resulted in significantly increased FOS production in the dorsomedialand ventromedial striatum (Fig. 7c).

Figure 7. Effect of NMDA and dopamine receptor antagonists on SKF 38393-triggered FOS production in the hippocampus (A), amygdala (B) and striatum (C).Mice were pretreated with saline, CGS 19755, SCH 23390, or sulpiride. No. of miceper group was four to six. Statistically significant differences: *p < 0.05 or greatervs. saline, ap < 0.05 or greater vs. SKF 38393.

A

B

C

saline

CGS-19755 10 mg/kg +

SCH 23390 1 mg/kg +

Sulpiride 50 mg/kg +SKF 38393 7.5 mg/kg

SKF 38393 7.5 mg/kg

SKF 38393 7.5 mg/kg

SKF 38393 7.5 mg/kg

No.

of

FOS+

nuc

lei/1

05 µ

m2

CA1 CA2 CA3 DG0

50

100

150

200

250

300

*

aa

a

No.

of

FOS+

nuc

lei/1

05 µ

m2

0255075

100125150175200

**

aaa

a

aaa

a

a

*

No.

of

FOS+

nuc

lei/1

05 µ

m2

DMC DLC VMC VLC0

10

20

30

40

50

60

**

*

*

*,a*,a

a aaaaaaa

*

*

CeA BLA BMA CoA MeA


71

DISCUSSION

Significant regional differences in the coupling of D1 and NMDA receptorstimulation with FOS production are demonstrated in this study. In most of theexamined forebrain areas only one of the employed agonists (for D1 receptors orNMDA receptors) was able, in the required presence of endogenous agonists for theother receptor, to induce FOS production. Thus, one receptor played a triggering, theother a permissive role. The requirement of both receptors and the assignment oftheir triggering or permissive role was demonstrated in experiments with receptor-specific antagonists.

In agreement with previous studies, the D1 receptor agonist was found toelicit FOS production in the striatum (Lange et al., 1997; Mons and Cooper, 1995;Ruskin and Marshall, 1994; Sgamboto et al., 1997). Full D1 receptor agonists havebeen shown to induce robust formation of c-fos mRNA or to increase production ofthe FOS protein in the striatum of intact rats (Le Moine et al., 1997; Nakazato et al.,1998; Wang and Ginty, 1996; Wirtshafter and Asin, 1994), whereas the partialdopamine D1 receptor agonist SKF 38393 triggered FOS only in the supersensitivestriatum (Cole et al., 1992; Robertson et al., 1992). In the present work, the recentlydeveloped partial D1 receptor agonist SKF-38393 hydrobromide, which crossesefficiently the blood brain barrier (Murray and Waddington, 1989), elicitedsubstantial FOS production in the normosensitive striatum. Accordingly, we foundthat i.p. injection of SKF 38393 hydrobromide, but not SKF-38393 hydrochlorideincreased FOS production in the brain (data not shown). Striatal FOS production inresponse to D1 receptor stimulation was previously found exclusively in neuronsexpressing the protein phosphatase inhibitor DARPP-32 (dopamine and cAMP-regulated phosphoprotein) (Beretta et al., 1992) which mediates the potentiation ofNMDA currents by activation of the protein kinase A pathway (Blank et al., 1997).It is possible that this interaction contributed to the permissive role of NMDAreceptors in the D1 receptor-triggered FOS response, in view of the particularly highamounts of DARPP-32 in areas such as caudate nucleus and amygdala (Ouimet etal., 1984). Accordingly, mice lacking DARPP-32 respond with significantly lowerstriatal FOS production to the D1/D2 receptor agonist amphetamine than wild-typecontrols (Fienberg et al., 1998). In addition, the observed interactions could be alsomediated by intracellular mechanisms encompassing the protein kinase C pathwayleading to FOS production upon stimulation of D1 receptors (Simpson and Morris,1995).

The mechanisms of a triggering or permissive action of D1 and NMDAreceptors on FOS production in a region-specific manner are unknown at this time.Interestingly, in striatal cultures, both NMDA and D1 receptor agonists are capableto trigger c-fos mRNA and FOS production (Conderelli et al., 1994; Konradi et al.,1994; Lerea and Mac Namara, 1993; Liu et al., 1995). In contrast to in vitroexperiments, in vivo administration of NMDA alone did not trigger any FOSproduction in the striatum. This finding was in full agreement with the inability ofNMDA to induce FOS after intrastriatal injection (Ciani et al., 1994). However,blockade of NMDA receptors prior to D1 receptor stimulation prevented D1receptor-mediated FOS production both in vivo and in vitro. Application of NMDAantagonists in striatal cultures results in decreased Ca2+ influx, pCREB levels, and

Chapter 4

72

FOS production (Das et al., 1997; Konradi et al., 1996). These findings demonstratethat both signalling cascades are functional in striatal neurons and converge mostprobably at pCREB which might regulate c-fos expression through the calcium andcAMP response element of the fos gene promoter.

It appears, that the regulation of FOS production in vivo occurs undercomplex circuitry mechanisms, probably by reciprocal interactions among brainstructures through axonal projections. It is known, for example, that electricalstimulation of the sensorimotor, auditory and limbic cortex elicits FOS production inthe striatum in areas corresponding to the primary projection site of the stimulatedcortical cells (Sgambato et al., 1997). It is also possible that in vivo the Mg2+ blockof NMDA receptors prevented FOS production, since striatal neurons are morehyperpolarized than their limbic counterparts (Hernandez-Lopez et al., 1997). Inaddition, the NMDA receptor function was found to be tonically inhibited by D2receptors in the striatum. Evidence supporting these observations was found in micelacking the D2 receptor. Tetanic stimulation, which induces dopamine receptor-dependent long-term depression (LTD) in wild-type mice (Calabresi et al., 1992),results in NMDA receptor-dependent long-term potentiation (LTP) in D2R-deficientmice (Calabresi et al., 1997). D2 receptors also inhibit glutamate release (Yamamotoand Davy, 1992), and its relief by sulpiride could additionally account for FOSproduction observed after administration of sulpiride alone. Thus, it seems likelythat blockade of D2 receptors facilitates FOS production by NMDA in the lateralcaudate nucleus, as observed in the present study, although it is also possible that theNMDA treatment modulated the response to sulpiride.

The FOS production pattern induced by sulpiride alone was consistent withprevious reports demonstrating, that typical neuroleptics, such as haloperidol, induceFOS production in the dorsolateral/lateral striatum (Semba et al., 1999), whereasatypical neuroleptics, such as sulpiride, elevate FOS levels in the medial striatum(Robertson et al., 1994). It was also found that D2 antagonists induce c-fos mRNAin the rat caudate putamen and nucleus accumbens and enhance c-fos mRNAformation in response to selective D1 agonists (Dragunow et al., 1990).Accordingly, it is proposed that D2 receptors, which are negatively coupled to thecAMP pathway (Stoof and Kebabian, 1981) tonically inhibit FOS production inresponse to D1 agonists. Our results demonstrated that simultaneous blockade of D2receptors and stimulation of D1 receptors enhanced FOS production in the medialcaudate nucleus. Although D2 receptors may exert a tonic inhibition of D1 receptorfunction in this brain region, it is equally likely that the observed increase of FOSproduction was due to additive effects of both treatments. Interestingly, co-activation of D1 and D2 receptors produced by exogenous administration ofnonselective dopamine receptor agonists, such as the psychostimulants amphetamineand cocaine, or combined treatment with selective D1 and D2 agonistssynergistically activates FOS production in a characteristic patchy pattern. Underthose experimental conditions, D2 antagonists were found to prevent c-fos mRNAexpression (Ruskin and Marshall, 1994). These findings suggest that tonic andphasic stimulation of D2 receptors may differentially affect the activity of D1receptors.

Combined treatment of mice with NMDA followed by SKF 38393 or areversed sequence of these agonists did not show evidence for additive effects. In


73

fact, in the striatum and some nuclei of the amygdala FOS levels were significantlylower than FOS levels after the injection of the D1 receptor agonist alone. Thisfinding seems paradoxical in view of the observation that NMDA receptor activitywas necessary for D1 receptor-triggered FOS production in these areas. However,Ca2+ influx may also exert multiple inhibitory effects on the D1 receptor-mediatedsignalling pathway. It was found that Ca2+ inactivates phosphorylated DARPP-32(Halpain et al., 1990). Ca2+ also inhibits the adenylyl cyclase V, which is present athigh levels and almost exclusively in the striatum (Glatt and Syder, 1993; Mons andCooper, 1995) and reduces dopamine-induced elevation of cAMP (Schinelli et al.,1994). Thus, synergistic or antagonistic effects between D1 and NMDA receptors onFOS production observed in the present study may be due to different interactionsbetween the Ca2+ and adenylyl cyclase signalling pathways.

As in the striatum, D1 receptor stimulation triggered NMDA receptor-dependent FOS responses in the basomedial, cortical and medial amygdala, whereasNMDA was ineffective. Although some of the mechanisms underlying theD1/NMDA receptor interactions may be shared by the amygdala and striatum,existing data demonstrate that the dopaminergic signalling markedly differs betweenthese two areas. D1 receptors mediate LTD in the striatum but LTP in the amygdala(Huang and Kandel, 1996). In contrast to the striatum, where D1 receptors appear tobe coupled to adenylyl cyclase (Glatt and Syder, 1993), the amygdaloid D1receptors were found to be positively coupled to the phosphoinositol pathway(Undie and Friedman, 1990). The inefficiency of NMDA to trigger FOS productionin the amygdala suggested that the NMDA system may be also inhibited as assumedfor the striatum. However, our results excluded the involvement of D2 receptors.Thus, it is remarkable that both brain areas exhibited identical FOS responses to D1and NMDA receptor agonists despite the described regional differences in receptorcoupling to signal transduction.

In the hippocampus and central amygdala, the triggering and permissiveroles of D1 and NMDA receptors were reversed. The D1 receptor agonist was notable to trigger FOS production, but played a permissive role, as demonstrated by theability of the D1 receptor antagonist to reduce significantly NMDA-triggered FOSproduction. It is well established that the hippocampal regulation of the fos geneexpression requires the activation of several regulatory elements within its promoter,which may only be achieved by multiple intracellular signalling pathways(Robertson et al., 1995). In contrast to NMDA receptor stimulation, which couldactivate all necessary elements (c-Sis inducible element, serum response element,calcium and cAMP response element) of the fos gene promoter (Ghosh et al., 1994),stimulation of the D1 receptor, which is mainly coupled to the cAMP signallingpathway, did not appear to be sufficient. However, the cAMP pathway may beimportant for the permissive role of D1 receptors.

Another type of dopamine/NMDA interaction was observed in the dentategyrus, which allowed for both D1 and NMDA receptors to trigger and permit FOSproduction. Furthermore, the D2 receptor antagonist fully prevented D1 receptor-triggered FOS production indicating a D1/D2 receptor synergism. A molecular basisfor D1/D2 synergistic interactions in the dentate gyrus, as well as some amygdalarnuclei, may be the D2 receptor-mediated release of arachidonic acid (Di Marzo et

Chapter 4

74

al., 1992; Piomelli et al., 1991) which was found to be required for FOS productionin distinct brain areas (Lerea et al., 1997).

The observed, highly differentiated and reciprocal dopamine and NMDAinteractions in the regulation of the FOS response may enable the brain to integratedopaminergic and glutamatergic signalling in a region-specific manner in vivo.Thus, the regional specificity of the roles of dopamine and NMDA receptors in thecontrol of FOS production may have broad implications for the neuronal plasticity,which occurs in the basal ganglia and the limbic system.

REFERENCES

Beretta S, Robertson HA and Graybiel AM (1992) Dopamine and glutamate agonists stimulate neuron-specific expression of Fos-like protein in the striatum. J. Neurophysiol. 68: 767-777.

Bergman H, Feingold A, Nini A, Raz A, Slovin H, Abeles M and Vaadia E (1998) Physiologicalaspects of information processing in the basal ganglia of normal and parkinsonian primates. TrendsNeurosci. 21: 32-38.

Blandini F and Greenamyre JT (1998) Prospects of glutamate antagonists in the therapy of Parkinson'sdisease. Fund. & Clin. Pharmacol. 12: 4-12.

Blank T, Nijholt I, Teichert U, Kuegler H, Behrsing H, Fienberg A, Greengard P and Spiess J (1997)The phosphoprotein DARPP-32 mediates cAMP-dependent potentiation of striatal N-methyl-D-aspartate responses. Proc. Natl. Acad. Sci. USA 94: 14859-14864.

Bozas E, Tritos N, Phillipidis H and Stylianopoulou F (1997) At least three neurotransmitter systemsmediate a stress-induced increase in c-fos mRNA in different rat brain areas. Cell & Mol. Neurobiol.17: 157-169.

Calabresi P, Maj R, Pisani A, Mercuri NB and Bernardi G (1992) Long-term synaptic depression in thestriatum: Physiological and pharmacological characterization. J. Neurosci. 12: 4224-4233.

Calabresi P, Saiardi A, Pisani A, Baik JH, Centonze D, Mercuri NB, Bernardi G and Borrelli E (1997)Abnormal synaptic plasticity in the striatum of mice lacking dopamine D2 receptors. J. Neurosci. 17:4536-4544.

Cepeda C, Buchwald NA and Levine MS (1993) Neuromodulatory actions of dopamine in theneostriatum are dependent upon the excitatory amino acid receptor sybtypes activated. Proc. Natl.Acad. Sci. USA 90: 9576-9580.

Cepeda C, Colwell CS, Itri JN, Chandler SH and Levine MS (1998) Dopaminergic modulation ofNMDA-induced whole cell currents in neostriatal neurons in slices: Contribution of calciumconductances. J. Neurophysiol. 79: 82-94.

Ciani E, Guarnieri T and Contestabile A (1994) Fos protein induction, neuropathology, andpharmacological protection after excitotoxic brain insult. Exp. Brain Res. 98: 421-430.

Cole AJ, Bhat RV, Patt C, Worley PF and Baraban JM (1992) D1 dopamine receptor activation ofmultiple transcription factor genes in rat striatum. J. Neurochem. 58: 1420-1426.



75

Condorelli DF, Albani PD, Amico C, Lukasaiuk K, Kaczmarek L and Giuffrida-Stella AM (1994)Glutamate receptor-driven activation of transcription factors in primary neuronal culture. Neurochem.Res. 19: 489-499.

Das S, Grunert M, Williams L and Vincent SR (1997) NMDA and D1 receptors regulate thephosphorylation of CREB and the induction of c-fos in striatal neurons in primary culture. Synapse 25:227-233.

Di Marzo V and Piomelli D (1992) Participation of prostaglandin E2 in dopamine D2 receptor-dependent potentiation of arachidonic acid release. J. Neurochem. 59: 379-382.

Dragunow M, Robertson GS, Faull RLM, Robertson HS and Jansen K (1990) D2 dopamine receptoragonist induce Fos and related proteins in rat striatal neurons. Neuroscience 37: 287-294.

Fienberg AA, Hiroi N, Mermelstein PG, Song W-J, Snyder GL, Nishi A, Cheramy A, O’Callaghan JP,Miller DB, Cole DG, Corbett R, Haile CN, Cooper DC, Onn SP, Grace AA, Ouimet CC, White FJ,Hyman SE, Surmeier DJ, Girault JA, Nestler EJ and Greengard P (1998) DARPP-32: Regulator of theefficacy of dopaminergic neurotransmittion. Science 281: 838-842.

Franklin KBJ and Paxinos G (1997) In: The mouse brain in stereotaxic coordinates. San Diego:Academic Press.

Fraser DD and MacVicar BA (1994) Dopamine modulates NMDA-activated currents in striatal andcortical neurons. Soc. Neurosci. Abstr. 20: 481.

Ghosh A, Ginty DD, Bading H and Greenberg ME (1994) Calcium regulation of gene expression inneural cells. J. Neurobiol. 25: 294-303.

Glatt CE and Syder SH (1993) Cloning and expression of an adenylyl cyclase localized to the corpusstriatum. Nature 367: 536-361.

Halpain S, Girault JA and Greengard P (1990) Activation of NMDA receptors inducesdephosphorylationof DARPP-32 in rat striatal slices. Nature 343: 369-372.

Hernandez-Lopez S, Bargas J, Surmeier DJ, Reyes A and Galarraga E (1997) D1 receptor activationenhances evoked discharge in neostriatal medium spiny neurons by modulating an L-type Ca2+conductance. J. Neurosci. 17: 3334-3342.

Huang Q, Zhou D, Chase K, Gusella JF, Aronin N and Difiglia M (1992) Immunohistochemicallocalization of the D-1 dopamine receptor in rat brain reveals its axonal transport, pre and postsynapticlocalization and prevalence in the basal ganglia, limbic system and thalamic reticular nucleus. Proc.Natl. Acad. Sci. USA 89: 11988-11992.

Huang YY and Kandel ER (1996) D1/D5 receptors mediate a protein synthesis-dependent late phase ofLTP in the amygdala. Proc. Soc. Neurosci. Abstr. 22: 332.

Keefe KA and Ganguly A (1998) Effects of NMDA receptor antagonists on D1 dopamine receptor-mediated changes in striatal immediate early gene expression: Evidence for involvement ofpharmacologically distal NMDA receptors? Dev. Neurosci. 20: 216-228.

Konradi C, Cole RL, Heckers S and Hyman SE (1994) Amphetamine regulates gene expression in ratstriatum via transcription factor CREB. J. Neurosci. 14: 5623-5634.

Konradi C, Leveque JC and Hyman SE (1996) Amphetamine and dopamine-induced immediate earlygene expression in striatal neurons depends on postsynaptic NMDA receptors and calcium. J. Neurosci.16: 4231-4239.

Chapter 4

76

Lange KW, Kornhuber J and Riederer P (1997) Dopamine/glutamate interactions in Parkinson'sdisease. Neurosci. Biobehav. Rev. 21: 393-400.

Laurie DJ, Putzke J, Zieglgaensberger W, Seeburg PH and Toelle TR (1995) The distribution of splicevariants of the NMDAR1 subunit mRNA in adult rat brain. Mol. Brain Res. 32: 94-108.

Le Moine C, Svenningsson P, Fredholm BB and Bloch B (1997) Dopamine-adenosine interactions inthe striatum amd the globus pallidus: inhibition of striatopallidal neurons through either D2 or A2Areceptors enhances D1 receptor-mediated effects on c-fos expression. J. Neurosci. 17: 8038-8048.

Lerea LS and MacNamara JO (1993) Ionotropic glutamate receptor subtypes activate c-fostranscription by distinct calcium-requiring intracellular signalling pathways. Neuron 10: 31-41.

Lerea LS, Carlson NG, Simonato M, Morrow JD, Roberts JL and MacNamara JO (1997) ProstaglandinF-2 is required for NMDA receptor-mediated induction of c-fos mRNA in dentate gyrus neurons. J.Neurosci. 17: 117-124.

Liu F-C, Takahashi H, Mckay RDG and Graybiel AM (1995) Dopaminergic regulation of transcriptionfactor expression in organotypic cultures of developing striatum. J. Neurosci. 15: 2367-2384.

Mele A, Castellano C, Cestari V and Oliverio A (1995) Interaction between the NMDA competitveantagonist CPP and the dopaminergic system in one-trial inhibitory avoidance in C56BL/6 mice.Neurobiol. Learn. Mem. 63: 143-148.

Mons N and Cooper DM (1995) Adenylate cyclases: critical foci in neuronal signalling. TrendsNeurosci. 18: 536-542.

Morgan JI and Curran T (1991) Stimulus-transcription coupling in the nervous system: involvement ofthe inducible proto-oncogenes fos and jun. Annu. Rev. Neurosci. 14: 421-451.

Murray AM and Waddington JL (1989) The induction of grooming and vacuous chewing by a series ofselective D-1 dopamine receptor agonists: two directions of D-1 and D-2 interaction. Eur. J.Pharmacol. 160: 377-384.

Nakazato E, Ohno M and Watanabe S (1998) MK-801 reverses Fos expression induced by the fulldopamine D1 receptor agonist SKF-82958 in the rat striatum. Eur. J. Pharmacol. 342: 209-212.


Piomelli D, Pilon C, Giros B, Sokoloff P, Martres MP and Schwartz JC (1991) Dopamine activation ofthe arachidonic acid cascade as a basis for D-1-D-2 receptor synergism. Nature 353: 164-167.

Radulovic J, Kammermeier J and Spiess J (1998) Relationship between FOS production and classicalfear conditioning: effects of novelty, latent inhibition, and unconditioned stimulus preexposure. J.Neurosci. 18: 7452-7461.

Robertson GS, Vincent SR and Fibiger HC (1992) D1 and D2 dopamine receptors differentiallyregulate c-fos expression in striatonigral and striatopallidal neurons. Neuroscience 49: 285-296.

Robertson GS, Matsumura H and Fibiger HC (1994) Induction patterns of Fos-like immunoreactivity inthe forebrain as predictors of atypical antipsychotic activity. J. Pharmacol. Exp. Ther. 271: 1058-1066.



77

Roullet P, Mele A and Ammassari-Teule M (1996) Involvement of glutamatergic and dopaminergicsystems in the reactivity of mice to spatial and non-spatial change. Psychopharmacology 126: 55-61.

Ruskin DN and Marshall JF (1994) Amphetamine-induced and cocaine-induced fos in the rat striatumdepends on D-2 dopamine-receptor activation. Synapse 18: 233-240.

Semba J, Sakai MW, Suhara T and Akanuma N (1999) Differential effects of acute and chronictreatment with typical and atypical neuroleptics on c-fos mRNA expression in rat forebrain regionsusing non-radioactive in situ hybridization. Neurochem. Int. 34: 269-277.

Schinelli S, Paolillo M and Corona GL (1994) Modulation of dopamine-induced cAMP production inrat striatal cultures by the calcium ionophore A23187 and by phorbol-12-myristate-13-acetate. Mol.Brain Res. 21: 162-166.

Sgambato V, Abo V, Rogard M, Besson MJ and Deniau JM (1997) Effect of electrical stimulation ofthe cerebral cortex on the expression of the Fos protein in the basal ganglia. Neuroscience 81: 93-112.

Shin SW, Park JW, Suh MH, Suh SI and Choe BK (1998) Persistent expression of Fas/FasL mRNA inthe mouse hippocampus after a single NMDA injection. J. Neurochem. 71: 1773-1776.

Simpson CS and Morris BJ (1995) Induction of c-fos and zif/268 gene expression in rat striatalneurons, following stimulation of D-1-like dopamine receptors, involves protein kinase A and proteinkinase C. Neuroscience 68: 97-106.

Starr MS (1995) Glutamate/dopamine D1/D2 balance in the basal ganglia and its relevance toParkinson’s disease. Synapse 19: 264-293.

Stoof JC and Kebabian JW (1981) Opposing roles for D-1 and D-2 dopamine receptors in efflux ofcyclic AMP from rat neostriatum. Nature 294: 366-368.

Tomitaka S-I, Hashimoto K, Narita N, Minabe Y and Tamura A (1995) Amantadine induces c-fos inrat striatum: reversal with dopamine D1 and NMDA receptor antagonists. Eur. J. Pharmacol. 285: 207-211.

Undie AS and Friedman E (1990) Stimulation of a dopamine D-1 receptor enhances inositolphosphates in rat brain. J. Pharmacol. Exp.Ther. 253: 987-992.

Wang JQ and McGinty JF (1996) Scopolamine augments c-fos and zif/268 messenger RNA expressioninduced by the full D1 dopamine receptor agonist SKF-82958 in the intact rat striatum. Neuroscience72: 601-616.

White IM, Christensen JR, Flory GS, Miller DW and Rebec GV (1995) Amphetamine, cocaine, anddizocilpine enhance performance on a lever-release conditioned avoidance response task in rats.Psychopharmacology 118: 324-331.

Wichmann T and Delong MR (1996) Functional and pathophysiological models of the basal ganglia.Curr. Opin. Neurobiol. 6: 751-758.

Wirtshafter D and Asin KE (1994) Interactive effects of stimulation of D1 and D2 dopamine receptorson Fos-like immunoreactivity in the normosensitive rat striatum. Brain Res. Bull. 35: 85-91.

Yamamoto BK and Davy S (1992) Dopaminergic modulation of glutamate release in striatum asmeasured by microdialysis. J. Neurochem. 58: 1736-1742.

CHAPTER 5

In vivo CREB phosphorylation mediated by dopamine

and NMDA receptor activation in mouse hippocampus

and caudate nucleus

Brain Research Gene Expression Patterns (2002) 1: 101-106

ABSTRACT

The pattern of CREB phosphorylation was investigated in the caudate nucleus andhippocampus 10 min or 3 h after i.p. injection of dopamine or NMDA receptoragonists alone, or in combination with antagonists. Ten minutes after C57BL/6Jmice were injected with either the dopamine D1 receptor agonist SKF-38393hydrobromide or NMDA, immunoreactivity of phosphorylated CREB (pCREB) wassignificantly increased in all parts of the caudate nucleus but not in hippocampalregions. However, 3 h after the injection of SKF-38393, pCREB levels in thecaudate nucleus did not differ significantly from the pCREB levels in controlanimals, whereas pCREB levels were still elevated 3 h after NMDA injection.Except for the D1 receptor antagonist SCH-23390, which induced CREBphosphorylation in the caudate nucleus, dopamine and NMDA receptor antagonistshad little effect on pCREB levels by themselves. However, the NMDA receptorantagonist CGS-19755 injected i.p. blocked both the NMDA- and SKF-38393-induced rise of pCREB levels in the caudate nucleus. Similarly, the D1 receptorantagonist SCH-23390 inhibited the effects produced by SKF-38393 or NMDA.Interestingly, the D2 receptor antagonist sulpiride also blocked the SKF-38393-triggered rise of pCREB. The results demonstrated that NMDA and dopaminereceptors modulate pCREB levels in the caudate nucleus and suggest mutualpermissive roles for both receptors.

Chapter 5

80

In earlier experiments, using FOS as a marker, we demonstrated theinterdependent relationship of dopamine and NMDA signaling in a brain region-specific manner (Radulovic et al., 2000). The promotor region of c-fos contains abinding site for the cAMP response element-binding protein (CREB). CREBphosphorylation (pCREB) was shown to mediate c-fos expression in response toagents that increase intracellular concentrations of cAMP or Ca2+ (Ginty et al.,1992; Sheng et al., 1991). Here, we investigated the regional distribution of pCREBimmunoreactivity after activation of NMDA and dopamine receptors in intact mousebrains to evaluate the possible regional correlation with FOS production. CREBphosphorylation was detected using a specific antibody raised against the serine-133-phosphorylated form of CREB (pCREB) (Ginty et al., 1993).

Unlike FOS production, phosphorylation of CREB was detectable in lowlevels in most neurons throughout the brain of untreated animals (data not shown).We have previously demonstrated (Radulovic et al., 2000) that FOS production inthe hippocampus was highly responsive to the activation of NMDA and dopaminereceptors. Surprisingly, i.p. injection of NMDA or the dopamine D1 receptor agonistSKF-38393 did not exhibit any significant effect on the number of pCREB-positivecells in the hippocampus 10 min or 3 h after injection when compared with vehicle-injected animals (Fig. 1).

Figure 1. Effect of dopamine or NMDA receptor agonists and antagonists on pCREBproduction in hippocampus. pCREB production in the hippocampus 10 min or 3 hafter systemic injection of vehicle or SKF-38393, SCH-23390, sulpiride, NMDA orCGS-19755 as measured by optical density. Bars represent mean ± SEM. Number ofmice per group was 6-13.

SKF38393vehicle

05

10152025

10 min 3h 10 min 3h

vehicleNMDA

05

10152025

10 min

vehicleSCH23390

10 min

vehicleCGS19755

10 min

vehicleSulpiride

Hippocampus

Opt

ical

den

sity

of

pCR

EB

pos

itive

nuc

lei

NMDA and dopamine modulate pCREB levels in mouse brain

81

Previous work has identified two control regions in the c-fos promoter, thecyclic AMP response element (CRE) and the serum response element (SRE), ascalcium-responsive promoter/enhancer elements (Sheng et al., 1988). On the basis ofour data it might be speculated, that in the hippocampus NMDA and SKF-38393enhanced FOS production mainly via SRE in view of the observation that neitherNMDA nor SKF-38393 induced phosphorylation on serine-133 of CREB whichinteracts in its phosphorylated form with CRE (Sheng et al., 1991). The D1 receptorantagonist SCH-23390, the NMDA receptor antagonist CGS-19755 and the D2receptor antagonist sulpiride injected alone did not affect CREB phosphorylation inthe hippocampus 10 min after injection (Fig. 1).

In contrast, the number of pCREB-positive cells in the caudate nucleus wasstrongly increased 10 min (p < 0.001) and 3 h (p < 0.05) after injection of NMDA.SKF-38393 also caused a significant enhancement of CREB phosphorylation 10 minafter the injection (p < 0.001), whereas 3 h after injection, no significant differencein pCREB-positive cells between SKF-38393- and vehicle-injected animals wasfound (Fig. 2,3). Injection of the antagonist SCH-23390 alone significantlyenhanced the number of pCREB-positive cells in the caudate nucleus 10 min afterinjection (p < 0.05). CGS-19755 did not significantly affect CREB phosphorylation(Fig. 2). After sulpiride treatment, the level of pCREB positive cells was notsignificantly increased (p = 0.08).

Figure 2. Effect of dopamine or NMDA receptor agonists and antagonists on pCREBproduction in the caudate nucleus. pCREB production in the caudate nucleus 10 minor 3 h after systemic injection of vehicle or SKF-38393, SCH-23390, sulpiride,NMDA or CGS-19755 as measured by the number of pCREB-positive nuclei. Thenumber of nuclei was normalized to the average of the vehicle groups. Bars representmean ± SEM Number of mice per group was 6-13. Statistically significant differences:*, p < 0.05 vs. vehicle.

*

0100200300400500

10 min

vehicleSCH23390

10 min 3h

NMDAvehicle

0100200300400500

10 min 3h

vehicleSKF38393

10 min

vehicleCGS19755

10 min

vehicleSulpiride

*

*

*Caudate nucleus

Rel

ativ

e in

crea

se in

pC

RE

B p

ositi

ve n

ucle

i (%

)

Chapter 5

82

The NMDA receptor antagonist CGS-19755 prevented the NMDA- andSKF-38393-triggered rise in CREB phosphorylation (Fig. 3). This result suggeststhat the D1 receptor signaling pathway needs Ca2+ influx through NMDA channelsto increase CREB phosphorylation in the caudate nucleus. Activation of D1receptors in the striatum has been demonstrated to increase phosphorylation ofNMDA receptors by a chain of events including cAMP-dependent phosphorylationof the dopamine and cAMP-regulated phosphoprotein (DARPP-32) which, byinhibiting protein phosphatase-1, is believed to inhibit dephosphorylation of NMDAreceptors, thus enhancing NMDA receptor-mediated currents (Blank et al., 1997;Snyder et al., 1998). The subsequent increase in Ca2+ influx after D1 receptoractivation is inhibited by the NMDA receptor antagonist preventing an SKF-38393-mediated rise of pCREB levels. The D1 receptor antagonist SCH-23390 blocked theincreased pCREB levels after SKF-38393 and NMDA injection (Fig. 3). Sulpiride,the D2 receptor antagonist, was also able to prevent the SKF-38393-inducedincrease of pCREB levels (Fig. 3). This finding was unexpected, since sulpiride andSKF-38393 given alone induced CREB phosphorylation in the caudate nucleus.Similarly, the enhancement of CREB phosphorylation by SCH-23390 and SKF-38393 in the caudate nucleus canceled each other out when both compounds wereapplied together. Activation of D1 receptors may increase the number of pCREBpositive cells via stimulation of adenylyl cyclase in view of the finding that D1receptors are positively coupled to adenylyl cyclase in striatal neurons (Arnauld etal., 1998). This stimulation of adenylyl cyclase can be blocked by the selective D1receptor antagonist SCH-23390. On the other hand, blockade of D1 receptors bySCH-23390 alone might shift the balance of D1/D2 receptor activation byendogenous dopamine towards activation of D2 receptors. Activation of D2receptors has been shown to enhance CREB phosphorylation by activation ofprotein kinase C and Ca2+/calmodulin-dependent protein kinase requiring DARPP-32 (Yan et al., 1999). In the presence of endogenous dopamine, the blockade of D2receptors results in an increased activity of adenylyl cyclase (Jaber et al., 1996;Vallone et al., 2000) which might be responsible for the increase of CREBphosphorylation after sulpiride treatment. The observation that blockade of D2receptors prevented SKF-38393-induced increase of pCREB levels suggestscooperative mechanisms of both receptors to modulate CREB phosphorylation. Thisview is supported by data of Jung and Schmauss (Jung and Schmauss, 1999)describing that pretreatment of mice with the D2 receptor antagonist eticlopridereduced the magnitude of FOS responses to D1 agonist stimulation in striatum.Thus, maximum FOS or pCREB responses after D1 receptor stimulation mayrequire a steady-state activity of D2 and possibly D3 receptors, which can also beblocked by sulpiride (Levant, 1997).

In combination with our previous findings (Radulovic et al., 2000), thepresent results show a clear brain region-specific interaction of NMDA and thedopamine signaling system on the level of immediate early genes. Several studieshave reported an interaction of both transmission systems in learning and memory(Adriani et al., 1998), especially in the encoding of spatial information. Since CREBhas been associated with various forms of memory formation (Frank and Greenberg,1994), it may represent a molecular target involved in the modulation of learningand memory processes by glutamatergic and dopaminergic neurotransmission.


83

Figure 3. Representative photomicrographs of the caudate nucleus of agonist andantagonist-injected animals. D1, D2 and NMDA receptor antagonists block the SKF-38393- or NMDA-induced production of pCREB in caudate nucleus. Animalsreceived either agonist alone or antagonist 30 min before agonist injection. Scale bar= 100 µm.

Chapter 5

84

METHODS AND PROTOCOLS

All animal procedures are in strict accordance with the guidelines of the animalexperimentation committee of the State Government of Lower Saxony.

SubjectsEight-week-old, male, inbred C57BL/6J mice ± 25 g were obtained from Elevage Janvier (Le

Genest St Isle, France). Animals were housed individually in macrolon cages placed in anenvironmentally controlled room (22 ± 1˚C, 55 ± 10% humidity) maintained on a 12 h light/dark cycle(light on between 7 am and 7 pm). Water and standard pelleted diet were available ad libitum. Miceunderwent treatment at least one week after adaptation to the housing conditions.

TreatmentTreatment was performed essentially as described (Radulovic et al., 2000). All drugs were

injected i.p. SKF-38393 (7.5 mg/kg) was injected as D1 receptor agonist and NMDA (30 mg/kg) wasused as NMDA receptor agonist. Antagonist studies were performed with D1 receptor antagonist SCH-23390 (1 mg/kg), D2 receptor antagonist sulpiride (50 mg/kg) and NMDA receptor antagonist CGS-19755 (10 mg/kg). When both agonist and antagonist were injected i.p., the antagonist was injected 30min before i.p. injection of the agonist. All substances were dissolved in phosphate-buffered saline(PBS) to their final concentration. When necessary, the pH of the injected solution was adjusted to 7.4with NaOH or HCl. Each mouse received 100 µl of drug solution. Control animals were injected with100 µl of the appropriate vehicle. Experiments were always performed at the same time of the day.

ImmunohistochemistryAt 10 min or 3 h after the final treatment animals were deeply anaesthetized with 0.02 ml/g

b/w. avertin (tribromethanol 1 g; amylalcohol 0.81 g; millipore 71.49 g; pH 7.4). Then, they weretranscardially perfused with ice cold 0.1 M phosphate buffer (PB; pH 7.4) followed by 4%paraformaldehyde solution (in 0.1 M PB, pH 7.4). Brains were postfixed and subsequently dehydratedin 10%, 20% and 30% sucrose in 0.01 M phosphate buffered saline. Immunohistochemical staining offree-floating coronal sections (50 µm) was performed as described previously (Radulovic et al., 2000)and summarized here. After elimination of endogenous peroxidase activity and a preincubation step,sections were incubated for 48 h with rabbit anti-pCREB antibody (1:3000 dilution; New EnglandBiolabs). Subsequently, sections were incubated with biotinylated goat anti-rabbit antibody (1:200;Vector ABC kit) and with the ABC complex (Vector ABC kit) before being reacted withdiaminobenzidine (DAB). Sections were mounted on gelatin-coated slides, air-dried, dehydrated,coverslipped with Eukitt, and examined using light microscopy.

Quantification and AnalysisThe number of pCREB positive cells was counted in mouse hippocampus and caudate

nucleus with a Macintosh-based image analysis system (NIH Image), as was described previously(Radulovic et al., 1998). Counting was performed at a magnification X10. The anteroposterior (AP)coordinates relative to bregma of the areas (Franklin and Paxinos, 1997) included for detailed analysiswere: AP +0.98 caudate nucleus and AP –1.34 hippocampus. In the caudate nucleus individual nucleiwere counted and expressed as number of pCREB positive nuclei per 0.1 mm2. The density of stainedcells in the hippocampus made it impossible to measure individual cells. Therefore, computerizedmicrodensity was used to determine the intensity of staining. The data reflect optical density of thepyramidal cell layer, determined after subtraction of the mean gray value found for the stratum oriens.Statistical analysis of immunohistochemical data was performed using ANOVA and the Bonferroni-Dunn test for posthoc comparisons.


85

ACKNOWLEDGMENTS

We would like to thank Dr. Jelena Radulovic for helpful discussions andBirgit Grafelmann and Eleanor Chen who worked as students in our laboratory.

REFERENCES

Adriani W, Felici A, Sargolini F, Roullet P, Usiello A, Oliverio A and Mele A (1998) N-methyl-D-aspartate and dopamine receptor involvement in the modulation of locomotor activity and memoryprocesses. Exp. Brain Res. 123: 52-59.

Arnauld E, Arsaut J and Demotes-Mainard J (1998) Conditional coupling of striatal dopamine D1receptor to transcription factors: ontogenic and regional differences in CREB activation. Mol. BrainRes. 60: 127-32.

Blank T, Nijholt I, Teichert U, Kügler H, Behrsing H, Fienberg A, Greengard P and Spiess J (1997)The phosphoprotein DARPP-32 mediates cAMP-dependent potentiation of striatal N-methyl-D-aspartate responses. Proc. Natl. Acad. Sci. USA 94: 14859-14864.

Frank DA and Greenberg ME (1994) CREB: a mediator of long-term memory from mollusks tomammals. Cell 79: 5-8.


Ginty DD, Bading H and Greenberg ME (1992) Trans-synaptic regulation of gene expression. Curr.Opin. Neurobiol. 2: 312-316.

Ginty DD, Kornhauser JM, Thompson MA, Bading H, Mayo KE, Takahashi JS and Greenberg ME(1993) Regulation of CREB phosphorylation in the suprachiasmatic nucleus by light and a circadianclock. Science 9: 238-241.

Jaber M, Robinson SW, Missale C and Caron MG (1996) Dopamine receptors and brain function.Neuropharmacology 35: 1503-1519.

Jung MY and Schmauss C (1999) Decreased c-fos responses to dopamine D(1) receptor agoniststimulation in mice deficient for D(3) receptors. J. Biol. Chem. 274: 29406-29412.

Levant B (1997) The D3 dopamine receptor: neurobiology and potential clinical relevance. Pharmacol.Rev. 49: 231-252.

Radulovic J, Blank T, Nijholt I, Kammermeier J and Spiess J (2000) In vivo NMDA/dopamineinteraction resulting in FOS production in the limbic system and basal ganglia of the mouse brain. Mol.Brain Res. 75: 271-280.

Radulovic J, Kammermeier J and Spiess J (1998) Relationship between Fos production and classicalfear conditioning: effects of novelty, latent inhibition, and unconditioned stimulus preexposure. J.Neurosci. 18: 7452-7461.

Sheng M, Dougan ST, McFadden G and Greenberg ME (1988) Calcium and growth factor pathways ofc-fos transcriptional activation require distinct upstream regulatory sequences. Mol. Cell Biol. 8: 2787-2796.

Chapter 5

86

Sheng M, Thompson MA and Greenberg ME (1991) CREB. A Ca2+-regulated transcription factorphosphorylated by calmodulin-dependent kinases. Science 252: 1427-1430.

Snyder GL, Fienberg AA, Huganir RL and Greengard P (1998) A dopamine/D1 receptor/protein kinaseA/dopamine- and cAMP-regulated phosphoprotein (Mr 32 kDa)/protein phosphatase-1 pathwayregulates dephosphorylation of the NMDA receptor. J. Neurosci. 18: 10297-10303.

Vallone D, Picetti P and Borrelli E (2000) Structure and function of dopamine receptors. Neurosci.Biobehav. Rev. 24: 125-132.

Yan Z, Feng J, Fienberg AA and Greengard P (1999) D(2) dopamine receptors induce mitogen-activated protein kinase and cAMP response element-binding protein phosphorylation in neurons. Proc.Natl. Acad. Sci. USA 96: 11607-11612.

CHAPTER 6

The corticotropin-releasing factor receptor 1

antagonist CP-154,526 markedly enhances associative

learning and paired-pulse facilitation immediately

after a stressful experience

Submitted

ABSTRACT

The neuropeptide corticotropin-releasing factor (CRF) coordinates theendocrine responses to stress as a major physiological regulator of the hypothalamic-pituitary-adrenal (HPA) axis. We assessed the effect of the non-peptidiccorticotropin-releasing factor receptor 1 (CRFR1) antagonist CP-154,526 incombination with exposure to an acute stressor on hippocampus-dependent memoryand hippocampal synaptic plasticity. Mice that received intraperitoneal injection ofCP-154,526 before exposure to 1 hr immobilization revealed improved context-dependent fear conditioning when trained immediately after immobilization incomparison to animals which were stressed without CP-154,526 treatment. It wasdetermined that exposure to the stressor reduced the amount of autophosphorylatedCa2+/calmodulin-dependent protein kinase II (CaMKII) in the hippocampal CA1area. When animals were pretreated with CP-154,526, the amount of hippocampalautophosphorylated CaMKII was elevated. Electrophysiological studies in thehippocampal CA1 region of stressed animals revealed no significant effects of theCP-154,526 pretreatment on various forms of long-term potentiation (LTP) but adrastic elevation of paired-pulse facilitation (PPF). The CP-154,526-inducedenhancements in fear conditioning and PPF could be prevented by the selectiveCaMKII inhibitor KN-62. Our results demonstrated that stress in combination withCP-154,526 pretreatment increased CaMKII activity and short-term synapticplasticity in the mouse CA1 area which correlated with improved hippocampus-dependent memory.

Chapter 6

88

INTRODUCTION

Exposure to stressful stimuli is known to affect hippocampal learning andmemory. However, it appears that the duration and intensity of the stressordetermine the direction of the stress effects on learning and memory and on synapticplasticity. It was demonstrated in numerous studies that induction of hippocampalLTP by high frequency stimulation or by multiple bursts of electrical pulsesdelivered in a theta-related pattern is inhibited after restraint stress and inescapabletail shock (Foy et al., 1987; Garcia et al., 1997; Kim et al., 1996; Shors and Dryver,1994). In contrast, long-term depression (LTD) observed after low frequencystimulation is known to be facilitated by such stress (Kim et al., 1996; Xu et al.,1997). In addition, several experiments suggest that stress also impairs short-termpotentiation in the CA1 area of hippocampal slices prepared immediately after thestress episode (Garcia et al., 1997; Kim et al., 1996) as well as in vivo (Diamond etal., 1994). Possible explanations for the impaired synaptic efficacy after stress rangefrom LTP saturation produced by the stress episode (Kim et al., 1996) to elevatedlevels of corticosterone which have been shown to decrease the magnitude of LTPand favor the induction of LTD (Kerr et al., 1994).

It has been shown that the activation of the HPA axis by stress is initiated bythe action of CRF (Spiess et al., 1981; Vale et al., 1981). However, there are no dataavailable on the contribution of the CRF receptor system to the described effects ofacute stress on synaptic plasticity. In this study, we used the CRFR1 antagonist CP-154,526 (Arborelius et al., 2000; Schulz et al., 1996) to elucidate the contribution ofCRF to changes in hippocampal synaptic plasticity and the possible correlation tohippocampal learning and memory in mice immediately after acute stress.


MaterialsMale Balb/c mice were obtained from Charles River (Sultzfeld, Germany). Double guide

cannulae (C235) were from Plastics One (Roanoke, VA, U.S.A.). CP-154,526 (butyl-[2,5-dimethyl-7-(2,4,6-trimethylphenyl)-7H-pyrrolo[2,3-d]pyrimidin-4-yl]-ethylamine) was provided by Dr. EricRonken (Solvay Pharmaceuticals, Netherlands). KN-62 was from Calbiochem (La Jolla, CA, U.S.A.).Antisauvagine-30 (Ruehmann et al., 1998) and [Glu11,16]astressin (Eckart et al., 2001) were synthesizedin our laboratory as described previously. All other drugs and salts were purchased from Sigma (St.Louis, MO, U.S.A.).

AnimalsExperiments were carried out on male Balb/c mice aged 9-12 weeks with an average weight

between 20-25 g. The mice were individually housed and maintained on a 12 hr light/dark cycle (lightson at 7 a.m.) with free access to food and water. All experimental procedures were in accordance withthe European Council Directive (86/609/EEC) by permission of the Animal Section Law enforced bythe District Government of Braunschweig, Lower Saxony, Germany.

Hippocampal Slice ElectrophysiologyTransverse hippocampal slices (400 µM) were obtained on a McIllwain tissue chopper and

kept submerged (minimum of 1 hr at room temperature before recordings) in artificial cerebrospinalfluid (aCSF) solution of the following composition (in mM): 130 NaCl, 3.5 KCl, 1.25 NaH2PO4, 1.5

CP-154,526 enhances learning and PPF after an acute stress

89

MgSO4, 2 CaCl2, 24 NaHCO3, and 10 glucose. Extracellular field potentials were recorded in achamber maintained at 32°C with recording electrodes pulled from borosilicate glass and filled with 2M NaCl (3-5 mΩ). All recordings were performed using a SEC-05L amplifier (npi Electronics,Germany). To record field potentials in the CA1 pyramidal cell body layer, Schaffer collaterals werestimulated with a bipolar electrode placed on the surface of the slice. At the beginning of eachexperiment, a stimulus response curve was established by increasing the stimulus intensity andmeasuring the amplitude of the population spike. Based on the input-output function, the stimulus wasadjusted to elicit a population spike with an amplitude of half of the maximum and was fixed at thislevel throughout the experiments. Traces were stored on a computer using Pulse 7.4 software (HEKA,Lambrecht, Germany) for off-line analysis.

CannulationDouble guide cannulae were implanted with the help of a stereotactic holder during 1.2%

avertin anesthesia (0.02 ml/g, i.p.) under aseptic conditions as previously described (Radulovic et al.,1999; Stiedl et al., 2000). Each double guide cannula with inserted dummy cannula and dust cap wasfixed to the skull by dental cement. The cannulae were placed into both lateral brain ventricles,anteroposterior (AP) –0.0 mm, lateral 1 mm, depth 3 mm or directed toward both dorsal hippocampi,AP –1.5 mm, lateral 1 mm, depth 2 mm (Franklin and Paxinos, 1997). The animals were allowed torecover for 4-5 d before the experiments started. On the day of the experiment, bilateral injections wereperformed using an infusion pump (CMA/Microdialysis) at a constant rate of 0.33 µl/min (finalvolume: 0.25 µl per side). Cannula placement was verified post hoc in all mice by injection ofmethylene blue. For electrophysiological experiments double cannula placement was verified byunilateral methylene blue injection just before the decapitation procedure.

StatisticsStatistical comparisons were made by using either Student`s t -test or ANOVA. Data were

expressed as mean ± SEM. Significance was determined at the level of p ≤ 0.05.

Drug TreatmentCP-154,526 (100 µl per injection) in an acidic vehicle (HCl, final concentration 0.1 N) was

administered (20 mg/kg, i.p.) 15 min prior to immobilization. Approximately 400 µg CP-154,526 wasapplied per mouse. CP-154,526 has been shown to penetrate the blood-brain barrier and to exhibitactivity in vivo (Schulz et al., 1996). KN-62 was dissolved in DMSO to a concentration of 4 mg/ml. Forcannula injection the stock was diluted in aCSF to a final concentration of 64 ng/µl. Mifepristone andspironolactone were dissolved in 1% EtOH and aCSF. Antisauvagine-30 was prepared as 80 µg/µlstock solution in 0.01 M HOAc and diluted to the final concentration of 800 ng/µl in aCSF prior toinjection. [Glu11,16]astressin was dissolved in aCSF.

Immobilization StressAn acute immobilization stress of mice consisted of taping their limbs to a Plexiglas surface

for 1 hr (Smith et al., 1995).

Fear ConditioningThe fear conditioning experiments were performed as previously described (Stiedl et al.,

2000) using a computer-controlled fear conditioning system (TSE, 303410, Bad Homburg, Germany).Fear conditioning was performed in a Plexiglas cage (36 x 21 x 20 cm) within a constantly illuminated(12 V, 10 W halogen lamp, 100-500 lux) fear conditioning box. In the conditioning box, a high-frequency loudspeaker (Conrad, KT-25-DT, Hirschau, Germany) provided constant background noise[white noise, 68 dB sound pressure level (SPL)]. The training (conditioning) consisted of a single trial.The mouse was exposed to the conditioning context (180 sec) followed by a tone (CS, 30 sec, 10 kHz,75 dB SPL, pulsed 5 Hz). After termination of the tone, a footshock (US, 0.7 mA, 2 sec, constantcurrent) was delivered through a stainless steel grid floor. The mouse was removed from the fearconditioning box 30 sec after shock termination to avoid an aversive association with the handlingprocedure. Memory tests were performed 24 hr after fear conditioning. Contextual memory was testedin the fear conditioning box for 180 sec without CS or US presentation (with background noise).Freezing, defined as the lack of movement except for respiration and heart beat, was assessed as thebehavioral parameter of the defensive reaction of mice (Blanchard and Blanchard, 1969; Bolles and

Chapter 6

90

Riley, 1973; Fanselow and Bolles, 1979) by a time-sampling procedure every 10 s throughout thememory tests. In addition, activity-derived measures (inactivity, mean activity, and exploratory area)were recorded by a photo beam system (10 Hz detection rate) controlled by the fear conditioningsystem.

Western BlottingCA1 areas of hippocampal slices were dissected out and homogenized. The insoluble

material was removed by centrifugation at 15,000 x g for 10 min at 4°C. Protein concentrations weredetermined with the Bradford assay (BioRad, Muenchen, Germany). Equal amounts of protein for eachgroup were separated on a 10% SDS gel and transferred to an Immobilon-P membrane using a semidrytransfer apparatus. The blot was probed using an anti-active CaMKII antibody (Promega, Madison, WI)or antibody directed against total CaMKII (Chemicon, Temecula, CA) and detected with alkalinephosphatase-conjugated second antibody. Western blots were developed using the chemiluminescencemethod.

RESULTS

Effects of CP-154,526, a selective CRFR1 antagonist, on fear conditioning andCaMKII activity

When mice were trained immediately after 1 hr immobilization theconditioned fear to context was mildly, however not significantly, attenuated incomparison to controls (p > 0.05). Interestingly, when mice received intraperitoneal(i.p.) injections of the specific CRFR1 antagonist CP-154,526 15 min before thestress session the fear conditioning to context was significantly enhanced whencompared to untreated mice. The injection of CP-154,526 alone without stressingthe mice had no effect on fear conditioning (Fig. 1A). Because context-dependentfear conditioning is a hippocampus-dependent learning task (Holland and Bouton,1999; Kim and Fanselow, 1992; Phillips and LeDoux, 1992) we next determined ifintrahippocampal (i.h.) injection of CRFR antagonists immediately beforeimmobilization could mimic the CP-154,526 effect. However, i.h. administration ofeither the non-selective CRFR antagonist [Glu11,16]astressin (Eckart et al., 2001) orthe CRFR2 antagonist antisauvagine-30 (Ruehmann et al., 1998) had no effect onconditioned fear to context (Fig. 1C). Similarly, both compounds did notsignificantly improve contextual fear conditioning when givenintracerebroventricularly (i.c.v.) before immobilization (p = n.s.; data not shown).Previous studies have shown that the CRF-induced increase in plasmaadrenocorticotropic hormone is blocked by CP-154,526 in rats (Schulz et al., 1996).We expected that the subsequent corticosterone reduction could be responsible forthe observed CP-154,526 effect, and therefore, we investigated the effects ofcorticosteroid receptor antagonists on conditioned fear to context. However,administration (i.h.) of the specific mineralocorticoid receptor antagonist,spironolactone (Kim et al., 1998), or the specific glucocorticoid receptor antagonist,mifepristone (Kim et al., 1998), immediately before immobilization had no effect onfear conditioning in comparison to vehicle-treated animals (Fig. 1D).


91

Figure 1. CP-154,526 enhances fear conditioning after immobilization stress. A,Mice were injected i.p. with either CP-154,526 (20 mg/kg ) or vehicle (0,1 N HCl) 15min before the 1 hr stress session and trained immediately after stress cessation.Control mice were naive, untreated mice. B, Animals preinjected with CP-154,526 orvehicle were injected i.h. with KN-62 (32 ng per mouse) immediately before the stresssession. C, Mice were injected i.h. with either vehicle (aCSF), antisauvagine-30 (400ng per mouse) or [Glu11,16]astressin (200 ng per mouse) immediately before the stresssession. D, Mice were injected i.h. with either vehicle (1% EtOH in aCSF),mifepristone (1 ng per mouse) or spironolactone (1 ng per mouse) immediately beforethe stress session. Data represent mean ± SEM from 4 to 9 animals. Statisticallysignificant differences: * p < 0.05.

In various forms of hippocampus-dependent memory, the involvement ofthe multifunctional CaMKII has been demonstrated (Cammarota et al., 1998; Silvaet al., 1992). Therefore, with the help of Western blots, we evaluated CaMKIIactivation in the CA1 area of mice that had experienced 1 hr immobilization withand without CP-154,526 treatment. We used antibodies that selectively recognizethe CaMKII autoinhibitory domain phosphorylated at threonine 286 (Thr286). Thisphosphorylation correlates with the activation of the enzyme. Acute stress resultedin a modest reduction of activated CaMKII, whereas the CP-154,526 pretreatmentcaused a substantial activation of CaMKII compared to controls (Fig. 2A). Micereceived i.h. injection of the selective CaMKII inhibitor KN-62 after the CP-154,526

0

25

50

75

0

25

50

75

Free

zing

(%

)

0

25

50

75

0

**

*BA

DC

Free

zing

(%

)*

*

25

50

75

antis

auva

gine-3

0+str

ess

vehic

le+str

ess

vehic

le+str

ess

mifepr

iston

e+str

ess

spiro

nolac

tone+

stres

s

[Glu

11,16

] astr

essin

+stres

s

vehic

le+str

ess

CP (1 hr

)

contr

ol

CP+stres

sstr

ess

vehic

le+str

ess

CP+stres

s

vehic

le+KN-6

2+str

ess

CP+KN-62+

stres

s

Chapter 6

92

injection to further explore whether elevated hippocampal CaMKII activity might beinvolved in the observed improved fear conditioning after CP-154,526 treatmentprior to immobilization stress. Now the learning improvement was completelyprevented when compared to animals that were treated with CP-154,526 alonebefore the stress session (Fig. 1B). Similarly, KN-62 antagonized the CP-154,526effect when injected i.c.v. (p < 0.05; data not shown). Again, activated CaMKIIlevels corresponded to enhanced learning ability. CaMKII activity was reducedwhen the CP-154,526 treatment was followed by KN-62 injection. Vehicle injectionhad no effect on CaMKII activity (Fig. 2B).

Figure 2. CP-154,526 injection increases the amount of active CaMKII followingacute stress. A, Representative CaMKII Western blots of area CA1 subregions fromcontrol (non-stressed) mice, mice that were subjected to 1 hr immobilization anddecapitated immediately after end of the stress session and from mice that werepretreated with CP-154,526 (i.p.; 20 mg/kg) 15 min before immobilization. Detectionwas performed using an antibody specific for Thr286-phosphorylated CaMKII (upperlane), or an antibody recognizing total CaMKII (lower lane). B, Mice were injectedwith CP-154,526 (i.p.; 20 mg/kg) and KN-62 (i.h.; 32 ng per mouse) (left), CP-154,526 (i.p.; 20 mg/kg) and aCSF (i.h.; vehicle for KN-62) (middle) and 0,1 N HCl(i.p.; vehicle for CP-154,526) and aCSF (i.h.; vehicle for KN-62) (right) before the 1hr stress session and decapitated immediately after end of the stress session. Resultsare representative of four independent Western blots


93

Effects of immobilization and CP-154,526 treatment on short-term and long-term synaptic plasticity

We examined the synaptic transmission in acute hippocampal slicepreparation to detect possible electrophysiological correlates for the behavioralfindings. A PPF study, in which double pulses were applied to the Schaffercollaterals at intervals of 10-200 ms, showed no difference in the CA1 pyramidalcell region in slices from stressed animals in comparison to slices from controlanimals. The ratio of the second response to the first response was significantlyenhanced in slices from animals that were pretreated with CP-154,526 before theimmobilization session. This enhancement was completely abolished when the CP-154,526 treatment was combined with an i.c.v. injection of KN-62 (Fig. 3A).

We next investigated the effect of stress and CP-154,526 on long-termsynaptic transmission in the CA1 area from hippocampal slices. After stimulation ofthe Schaffer-collateral pathway by two 100 Hz (1 s) trains, LTP was notsignificantly different in slices from stressed animals compared to LTP induced inslices from animals that were pretreated with CP-154,526 before immobilization(Fig. 3B). Similarly, LTP induced by theta burst stimulation (TBS), consisting of 10x 100 Hz bursts, was not significantly different in slices from stressed animals, non-stressed controls or CP-154,526-pretreated and subsequently stressed animals (Fig.3C). In contrast to the moderate inhibitory effect of stress on LTP induced by thestandard 100 Hz tetanus protocol, we found modest facilitation of LTP after acutestress when LTP was induced by TBS, consisting of 5 x 100 Hz bursts. However,there was no significant difference in the degree of LTP between slices obtainedfrom vehicle-treated and CP-154,526-treated mice (Fig. 3D).

DISCUSSION

We showed here that pretreatment of mice with the selective non-peptidergic CRFR1 antagonist CP-154,526 enhanced context-dependent fearconditioning after immobilization stress which correlated with elevated PPF in theCA1 area of hippocampal brain slices. Both, elevated short-term synaptictransmission as well as increased conditioned fear to context were prevented byinhibition of CaMKII with KN-62. CP-154,526 treatment caused an increase inCaMKII activity after stress but had no detectable effect on various types of LTP. Asimilar observation is made in mutant mice expressing CaMKII-Asp286 which ismore active in the absence of Ca2+ than CaMKII-Thr286 (Mayford et al., 1995).The mutant mice exhibit normal LTP in response to stimulation at 100 Hz. Incontrast to our results, Mayford et al. (1995) did not observe significant differencesin PPF between these transgenic and wild-type animals whereas in our experimentsincreased PPF was prevented by inhibition of CaMKII activity. On the other hand,mice heterozygous for a αCaMKII mutation exhibit decreased PPF, highly reducedcontextual fear conditioning, but normal LTP in the CA1 region (Silva et al., 1996).PPF is a short-lasting (< 1 s) enhancement in synaptic strength that is thought to bemediated by a presynaptic mechanism (Kamiya and Zucker, 1994). Activation ofprotein kinases in presynaptic terminals, particularly CaMKII was shown to

Chapter 6

94

correlate with neurotransmitter release (Nichols et al., 1990). For example, in ratstriatum the enhanced dopamine release after repeated amphetamine treatment ismediated by increased CaMKII activity (Kantor et al., 1999).

Figure 3. A, Plot of paired-pulse facilitation (amplitude of population spike responseto second stimulus divided by amplitude of population spike response to first stimulus)in the CA1 pyramidal cell region. Hippocampal slices were prepared from nonstressed control mice (n = 16), stressed mice (immediately after 1 hr immobilizationstress, n = 15) and stressed mice which were pretreated with either CP-154,526 (i.p.;20 mg/kg; 15 min before stress, n = 13) or CP-154,526 and KN-62 (i.c.v.; 32 ng permouse ; 0 min before stress; n = 5). There were no significant effects when vehicle forKN-62 and CP-154,526 was injected alone. Control versus stressed mice: p < 0.05;control versus CP-154,526 + stressed: p < 0.0001; control versus CP-154,526 + KN-62 + stressed: p = 0.18; stressed versus CP-154,526 + stressed: p < 0.05; stressedversus CP-154,526 + KN-62 + stressed: p = 0.17; CP-154,526 + stressed versus CP-154,526 + KN-62 + stressed: p < 0.005.B, LTP induced by two 100 Hz, 1 s trains (10 s interval): 211% ± 18% (control; n =6); 136% ± 14% (stressed mice; n = 6); 145% ± 13% (CP-154,526 + stressed; n =7). Control versus CP-154,526 + stressed: p < 0.05; control versus stressed: p <0.05; stressed versus CP-154,526 + stressed: p = 0.673.C, LTP induced by theta burst stimulation (TBS) consisting of 10 x 100 Hz bursts (10diphasic pulses per burst, 200 ms interburst interval): 133% ± 14% (control; n = 5);129% ± 12% (stressed mice; n = 6); 161% ± 23% (CP-154,526 + stressed; n = 4).There were no significant differences between groups.D, LTP induced by TBS consisting of 5 x 100 Hz bursts: 105% ± 11% (control; n =5); 137% ± 18% (stressed mice; n = 6); 133% ± 13% (stressed + CP-154,526; n =8). There were no significant differences between groups. LTP values are thepercentage of prestimulation baseline measured 1 hr after stimulation (mean ± SEM;n = number of animals

0 20 40 60 800

100

200

3002x100 Hz (HFS)

Time (min)

PS A

mpl

itude

(% o

f ba

sal P

S)

0 50 100 150 200

100

200

Interpulse Interval (ms)Perc

ent F

acili

tatio

n

(PS2

/PS1

x 1

00)

CP-154,526+ stress

stress CP-154,526 +KN-62 + stress

non stressedcontrol

A B

0 20 40 60 800

100

200

300

Time (min)

10x100Hz (TBS)

PS A

mpl

itude

(% o

f ba

sal P

S)

C D

0 20 40 60 800

100

200

300 5x100Hz (TBS)

PS A

mpl

itude

(% o

f ba

sal P

S)

Time (min)


95

There are several possible target molecules to regulate neurotransmitterrelease in the hippocampal CA1 area after phosphorylation by CaMKII. The v-SNARE protein synaptotagmin may play the role of a calcium sensor in exocytosis(Verona et al., 2000) and has been demonstrated to be a substrate for CaMKII inisolated synaptic vesicles (Popoli, 1993). Synapsin I, which is found exclusively inneuronal presynaptic terminals (Nayak et al., 1996) is believed to link synapticvesicles to the presynaptic cytoskeleton and restricts the availability of those vesiclesfor release. The on-vesicle phosphorylation of synapsin I by CaMKII and thesubsequent dissociation of the protein from the vesicle membrane may represent aprompt and efficient mechanism for the modulation of neurotransmitter release andpresynaptic plasticity (Stefani et al., 1997).).The intriguing question remains howCP-154,526 enhances hippocampal CaMKII activity and hippocampus-dependentmemory in combination with immobilization. Considering the i.h. injectionexperiments it seems unlikely that CP-154,526 was acting directly in thehippocampus because neither inhibition of hippocampal CRF receptors by[Glu11,16]astressin or antisauvagine-30 nor inhibition of hippocampal corticosteroidreceptors by specific inhibitors enhanced contextual fear conditioning afterimmobilization as did CP-154,526. It has further been shown that i.h. injection ofhuman/rat CRF enhances conditioned fear to context (Radulovic et al., 1999). Inview of this observation, it is unlikely that the blockade of hippocampal CRFreceptors by CP-154,526 elicited a similar effect. Therefore it is assumed that CP-154,526 blocked CRFR1 at one or more different non-hippocampal sites which werenot reached by our local i.h. and i.c.v. injections. CP-154,526 might potentiallyinfluence hippocampal function via the amygdala which contains CRFR1 protein(Radulovic et al., 1998) and mRNA (Van Pett et al., 2000) and projects to thehippocampal CA1 area (Aggleton, 1986; Pikkarainen et al., 1999). In a recent study,it was demonstrated that amygdalar lesions block stress effects on CA1 LTP with noeffect in unstressed animals (Kim et al., 2001). Similarly, CP-154,526 was onlyeffective in combination with a stressful event but did not affect synaptic plasticityin unstressed animals. It may be speculated, that CP-154,526 reduced synaptictransmission from the amygdala to the hippocampus during immobilization whichmight have caused the changes in hippocampal CaMKII activity. In the amygdalah/rCRF increases neuronal excitability (Rainnie et al., 1992). This increase ofexcitability may enhance neuronal transmission from the amygdala to thehippocampus and subsequently induce elevated Ca2+ influx in neurons of the CA1region during exposure to immobilization. The increase of intracellular Ca2+initially increases CaMKII autophosphorylation but also produces a delayed,phosphatase-dependent decrease in the levels of autophosphorylated CaMKII(Dosemeci and Reese, 1993), an effect most probably mediated by proteinphosphatase 1 (Shen et al., 2000) or Ca2+/calmodulin-dependent protein kinasephosphatase (CaMKPase) (Ishida et al., 1998). Since h/rCRF increases neuronalexcitability in the amygdala (Rainnie et al., 1992), the blockade of CRF receptors byCP-154,526 during immobilization would be presumed to prevent the enhancementof neuronal transmission from the amygdala to the hippocampus. In the presence ofCP-154,526 a slower rise of intracellular Ca2+ in the hippocampus and thus, aslower activation of CaMKII is expected (Miller and Kennedy, 1986). BecauseCaMKPase has to be phosphorylated by CaMKII for activation, delayed activation

Chapter 6

96

of CaMKII would particularly slow down activation of CaMKPase and thesubsequent dephosphorylation of autophosphorylated Thr286 of CaMKII(Kameshita et al., 1999). A delayed activation of CaMKII and the subsequentdelayed dephosphorylation would provide a possible explanation for elevatedCaMKII activity observed in the hippocampus of CP-154,526-injected animalsfollowing immobilization.

Taken together our results indicated a central role for CRFR1 outside thehippocampus in the regulation of hippocampal CaMKII activity during a period ofacute stress with a decisive influence on short-term synaptic plasticity andhippocampus-dependent memory.

ACKNOWLEDGMENTS

We are grateful to Dr. Jeansok Kim for constructive suggestions on thismanuscript.

REFERENCES

Aggleton JP (1986) A description of the amygdalo-hippocampal interconnections in the macaquemonkey. Exp. Brain Res. 64: 515-526.

Arborelius L, Skelton KH, Thrivikraman KV, Plotsky PM, Schulz DW and Owens MJ (2000) Chronicadministration of the selective corticotropin-releasing factor 1 receptor antagonist CP-154,526:behavioral, endocrine and neurochemical effects in the rat. J. Pharmacol. Exp. Ther. 294: 588-597.

Blanchard RJ and Blanchard DC (1969) Passive and active reactions to fear-eliciting stimuli. J. Comp.Physiol. Psychol. 68: 129-135.

Bolles RC and Riley AL (1973) Freezing as an avoidance response: another look at the operant-respondent distinction. Learn. Motiv. 4: 268-275.

Cammarota M, Bernabeu R, Levi De Stein M, Izquierdo I and Medina JH (1998) Learning-specific,time-dependent increases in hippocampal Ca2+/calmodulin-dependent protein kinase II activity andAMPA GluR1 subunit immunoreactivity. Eur. J. Neurosci. 10: 2669-2676.

Diamond DM and Rose GM (1994) Stress impairs LTP and hippocampal-dependent memory. Ann.N.Y. Acad. Sci. 746: 411-414.

Dosemeci A and Reese TS (1993) Inhibition of endogenous phosphatase in a postsynaptic densityfraction allows extensive phosphorylation of the major postsynaptic density protein. J. Neurochem. 61:550-555.

Eckart K, Jahn O, Radulovic J, Tezval H, van Werven L and Spiess J (2001) A single amino acidserves as an affinity switch between the receptor and the binding protein of corticotropin-releasingfactor: Implications for the design of agonists and antagonists. Proc. Natl. Acad. Sci. USA 98: 11142-11147.


97

Fanselow MS and Bolles RC (1979) Naloxone and shock-elicited freezing in the rat. J. Comp. Physiol.Psychol. 93: 736-744.

Foy MR, Stanton ME, Levine S and Thompson RF (1987) Behavioral stress impairs long-termpotentiation in rodent hippocampus. Behav. Neur. Biol. 48: 138-149.


Garcia R, Musleh W, Tocco G, Thompson RF and Baudry M (1997) Time-dependent blockade of STPand LTP in hippocampal slices following acute stress. Neurosci. Lett. 233: 41-44.

Holland PC and Bouton ME (1999) Hippocampus and context in classical conditioning. Curr. Opin.Neurobiol. 9: 195-202.

Ishida A, Kameshita I and Fujisawa H (1998) A novel protein phosphatase that dephosphorylates andregulates Ca2+/calmodulin-dependent protein kinase II. J. Biol. Chem. 273: 1904-1910.

Kameshita I, Ishida A and Fujisawa H (1999) Phosphorylation and activation of Ca2+/calmodulin-dependent protein kinase phosphatase by Ca2+/calmodulin-dependent protein kinase II. FEBS Lett.456: 249-252.

Kamiya H and Zucker RS (1994) Residual Ca2+ and short-term synaptic plasticity. Nature 371: 603-606.

Kantor L, Hewlett GH and Gnegy ME (1999) Enhanced amphetamine- and K+-mediated dopaminerelease in rat striatum after repeated amphetamine: differential requirements for Ca2+- and calmodulin-dependent phosphorylation and synaptic vesicles. J. Neurosci. 19: 3801-3808.

Kerr DS, Huggett AM and Abraham WC (1994) Modulation of hippocampal long-term potentiationand long- term depression by corticosteroid receptor activation. Psychobiol. 22: 123-133.


Kim JJ, Foy MR and Thompson RF (1996) Behavioral stress modifies hippocampal plasticity throughNMDA receptor activation. Proc. Natl. Acad. Sci. USA 93: 4750-4753.

Kim PJ, Cole MA, Kalman BA and Spencer RL (1998) Evaluation of RU28318 and RU40555 asselective mineralocorticoid receptor and glucocorticoid receptor antagonists, respectively: receptormeasures and functional studies. J. Steroid Biochem. Mol. Biol. 67: 213-222.

Kim JJ, Lee HJ, Han JS and Packard MG (2001) Amygdala is critical for stress-induced modulation ofhippocampal long-term potentiation and learning. J. Neurosci. 21: 5222-5228.

Miller SG and Kennedy MB (1986) Regulation of brain type II Ca2+/calmodulin-dependent proteinkinase by autophosphorylation: a Ca2+-triggered molecular switch. Cell 28: 861-870.

Mayford M, Wang J, Kandel ER and O`Dell TJ (1995) CaMKII regulates the frequency-responsefunction of hippocampal synapses for the production of both LTD and LTP. Cell 81: 891-904.

Nayak AS, Moore CI and Browning MD (1996) Ca2+/calmodulin-dependent protein kinase IIphosphorylation of the presynaptic protein synapsin I is persistently increased during long-termpotentiation. Proc. Natl. Acad. Sci. USA 93: 15451-15456.

Nichols RA, Sihra TS, Czernik AJ, Nairn AC and Greengard P (1990) Calcium/calmodulin-dependentprotein kinase II increases glutamate and noradrenaline release from synaptosomes. Nature 343: 647-651.

Chapter 6

98


Pikkarainen M, Ronkko S, Savander V, Insausti R and Pitkanen A (1999) Projections from the lateral,basal, and accessory basal nuclei of the amygdala to the hippocampal formation in rat. J. Comp.Neurol. 403: 229-260.

Popoli M (1993) Synaptotagmin is endogenously phosphorylated by Ca2+/calmodulin protein kinase IIin synaptic vesicles. FEBS Lett. 317: 85-88.

Radulovic J, Ruehmann A, Liepold T and Spiess J (1999) Modulation of learning and anxiety bycorticotropin-releasing factor (CRF) and stress: differential roles of CRF receptors 1 and 2. J. Neurosci.19: 5016-5025.

Radulovic J, Sydow S and Spiess J (1998) Characterization of native corticotropin-releasing factorreceptor type 1 (CRFR1) in the rat and mouse central nervous system. J. Neurosci. Res. 54: 507-521.

Rainnie DG, Fernhout BJ and Shinnick-Gallagher P (1992) Differential actions of corticotropinreleasing factor on basolateral and central amygdaloid neurones, in vitro. J. Pharmacol. Exp. Ther. 263:846-858.

Ruehmann A, Bonk I, Lin CR, Rosenfeld MG and Spiess J (1998) Structural requirements for peptidicantagonists of the corticotropin-releasing factor receptor (CRFR): development of CRFR2beta-selectiveantisauvagine-30. Proc. Natl. Acad. Sci. USA 95: 15264-15269.

Schulz DW, Mansbach RS, Sprouse J, Braselton JP, Collins J, Corman M, Dunaiskis A, Faraci S,Schmidt AW, Seeger T, Seymour P, Tingley FD 3rd, Winston EN, Chen YL and Heym J (1996) CP-154,526: a potent and selective nonpeptide antagonist of corticotropin releasing factor receptors. Proc.Natl. Acad. Sci. USA 93: 10477-10482.

Shen K, Teruel MN, Connor JH, Shenolikar S and Meyer T (2000) Molecular memory by reversibletranslocation of calcium/calmodulin–dependent protein kinase II. Nat. Neurosci. 3: 881-886.

Shors TJ and Dryver E (1994) Effect of stress and long-term potentiation (LTP) on subsequent LTPand the theta burst response in the dentate gyrus. Brain Res. 15: 232-238.

Silva AC, Paylor R, Wehner J and Tonegawa S (1992) Impaired spatial learning in α-calcium-calmodulin-kinase II mutant mice. Science 257: 206-211.

Silva AJ, Rosahl TW, Chapman PF, Marowitz Z, Friedman E, Frankland PW, Cestari V, Cioffi D,Sudhof TC and Bourtchuladze R (1996) Impaired learning in mice with abnormal short-lived plasticity.Curr. Biol. 6: 1509-1518.

Smith MA, Makino S, Kvetnansky R and Post RM (1995) Stress and glucocorticoids affect theexpression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs in the hippocampus. J.Neurosci. 15: 1961-1970.


Stefani G, Onofri F, Valtorta F, Vaccaro P, Greengard P and Benfenati F (1997) Kinetic analysis of thephosphorylation-dependent interactions of synapsin I with rat brain synaptic vesicles. J. Physiol. 504:501-515.

Stiedl O, Birkenfeld K, Palve M and Spiess J (2000) Impairment of conditioned contextual fear ofC57BL/6J mice by intracerebral injections of the NMDA receptor antagonist APV. Behav. Brain Res.116: 157-168.


99

Verona M, Zanotti S, Schafer T, Racagni G and Popoli M (2000) Changes of synaptotagmin interactionwith t-SNARE proteins in vitro after calcium/calmodulin-dependent phosphorylation. J. Neurochem.74: 209-221.

Vale W, Spiess J, Rivier C and Rivier J (1981) Characterization of a 41-residue ovine hypothalamicpeptide that stimulates secretion of corticotropin and β–endorphin. Science 213: 1394-1397.



Xu L, Holscher C, Anwyl R and Rowan MJ (1998) Glucocorticoid receptor and protein/RNAsynthesis-dependent mechanisms underlie the control of synaptic plasticity by stress. Proc. Natl. Acad.Sci. USA 95: 3204-3208.

CHAPTER 7

Priming of long-term potentiation in mouse

hippocampus by corticotropin-releasing factor and

acute stress: implications for hippocampus-dependent

learning

Journal of Neuroscience (2002) 22: 3788-3794

ABSTRACT

In the present experiments we characterized the action of human/ratcorticotropin-releasing factor (h/rCRF) and acute stress (1 hr immobilization) onhippocampus-dependent learning and on synaptic plasticity in mouse hippocampus.We first showed that h/rCRF application and acute stress facilitated (primed) long-term potentiation of population spikes (PS-LTP) in the mouse hippocampus andenhanced context-dependent fear conditioning. Both the priming of PS-LTP and theimprovement of context-dependent fear conditioning were prevented by the CRFreceptor antagonist [Glu11,16]astressin. PS-LTP priming and improved learning werealso reduced by the protein kinase C inhibitor bisindolylmaleimide I. Acute stressinduced the activation of Ca2+/calmodulin-dependent kinase II (CaMKII) 2 hr afterthe end of the stress session. The CaMKII inhibitor KN-62 antagonized the stress-mediated learning enhancement, however, with no effect on PS-LTP persistence.Thus, long-lasting increased neuronal excitability as reflected in PS-LTP primingappeared to be essential for the enhancement of learning in view of the observationthat inhibition of PS-LTP priming was associated with impaired learning.Conversely, it was demonstrated that inhibition of CaMKII activity reducedcontextual fear conditioning without affecting PS-LTP priming. This observationsuggests that priming of PS-LTP and activation of CaMKII represent two essentialmechanisms that may contribute independently to long-term memory.

Chapter 7

102

INTRODUCTION

Corticotropin-releasing factor (CRF) is a 41 amino acid neuropeptide that issynthesized in the hypothalamus and mediates the release of adrenocorticotropichormone from the anterior pituitary (Spiess et al., 1981; Vale et al., 1981). Theanatomical distribution of CRF in the brain suggests that this peptide not onlystimulates the release of corticotropin from the pituitary but also may modulate theneuronal activity of various other brain areas (Chalmers et al., 1995; Chang et al.,1993; De Souza et al., 1985; Potter et al., 1994). CRF has been shown to modulatelearning, food intake, arousal, startle and fear responses, general motor activity,body temperature, and sexual activity (Buwalda et al., 1997; Heinrichs et al., 1995;Holahan et al., 1997; Linthorst et al., 1997; Radulovic et al., 1999). Exogenousapplication of CRF to hippocampal slices reduces the slow afterhyperpolarizationand spike frequency accommodation (Aldenhoff et al., 1983; Haug and Storm, 2000)and enhances the amplitude of CA1 population spikes evoked by stimulation of theSchaffer collateral pathway in rats (Hollrigel et al., 1998). Recent studies havedemonstrated that CRF produces a long-lasting enhancement of synaptic efficacy inthe rat hippocampus in vivo (Wang et al., 1998, 2000). CRF has also been implicatedin learning in view of the observation that CRF injection into the mousehippocampus a few minutes before training enhances classical fear conditioningsignificantly (Radulovic et al., 1999). When injected directly into the dentate gyrusof the hippocampus, CRF improves the retention of a one-way inhibitory avoidancelearning in rats (Lee et al., 1992). However, no electrophysiological studies on thefunction of CRF in the mouse hippocampus have been performed to date.

The following series of experiments were aimed at further defining theeffect of acute stress and human/rat CRF (h/rCRF) on hippocampus-dependentlearning and on long-term synaptic plasticity in mouse hippocampus. In view of thepossibility that acute stress can induce changes in thresholds for synaptic plasticitynecessary for long-term potentiation (LTP) induction (Foy et al., 1987; Kim et al.,1996; Kim and Yoon, 1998), which has been referred to as “metaplasticity”(Abraham and Bear, 1996), we investigated the effects of h/rCRF andimmobilization stress on the induction and persistence of LTP of population spikes(PS-LTP). The threshold for hippocampus-dependent synaptic plasticity andmemory storage is thought to be determined by protein phosphorylation (Huang,1998). In particular, activation of protein kinase C (PKC) (Wang and Feng, 1992),Ca2+/calmodulin-dependent kinase II (CaMKII) (Malenka et al., 1989), or both(Malinow et al., 1989) has been suggested to be indispensable for induction ofexcitatory postsynaptic field potential (fEPSP)-LTP in the hippocampal CA1 region.Thus, we assessed the roles of PKC and CaMKII, in the regulation of hippocampallong-term synaptic plasticity and in the performance of mice in a hippocampus-dependent learning task.

CRF primes PS-LTP in mouse hippocampus

103


Animals

Experiments were performed on 9-12 week-old male Balb/c mice (Charles River, Sultzfeld,Germany). The mice were individually housed and maintained on a 12 hr light/dark cycle (lights on at7 am) with free access to food and water ad libitum. All experimental procedures were in accordancewith the European Council Directive (86/609/EEC) by permission of the Animal Section Law enforcedby the District Government of Braunschweig (Lower Saxony, Germany).

Hippocampal Slice ElectrophysiologyMice were briefly anesthetized with isofluran and then decapitated. In <1 min the skull was

opened, and the brain was removed and transferred to ice cold artificial CSF solution of the followingcomposition (in mM): 130 NaCl, 3.5 KCl, 1.25 NaH2PO4, 1.5 MgSO4, 2 CaCl2, 24 NaHCO3 and 10glucose, equilibrated with 95% O2/5% CO2 (pH 7.4). Hippocampi were dissected from the chilled brainhemispheres on ice. Transverse hippocampal slices (400 µm) were obtained on a McIlwain tissuechopper (The Mickle Laboratory Engineering Co. LTD, Surrey, England) and kept submerged(minimum of 1 hr at room temperature before recordings) in aCSF. Extracellular field potentials wererecorded in a recording chamber maintained at 32°C with recording electrodes pulled from borosilicateglass and filled with 2 M NaCl (3-5 mΩ). All recordings were made using a SEC-05L amplifier (npiElectronics, Tamm, Germany). To record field potentials in the CA1 pyramidal cell body layer,Schaffer collaterals were stimulated with a bipolar electrode placed on the surface of the slice. At thebeginning of each experiment, a stimulus-response curve was established by increasing the stimulusintensity and measuring the amplitude of the population spike. On the basis of the input-outputfunction, the stimulus was adjusted to elicit a population spike with an amplitude of half the maximumand was fixed at this level throughout the experiments. PS-LTP was induced by theta burst stimulation(TBS) at the test pulse intensity, consisting of 5 x 100 Hz bursts (five diphasic pulses per burst) with a200 msec interburst interval. Traces were stored on a computer using Pulse 7.4 software (Heka,Lambrecht, Germany) for off-line analysis. Short-term potentiation (STP) and PS-LTP were measured5 and 60 min after tetanic stimulation, respectively.

CannulationDouble guide cannulas (C235, Plastics One, Roanoke, VA) were implanted using a

stereotactic holder during 1.2% avertin anesthesia (0.02 ml/g, i.p.) under aseptic conditions asdescribed previously (Radulovic et al., 1999; Stiedl et al., 2000). Each double guide cannula withinserted dummy cannula and dust cap was fixed to the skull with dental cement. The cannulas wereplaced into both lateral brain ventricles, with anteroposterior (AP) coordinates zeroed at Bregma (AP, 0mm; lateral, 1 mm; depth, 3 mm) or directed toward both dorsal hippocampi (AP, –1.5 mm; lateral, 1mm; depth, 2 mm) (Franklin and Paxinos, 1997). The animals were allowed to recover for 4-5 d beforethe experiments started. On the day of the experiment, bilateral injections were performed using aninfusion pump (CMA/100, CMA/Microdialysis, Solna, Sweden) at a constant rate of 0.33 µl/min (finalvolume, 0.25 µl/side). Cannula placement was verified post hoc in all mice by injection of methyleneblue dye. For electrophysiological experiments, double guide cannula placement was verified byunilateral methylene blue injection. We never observed any effects of cannulation itself or vehicleinjection.

Drugsh/rCRF (Rühmann et al., 1996) and [Glu11,16]astressin (Eckart et al., 2001) were synthesized

in our laboratory as described. KN-62 and bisindolylmaleimide I (BIS-I) were from Calbiochem (SanDiego, CA).

Drug Treatment[Glu11,16]astressin was dissolved in aCSF solution. h/rCRF stock solutions were prepared in

10 mM acetic acid. Final dilutions in aCSF to 400 ng/µl were prepared immediately before theexperiments. The final pH of the peptide solution was 7.4. KN-62 was dissolved in DMSO to aconcentration of 4 mg/ml. For injection the stock was diluted in aCSF to a final concentration of 64ng/µl. BIS-I was stored as 1 mM stock solution in DMSO. For injection the solution was diluted with

Chapter 7

104

aCSF to a final concentration of 0.4 nmol/µl. DMSO in the concentrations used did not exhibit anysignificant effect by itself on synaptic responses or learning.

Immobilization StressAn acute immobilization stress of mice consisted of taping their limbs to a plastic surface for

1 hr (Smith et al., 1995).

Fear ConditioningThe fear-conditioning experiments were performed as described previously (Stiedl et al.,

2000) using a computer-controlled fear-conditioning system (TSE, Bad Homburg, Germany). Fearconditioning was performed in a Plexiglas cage (36 x 21 x 20 cm) within a fear-conditioning box thatwas constantly illuminated (12 V, 10 W halogen lamp, 100-500 lux). In the conditioning box, a high-frequency loudspeaker (KT-25-DT, Conrad, Hirschau, Germany) provided constant background noise[white noise, 68 dB sound pressure level (SPL)]. The training (conditioning) consisted of a single trial.The mouse was exposed to the conditioning context (180 sec) followed by a tone [conditioned stimulus(CS) 30 sec, 10 kHz, 75 dB SPL, pulsed 5 Hz]. After termination of the tone, a foot shock[unconditioned stimulus(US) 0.7 mA, 2 sec, constant current] was delivered through a stainless steelgrid floor. The mouse was removed from the fear-conditioning box 30 sec after shock termination toavoid an aversive association with the handling procedure. Under these conditions, the context servedas background stimulus. Background contextual fear conditioning but not foreground contextual fearconditioning, in which the tone is omitted during training, has been shown to involve the hippocampus(Phillips and LeDoux, 1994). Memory tests were performed 24 hr after fear conditioning. Contextualmemory was tested in the fear conditioning box for 180 sec without CS or US presentation (withbackground noise). Freezing, defined as the lack of movement except for respiration and heart beat,was assessed as the behavioral parameter of the defensive reaction of mice (Blanchard and Blanchard,1969; Bolles and Riley, 1973; Fanselow and Bolles, 1979) by a time-sampling procedure every 10 secthroughout memory tests. In addition, activity-derived measures (inactivity, mean activity, andexploratory area) were recorded by a photo beam system (10 Hz detection rate) controlled by the fear-conditioning system.

Western BlottingHippocampal slices were prepared as described above. CA1 subregions of hippocampal slices

were dissected out and immediately homogenized at 4°C with a plastic homogenizer in homogenizationbuffer containing 50 mM Tris-HCl, pH 8.0, 10 mM EDTA, 4 mM EGTA, 15 mM sodium phosphate,100 mM β-glycerophosphate, 10 mM sodium fluoride, 2 µg/ml aprotinin, 2 µg/ml leupeptin, 0.7 µg/mlpepstatin and 1 mM phenylmethylsulfonyl fluoride, pH 7.4. The insoluble material was removed bycentrifugation at 15,000 x g for 10 min at 4°C. Protein concentrations were determined with a Bradfordassay (BioRad, Munich, Germany). Equal amounts of protein for each group were separated on a 10%SDS gel and transferred to an Immobilon-P membrane (Millipore Corporation, Bedford, MA) using asemidry transfer apparatus. The blot was probed using an anti-active CaMKII antibody (Promega,Madison, WI) or antibody directed against total CaMKII (Chemicon, Temecula, CA) and detected withhorseradish peroxidase-conjugated second antibody. Western blots were developed using thechemiluminescence method.

StatisticsStatistical comparisons were made using Student’s t test and ANOVA. Data were expressed

as mean ± SEM. Significance was determined at the level of p < 0.05.

RESULTS

Activation of CRF receptors primes PS-LTPUsing intracellular recordings, we have found previously that h/rCRF

increased pyramidal cell excitability in the hippocampal CA1 area of Balb/c mice,


107

Recordings from slices obtained 2 hr after immobilization revealed that[Glu11,16]astressin (104 ± 6%, n = 5) and BIS-I (108 ± 7%, n = 5) (Fig. 2B) hadcompletely blocked the persistence of PS-LTP. These values were not significantlydifferent from those of non stressed controls (Fig. 2A). In addition, STP wassignificantly attenuated in slices from stressed animals after injection of[Glu11,16]astressin (140 ± 5%, n = 5, p < 0.05) (Fig. 2B). Western blot experimentsshowed that exposure of the animals to 1 hr immobilization resulted in elevatedimmunoreactivity of active, phosphorylated CaMKII, with a maximum at 2 hr afterthe stress session (Fig. 3). However, intracerebroventricular injection of the selectiveCaMKII inhibitor KN-62 before the stress session did not significantly reduce PS-LTP persistence (155 ± 10%, n = 5) (Fig. 2B) compared with PS-LTP induced inslices from stressed animals (Fig. 2A).

Figure 3. Acute stress induces activation of CaMKII in the hippocampal CA1 area.A , CA1 homogenates dissected from animals at several time points after 1 hrimmobilization were probed with an antibody specific for Thr286-phosphorylatedCaMKII, or an antibody recognizing total CaMKII. Non-stressed mice are shown ascontrols. Results are representative of four independent Western blots. B, the bargraph summarizes Western blot data of four experiments and shows the levels ofThr286-phosphorylated CaMKII expressed as a percentage of nonstressed controls.The dashed line represents the normalized average of nonstressed controls (set to100%). Statistically significant differences: * p < 0.05 versus nonstressed controls.

Acute stress and fear conditioningThe previous experiments have shown that acute stress facilitated PS-LTP in

the hippocampus and that this facilitation was prevented only by BIS-I and[Glu11,16]astressin but not by KN-62. In the next series of experiments, weinvestigated whether the same compounds exhibited effects on learning ofconditioned fear. Thus, mice received bilateral aCSF intrahippocampal injectionsimmediately before 1 hr immobilization and were trained at 0, 1, 2, and 3 hr aftertermination of the stress session. At 1 hr (p < 0.01; n = 11), 2 hr (p < 0.001; n = 12)and 3 hr (p < 0.001; n = 10) after exposure to immobilization, contextual fear wassignificantly enhanced compared with non-stressed controls (n = 30) (Fig. 4).

P-Thr286

CaMKII

Total CaMKII

non-

stres

sed

0 hr1 h

r2h

r3 h

r

A*

2hr

0 hr

1 hr

non-

stres

sed0

50100150200250

3 hr

P-T

hr28

6 C

aMK

II(%

of

amou

nt in

non-

stre

ssed

con

trol

s)B

Chapter 7

108

Injection of KN-62 intrahippocampally before stress significantly reduced freezing 2hr (p < 0.001; n = 8) and 3 hr (p < 0.01; n = 5) after the end of the stress sessioncompared with stressed, aCSF-injected mice (Fig.4). Similarly, freezing was reduced2 hr after exposure to immobilization (p < 0.005; n = 8) when KN-62 was injectedintracerebroventricularly (Fig.4). However, KN-62 injection alone(intrahippocampally) without exposing the animal to immobilization significantlyreduced freezing 3 hr after injection compared with non stressed, aCSF-injectedcontrols (p < 0.001, n = 5; data not shown). Freezing was also significantly reduced2 hr after the stress session when mice were injected before stress with BIS-Iintrahippocampally (p < 0.01; n = 5) and intracerebroventricularly (p < 0.01; n = 9)(Fig. 4). Likewise, [Glu11,16]astressin significantly reduced conditioned contextualfear 2 hr after the end of immobilization when injected intrahippocampally (p <0.001; n = 5) and intracerebroventricularly (p < 0.01; n = 8) compared with stressed,aCSF-injected mice. Both compounds, BIS-I and [Glu11,16]astressin, had nosignificant effect on freezing when mice were injected intrahippocampally andintracerebroventricularly without immobilization (data not shown).

Figure 4. Inhibition of CRF receptors, PKC and CaMKII in the dorsal hippocampusprevent stress-mediated enhancement of context-dependent fear conditioning. A, Micewere injected intrahippocampally (i.h.) or intracerebroventricularly (i.c.v.) withaCSF, KN-62 or BIS-I immediately before the start of the stress session and trained atseveral timepoints after the end of the stress session. B , mice were injectedintrahippocampally or intracerebroventricularly with [Glu11,16]astressin immediatelybefore the stress session and trained 2 hr after immobilization. Freezing wasmeasured in the memory test performed 24 hr after training. Statistically significantdifferences: * p < 0.05 versus aCSF; ** p < 0.05 versus control.

**

*

**

2 hr i.h.

[Glu 11,16]astr.+ stress

2 hri.c.v.

* * * **

** ** ****

3 hr2 hr1 hr0 hr0

1020304050607080

Free

zing

(%

)

2 hri.h. i.c.v.

A B

*

con con

aCSF + stressKN-62 + stressBIS + stress


109

DISCUSSION

Recent behavioral studies have shown that stress (Conrad et al., 1999; Deaket al., 1999) and intracerebroventricular injection of h/rCRF (Radulovic et al., 1999)enhance, under defined conditions, contextual fear conditioning, which represents ahippocampus-dependent learning task in rats (Holland and Bouton, 1999; Kim andFanselow, 1992; Phillips and LeDoux, 1992) and mice (Abel et al., 1997; Chen etal., 1996; Logue et al., 1997). At the same time, stress blocks hippocampal highfrequency stimulation (HFS)-induced LTP and facilitates long-term depression(LTD) (Xu et al., 1997). In the present experiments, we found that activation of CRFreceptors mediated the PS-LTP priming observed after acute stress in the CA1 areaof the mouse hippocampus. CRF can be directly secreted from nerve terminalslocated in the hippocampus during exposure to stress. Specifically, numerous, largeCRF-immunoreactive neurons have been found in the hippocampal CA1 and CA3region (Merchenthaler , 1984; Swanson et al., 1983).

A similar priming effect has been observed after activation of group Imetabotropic glutamate receptors (mGluRs). The mGluR agonist 1S,3R-aminocyclopentanedicarboxylic acid (ACPD) was found to facilitate the persistenceof fEPSP-LTP in area CA1 of rat hippocampal slices (Cohen and Abraham, 1996;Cohen et al., 1999; Raymond et al., 2000). In contrast to ACPD, h/rCRF did notsignificantly enhance STP.

It is important to note that in all our electrophysiological experiments, brainslices were equilibrated in aCSF for at least 2-3 hr after slice preparation before PS-LTP was induced. Consequently, it appears that the mechanisms critical for PS-LTPpriming either have a slow turnover rate or reveal high stability to be detectable inbrain slices from stressed mice even hours after preparation. This assumption is alsoconfirmed by the finding that bath-applied h/rCRF was still effective at PS-LTPpriming at least 2 hr after wash-out (Blank and Nijholt, unpublished observations).Recently, it was shown that previous activity which generates protein synthesis-dependent LTP may also prime the persistence of LTP in a second input (Frey andMorris, 1997). Similarly, activation of group I mGluRs facilitates the persistence ofLTP in area CA1 of rat hippocampal slices by triggering de novo protein synthesisfrom existing mRNA (Raymond et al., 2000). In the case of mGluR-mediatedenhancement of LTP the new proteins seem to be synthesized in close proximity tothe activated synapses. This observation implies for our experiments that newlysynthesized proteins would have to be preserved or protein synthesis would have tocontinue during equilibration of the slices in aCSF to still affect PS-LTP persistenceseveral hours after slice preparation.

The capability of the PKC inhibitor BIS-I to prevent PS-LTP priming byh/rCRF and stress is consistent with the described priming effect of group I mGluRactivation which can lead to the liberation of inositol triphosphate and to thesubsequent activation of PKC in hippocampal slices (Cohen et al., 1998).Interestingly, the PKC inhibitor BIS-I prevented only h/rCRF-mediated primingwhen it was applied 1 hr before administration of h/rCRF. BIS-I had no effect onPS-LTP persistence when applied to slices already primed after TBS. This findingcontradicts the hypothesis that persistent activation of PKC underlies the observedPS-LTP maintenance (Colley et al., 1990; Wang and Feng, 1992).

Chapter 7

110

On the basis of the observation that [Glu11,16]astressin and BIS-I incombination with h/rCRF or stress reduced STP in hippocampal slices, it appearedpossible that hippocampal CRF receptors and PKC were tonically activated. Insupport of this hypothesis, activation of PKC has been shown to be involved in LTPinduction in rat hippocampal CA1 cells (Hvalby et al., 1994). However, BIS-I and[Glu11,16]astressin exhibited no effect on STP or contextual fear conditioning whenapplied individually, findings that argue against a possible involvement of tonicallyactive hippocampal CRF receptors or tonically active PKC. It seems notable that[Glu11,16]astressin and BIS-I affected induction of PS-LTP only when neuronalactivity was enhanced, but not under baseline conditions. However, in the presenceof BIS-I or [Glu11,16]astressin, only a limited enhancement of neuronal activity canbe expected, which most likely generated only a modest elevation of intracellularCa2+ concentrations. Low levels of Ca2+ favor the activation of proteinphosphatases, which, in turn, would lower the probability of LTP induction (Kato etal., 1999; Mulkey et al., 1994).

In addition, our findings revealed that pharmacological inhibition ofhippocampal CRF receptors and PKC impaired the observed stress-mediatedlearning enhancement. These data are consistent with previous observationsindicating that the contextual learning impairment of DBA mice can be reversed byactivation of hippocampal PKC as determined by the fear conditioning task(Fordyce et al., 1995).

In the present study, we found elevated levels of active CaMKII in themouse CA1 area 2 hr after exposure to immobilization. Injection of the selectiveCaMKII inhibitor KN-62 did not prevent priming of PS-LTP persistence observedwhen hippocampal brain slices were prepared 2 hr after the end of the stress session.Although CaMKII activity was increased at this time, there was no evidence thatbaseline synaptic transmission was altered. Conflicting data about the effects ofCaMKII activation on basal synaptic transmission and LTP induction in the CA1region of the hippocampus have been reported. For example, it was found thatincreasing CaMKII activity in CA1 neurons by viral transfection (Pettit et al., 1994)or by injection of the active enzyme (Lledo et al., 1995) results in an enhancement ofsynaptic transmission and impairment of LTP induction. Direct injection of aconstitutively active form of CaMKII into postsynaptic CA1 neurons potentiatesevoked EPSCs significantly within 15-30 min (Lledo et al., 1995). In contrast, inCaMKII-Asp286 transgenic mice expressing activated CaMKII, no change in basalsynaptic transmission has been observed (Mayford et al., 1995). Nevertheless,transgenic expression of activated CaMKII eliminates LTP in the range of 5 to 10Hz with no effect on LTP in response to 100 Hz tetanus stimulation (Mayford et al.,1995). In previous experiments (Blank and Nijholt, unpublished observations), wefound significant PS-LTP impairment in the CA1 area of stressed mice (2 hr afterthe end of immobilization) when PS-LTP was induced with the standard 100 HzHFS protocol, which produced saturating PS-LTP in control animals. At that time,we also found elevated CaMKII activity in the hippocampal CA1 area, asdemonstrated in the present study. Our finding that the facilitatory effects of acutestress on hippocampus-dependent learning paralleled those of PS-LTP induced byTBS but not by HFS might be explained by the fact that the impact of GABAergictransmission on LTP expression is highly dependent on tetanization parameters


111

(Chapman et al., 1998). However, reduced GABAergic inhibition would be expectedto result in elevated basal synaptic transmission in brain slices of stressed animals.We did not observe such effect. If acute stress already induces maximal LTP-likephenomena in the mouse hippocampus, strong stimulation of hippocampal brainslices would not be expected to produce additional LTP (Kim et al., 1996). Thus, thequestion may not be whether TBS instead of HFS is responsible for facilitation ofLTP in brain slices obtained from stressed animals but rather whether to use weakLTP stimulation instead of a strong stimulation protocol. In support of thishypothesis, Cohen and Abraham (1996) described that activation of mGluRs onlyfacilitates the induction of LTP induced by weak TBS but does not enhance LTPinduced by strong stimulation.

LTP is the most extensively studied form of neuroplasticity and is widelybelieved to be the substrate for learning and memory (Bear and Abraham, 1996;Bliss and Collingridge, 1993; Maren and Baudry, 1995). However, a definitivelinkage of LTP to learning or memory has not been achieved. Our data show a clearcorrelation between improved learning and facilitation of TBS-induced PS-LTPpersistence, and both phenomena appeared to be PKC-dependent. At the same time,inhibition of CaMKII activity prevented learning improvement without impairmentof PS-LTP persistence. The performance in a hippocampus-dependent task wasaffected without interference with the long-lasting enhancement of hippocampalsynaptic transmission. This result suggests that the underlying synaptic mechanismsof PS-LTP priming seem to be necessary, but not sufficient, for learningenhancement.

ACKNOWLEDGMENTS

We are grateful to Thomas Liepold and Hossein Tezval for amino acidanalysis and peptide synthesis. We thank Stefanie Vollstädt for excellent technicalassistance.

REFERENCES

Abel T, Nguyen PV, Barad M, Deuel TA, Kandel ER and Bourtchouladze R (1997) Geneticdemonstration of a role for PKA in the late phase of LTP and in hippocampus-based long-termmemory. Cell 88: 615-626.

Abraham WC and Bear MF (1996) Metaplasticity: the plasticity of synaptic plasticity. TrendsNeurosci. 19: 126-130.

Aldenhoff JB, Gruol DL, Rivier J, Vale W and Siggins GR (1983) Corticotropin-releasing factordecreases postburst hyperpolarizations and excites hippocampal neurons. Science 221: 875-877.

Bear MF and Abraham WC (1996) Long-term depression in hippocampus. Annu. Rev. Neurosci. 19:437-462.

Chapter 7

112


Bliss TV and Collingridge GL (1993) A synaptic model of memory: long-term potentiation in thehippocampus. Nature 361: 31-39.

Bolles RC and Riley AL (1973) Freezing as an avoidance response: another look at the operant-respondent distinction. Learn. Motiv. 4: 268-275.

Buwalda B, de Boer SF, Van Kalkeren AA and Koolhaas JM (1997) Physiological and behavioraleffects of chronic intracerebroventricular infusion of corticotropin-releasing factor in the rat.Psychoneuroendocrinology 22: 297-309.

Chalmers DT, Lovenberg TW and De Souza EB (1995) Localization of novel corticotropin-releasingfactor receptor (CRFR2) mRNA expression to specific subcortical nuclei in rat brain: comparison withCRF1 receptor mRNA expression. J. Neurosci. 15: 6340-6350.

Chang CP, Pearse RV, O’Connell S and Rosenfeld MG (1993) Identification of a seven transmembranehelix receptor for corticotropin-releasing factor and sauvagine in mammalian brain. Neuron 11: 1187-1195.

Chapman CA, Perez Y and Lacaille JC (1998) Effects of GABA(A) inhibition on the expression oflong-term potentiation in CA1 pyramidal cells are dependent on tetanization parameters. Hippocampus8: 289-298.

Chen C, Kim JJ, Thompson RF and Tonegawa S (1996) Hippocampal lesions impair contextual fearconditioning in two strains of mice. Behav. Neurosci. 110: 1177-1180.

Cohen AS and Abraham WC (1996) Facilitation of long-term potentiation by prior activation ofmetabotropic glutamate receptors. J. Neurophysiol. 76: 953-962.

Cohen AS, Coussens CM, Raymond CR and Abraham WC (1999) Long-lasting increase in cellularexcitability associated with the priming of LTP induction in rat hippocampus. J. Neurophysiol. 82:3139-3148.

Cohen AS, Raymond CR and Abraham WC (1998) Priming of long-term potentiation induced byactivation of metabotropic glutamate receptors coupled to phospholipase C. Hippocampus 8: 160-170.

Colley PA, Sheu FS and Routtenberg A (1990) Inhibition of protein kinase C blocks two componentsof LTP persistence, leaving initial potentiation intact. J. Neurosci. 10: 3353-3360.

Conrad CD, LeDoux JE, Magarinos AM and McEwen BS (1999) Repeated restraint stress facilitatesfear conditioning independently of causing hippocampal CA3 dendritic atrophy. Behav. Neurosci. 113:902-913.

Deak T, Nguyen KT, Ehrlich AL, Watkins LR, Spencer RL, Maier SF, Licinio J, Wong ML, ChrousosGP, Webster E and Gold PW (1999) The impact of the nonpeptide corticotropin-releasing hormoneantagonist antalarmin on behavioral and endocrine responses to stress. Endocrinology 140: 79-86.

De Souza EB, Insel TR, Perrin MH, Rivier J, Vale WW and Kuhar MJ (1985) Corticotropin-releasingfactor receptors are widely distributed within the rat central nervous system: an autoradiographic study.J. Neurosci. 5: 3189-3203.



113

Eckart K, Radulovic J, Radulovic M, Jahn O, Blank T, Stiedl O and Spiess J (1999) Actions of CRFand its analogs. Curr. Med. Chem. 6: 1035-1053.


Fordyce DE, Clark VJ, Paylor R and Wehner JM (1995) Enhancement of hippocampally-mediatedlearning and protein kinase C activity by oxiracetam in learning-impaired DBA/2 mice. Brain Res. 672:170-176.

Foy MR, Stanton ME, Levine S and Thompson RF (1987) Behavioral stress impairs long-termpotentiation in rodent hippocampus. Behavioral & Neural Biology 48: 138-149.


Frey U and Morris RGM (1997) Synaptic tagging and long-term potentiation. Nature 385: 533-536.

Haug T and Storm JF (2000) Protein kinase A mediates the modulation of the slow Ca(2+)-dependentK(+) current, I(sAHP), by the neuropeptides CRF, VIP, and CGRP in hippocampal pyramidal neurons.J. Neurophysiol. 83: 2071-2079.

Heinrichs SC, Menzaghi F, Merlo Pich E, Britton KT and Koob GF (1995) The role of CRF inbehavioral aspects of stress. Ann. N. Y. Acad. Sci. 771: 92-104.

Holahan MR, Kalin NH and Kelley AE (1997) Microinfusion of corticotropin-releasing factor into thenucleus accumbens shell results in increased behavioral arousal and oral motor activity.Psychopharmacology 130: 189-196.

Holland PC and Bouton ME (1999) Hippocampus and context in classical conditioning. Curr. Opin.Neurobiol. 9: 195-202.

Hollrigel GS, Chen K, Baram TZ and Soltesz I (1998) The pro-convulsant actions of corticotropin-releasing hormone in the hippocampus of infant rats. Neuroscience 84: 71-79.

Huang EP (1998) Synaptic plasticity: going through phases with LTP. Curr. Biol. 8: R350-2.

Hvalby O, Hemmings HC Jr, Paulsen O, Czernik AJ, Nairn AC, Godfraind JM, Jensen V, Raastad M,Storm JF, Andersen P and Greengard P (1994) Specificity of protein kinase inhibitor peptides andinduction of long-term potentiation. Proc. Natl. Acad. Sci. USA 91: 4761-4765.

Kato K, Li ST and Zorumski CF (1999) Modulation of long-term potentiation induction in thehippocampus by N-methyl-D-aspartate-mediated presynaptic inhibition. Neuroscience 92: 1261-1272.



Kim JJ and Yoon KS (1998) Stress: metaplastic effects in the hippocampus. Trends Neurosci. 21: 505-509.

Koob GF and Heinrichs SC (1999) A role for corticotropin releasing factor and urocortin in behavioralresponses to stressors. Brain Res. 848: 141-152.

Lee EH, Hung HC, Lu KT, Chen WH and Chen HY (1992) Protein synthesis in the hippocampusassociated with memory facilitation by corticotropin-releasing factor in rats. Peptides 13: 927-937.

Chapter 7

114

Linthorst ACE, Flachskamm C, Hopkins SJ, Hoadley ME, Labeur MS, Holsboer F and Reul JMHM(1997) Long-term intracerebroventricular infusion of corticotropin-releasing hormone altersneuroendocrine, neurochemical, autonomic, behavioral, and cytokine responses to a systemicinflammatory challenge. J. Neurosci. 17: 4448-4460.

Lledo PM, Hjelmstad GO, Mukherji S, Soderling TR, Malenka RC and Nicoll RA (1995)Calcium/calmodulin-dependent kinase II and long-term potentiation enhance synaptic transmission bythe same mechanism. Proc. Natl. Acad. Sci. USA 92: 11175-11179.

Logue SF, Paylor R and Wehner JM (1997) Hippocampal lesions cause learning deficits in inbred micein the Morris water maze and conditioned-fear task. Behav. Neurosci. 111: 104-113.

Malenka RC, Kauer JA, Perkel DJ, Mauk MD, Kelly PT, Nicoll RA and Waxham MN (1989) Anessential role for postsynaptic calmodulin and protein kinase activity in long-term potentiation. Nature340: 554-557.

Malinow R, Schulman H and Tsien RW (1989) Inhibition of postsynaptic PKC or CaMKII blocksinduction but not expression of LTP. Science 245: 862-866.

Maren S and Baudry M (1995) Properties and mechanisms of long-term synaptic plasticity in themammalian brain: relationships to learning and memory. Neurobiol. Learn. Mem. 63: 1-18.

Mayford M, Wang J, Kandel ER and O`Dell TJ (1995) CaMKII regulates the frequency-responsefunction of hippocampal synapses for the production of both LTD and LTP. Cell 81: 891-904.

Merchenthaler I (1984) Corticotropin releasing factor (CRF)-like immunoreactivity in the rat centralnervous system. Extrahypothalamic distribution. Peptides 5: 53-69.

Mulkey RM, Endo S, Shenolikar S and Malenka RC (1994) Involvement of a calcineurin/inhibitor-1phosphatase cascade in hippocampal long-term depression. Nature 369: 486-488.

Pettit DL, Perlman S and Malinow R (1994) Potentiated transmission and prevention of further LTP byincreased CaMKII activity in postsynaptic hippocampal slice neurons. Science 266: 1881-1885.


Phillips RG and LeDoux JE (1994) Lesions of the dorsal hippocampal formation interfere withbackground but not foreground contextual fear conditioning. Learn. Mem. 1: 34-44.

Potter E, Sutton S, Donaldson C, Chen R, Perrin M, Lewis K, Sawchenko PE and Vale W (1994)Distribution of corticotropin-releasing factor receptor mRNA expression in the rat brain and pituitary.Proc. Natl. Acad. Sci. USA 91: 8777-8781.

Radulovic J, Rühmann A, Liepold T and Spiess J (1999) Modulation of learning and anxiety bycorticotropin-releasing factor (CRF) and stress: differential roles of CRF receptors 1 and 2. J. Neurosci.19: 5016-5025.

Raymond CR, Thompson V, Tate WP and Abraham WC (2000) Metabotropic glutamate receptorstrigger homosynaptic protein synthesis to prolong long-term potentiation. J. Neurosci. 20: 969-976.

Rühmann A, Köpke AK, Dautzenberg FM and Spiess J (1996) Synthesis and characterization of aphotoactivatable analog of corticotropin-releasing factor for specific receptor labeling. Proc. Natl.Acad. Sci. USA 93: 10609-10613.


115

Smith MA, Makino S, Kvetnansky R and Post RM (1995) Stress and glucocorticoids affect theexpression of brain-derived neurotrophic factor and neurotrophin-3 mRNAs in the hippocampus. J.Neurosci. 15: 1961-1970.



Swanson LW, Sawchenko PE, Rivier J and Vale WW (1983) Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study.Neuroendocrinology 36: 165-186.

Toullec D, Pianetti P, Coste H, Bellevergue P, Grand-Perret T, Ajakane M, Baudet V, Boissin P,Boursier E, Loriolle F, Duhamel L, Charon D and Kirilovsky J (1991) The bisindolylmaleimide GF109203X is a potent and selective inhibitor of protein kinase C. J. Biol. Chem. 266: 15771-15781.

Turnbull AV and Rivier C (1997) Corticotropin-releasing factor (CRF) and endocrine responses tostress: CRF receptors, binding protein, and related peptides. Proc. Soc. Exp. Biol. Med. 215: 1-10.

Vale W, Spiess J, Rivier C and Rivier J (1981) Characterization of a 41-residue ovine hypothalamicpeptide that stimulates secretion of corticotropin and β-endorphin. Science 213: 394-1397.

Wang JH and Feng DP (1992) Postsynaptic protein kinase C essential to induction and maintenance oflong-term potentiation in the hippocampal CA1 region. Proc. Natl. Acad. Sci. USA 89: 2576-2580.

Wang HL, Tsai LY and Lee EH (2000) Corticotropin-releasing factor produces a protein synthesis-dependent long-lasting potentiation in dentate gyrus neurons. J. Neurophysiol. 83: 343-349.

Wang HL, Wayner MJ, Chai CY and Lee EH (1998) Corticotropin-releasing factor produces a long-lasting enhancement of synaptic efficacy in the hippocampus. Eur. J. Neurosci. 10: 3428-3437.


CHAPTER 8

Modulation of neuronal excitability and associative

learning by different corticotropin-releasing factor-

signaling cascades in mouse hippocampus

Submitted

ABSTRACT

Corticotropin-releasing factor (CRF) exerts a key neuroregulatory control onstress responses in various regions of the mammalian brain including thehippocampus. However, the signal transduction pathways that mediate the responsesto CRF in the hippocampus are largely uncharacterized. In the present study, weinvestigated the effects of human/rat CRF (h/rCRF) on hippocampal neuronalexcitability and hippocampus-dependent learning in two mouse inbred strains,Balb/c and C57BL/6N. In hippocampal brain slices from mice of both inbred strains,h/rCRF increased the neuronal activity of pyramidal cells in the CA1 region. Thisenhancement was blocked by the CRF receptor antagonist [Glu 11,16]astressin.Pretreatment of slices with bisindolylmaleimide I, a protein kinase C (PKC)inhibitor, prevented the h/rCRF effect only in slices from Balb/c mice but not inslices from C57BL/6N mice. On the other hand, H-89, an inhibitor of cAMP-dependent protein kinase (PKA), abolished the h/rCRF effect in pyramidal neuronsfrom C57BL/6N mice with no effect in slices from Balb/c mice. In support of thefinding, that, depending on the mouse inbred strain, h/rCRF enhanced hippocampalneuronal activity either through activation of the PKC- or PKA-pathway, wedemonstrated that h/rCRF elevated PKA activity in hippocampal slices fromC57BL/6N mice. In contrast, h/rCRF increased endogenous PKC activity in thehippocampus of Balb/c mice. Additional experiments using hippocampal membranesuspensions to analyze receptor-mediated guanine-nucleotide binding proteinactivation revealed that, upon stimulation by h/rCRF, CRF receptors mainly coupledto Gq/11 in the hippocampus of Balb/c mice and to Gs, Gq/11 and Gi in thehippocampus of C57BL/6N mice. Following intracerebroventricular injection of

Chapter 8

118

h/rCRF, context-dependent fear conditioning was improved in Balb/c mice, butunchanged in C57BL/6N mice.

Our findings demonstrated, that CRF receptors in mouse hippocampuscouple to various G proteins to activate at least two different intracellular signalingpathways. These differences may contribute to contrasting effects of h/rCRF onhippocampus-dependent learning observed in various mouse inbred strains.

CRF mediated excitability of mouse hippocampal neurons

119

INTRODUCTION

Corticotropin releasing factor (CRF) is a 41 amino acid neuropeptide, whichhas been implicated in both physiological and behavioral responses to stress. (Spiesset al., 1981; Vale et al., 1981). Two G protein-coupled receptor subtypes, CRFR1(Chang et al., 1993; Chen et al., 1993) and CRFR2 (Lovenberg et al., 1995), wererecently cloned. The expression pattern of CRFR2 in brain is distinct and morerestricted than that of CRFR1 (Van Pett et al., 2000). However, mRNA encoding forboth receptors is colocalized in several brain regions like the hippocampus (Van Pettet al., 2000). Mice deficient for CRFR1 display decreased anxiety-like behavior andhave an impaired stress response (Smith et al., 1998; Timpl et al., 1998). Studies ofmice lacking the CRFR2 gene suggest that this receptor is responsible for themodulation of stress responses and maintenance of homeostasis after hypothalamic-pituitary-adrenal (HPA) axis activation (Bale et al., 2000; Coste et al., 2000;Kishimoto et al., 2000). During exposure to stress CRF can be directly secreted fromnerve terminals located in the hippocampus. Specifically, numerous, large CRF-immunoreactive neurons have been found in the hippocampal CA1 and CA3 region(Merchenthaler I, 1984; Swanson et al., 1983). Previous studies have shown themodulation of hippocampus-dependent learning and memory by CRF. Human/ratCRF (h/rCRF) directly injected into the dentate gyrus consistently enhancedmemory retention in rats in a one-way passive avoidance task (Lee et al., 1993).Injection of h/rCRF into the dorsal hippocampus shortly before the trainingenhanced context- and tone-dependent fear conditioning in mice through CRFR1(Radulovic et al., 1999). Besides the effects on hippocampal learning tasks, CRFexerts a profound effect on hippocampal neuronal activity. In recent studies, it wasdemonstrated that h/rCRF produces a long-lasting enhancement of synaptic efficacyin the rat hippocampus in vivo (Wang et al., 1998, 2000). h/rCRF reversiblyincreases the spiking of rat hippocampal pyramidal cells (Aldenhoff et al., 1983) andenhances the amplitude of CA1 population spikes evoked by stimulation of theSchaffer collateral pathway (Hollrigel et al., 1998). We recently showed thatapplication of h/rCRF facilitates the induction and stability of long-term potentiation(LTP) in area CA1 of mouse hippocampal slices (Blank et al., in press).

To examine the signal transduction pathways of h/rCRF in mousehippocampus, we studied the G protein and second-messenger activation followingCRF receptor stimulation in hippocampi of two mouse inbred strains, C57BL/6Nand Balb/c. We have chosen these two inbred strains because C57BL/6 and Balb/cmice have repeatedly been found to differ strongly in several behavioral responses(Beuzen and Belzung, 1995; Oliverio et al., 1973; Peeler and Nowakowski, 1987) aswell as in neurodevelopmental and neurochemical parameters (Nowakowski, 1984).For example, Balb/c mice exhibit stronger anxiety-like responses in the light/darkchoice test (Beuzen and Belzung, 1995), in the open-field paradigm (Oliverio et al.,1973), and in a runway traversal locomotor activity test (Peeler and Nowakowski,1987). The impact of h/rCRF on neuronal excitability of CA1 pyramidal cells wasinvestigated in hippocampal slices from both mouse inbred strains. Finally, weinvestigated the influence of h/rCRF on hippocampus-dependent learning inC57BL/6N and Balb/c mice.

Chapter 8

120


AnimalsExperiments were carried out with male Balb/c and C57BL/6N mice (Charles River,

Sultzfeld, Germany) 9-12 weeks old. The mice were individually housed and maintained on a 12 hrlight/dark cycle (lights on at 7 am) with free access to food and water. All experimental procedureswere in accordance with the European Council Directive (86/609/EEC) by permission of the AnimalSection Law enforced by the District Government of Braunschweig, Lower Saxony, Germany.

ElectrophysiologyMice were briefly anesthetized with isofluran and then decapitated. In less than one minute

the skull was opened, the brain removed and transferred to ice cold artificial cerebrospinal fluid (aCSF)solution of the following composition (in mM): 130 NaCl, 3.5 KCl, 1.25 NaH2PO4, 1.5 MgSO4, 2CaCl2, 24 NaHCO3, 10 glucose, equilibrated with 95% O2/5% CO2 (pH 7.4). Hippocampi weredissected from the chilled brain hemispheres on ice. Transverse hippocampal slices (400 µM) wereobtained on a McIlwain tissue chopper and kept submerged (minimum of 1 hr at room temperaturebefore recordings) in aCSF.

Conventional intracellular recording techniques were employed using glass microelectrodesfilled with 3 M potassium acetate. Microelectrodes were pulled from borosilicate glass capillaries(WPI, Sarasota, FL) on a horizontal electrode puller (Zeitz-Instrumente, Augsburg, Germany). Themicroelectrode tip resistances ranged from 60 to 100 MΩ for recordings from mouse hippocampalneurons. Intracellular signals were recorded with a single-electrode voltage-clamp amplifier (SEC-05L,npi Electronics, Germany) which performed current-clamp measurements at high switching frequenciesin the range of 25-30 kHz. Bridge balance was monitored throughout the experiment and adjusted asrequired. Traces were stored on a computer using Pulse 7.4 software (HEKA, Lambrecht, Germany) foroff-line analysis. For intracellular recordings only neurons were included, which exhibitedovershooting action potentials, stable membrane potentials of at least -60 mV and input resistances of ≥35 MΩ. Input resistance was determined by measuring the voltage deflection at the end of a 100 mshyperpolarization current step (-0.2 nA). Depolarizing current pulses of 3-5 ms duration were injectedthrough the recording electrode to elicit single action potentials. Spike frequency adaptation wasinvestigated by injecting each cell with a series of 600 ms depolarizing current pulses (0.2-1 nA;increment 100 pA). In order to compare neuronal responses, the membrane potential of each cell wasmanually clamped to -65 mV by discontinuous current injection.

Drugsh/rCRF (Rühmann et al., 1996) and [Glu 11,16]astressin (Eckart et al., 2001) were synthesized

in our laboratory as described. H-89 and bisindolylmaleimide I (BIS-I) were from Calbiochem (SanDiego, CA). Phorbol 12,13-dibutyrate (PDBu) and 4α−phorbol were both from Sigma (St. Louis, MO).

Drug Treatment[Glu 11,16]astressin was dissolved in aCSF to a solution of 280 µM. h/rCRF stock solutions

were prepared in 10 mM acetic acid. For cannula injections final dilutions in aCSF to 400 ng/µl wereprepared immediately before the experiments. The final pH of the peptide solution was 7.4. BIS-I wasstored as 1 mM stock solution in dimethylsulfoxide (DMSO). For injection, the solution was dilutedwith aCSF to a final concentration of 0.4 nmol/µl. Phorbol 12,13-dibutyrate (PDBu) and 4α−phorbolwere both dissolved in DMSO to 5 µg/µl. For injection, the solutions were diluted with aCSF to a finalconcentration of 10 ng/µl.

CannulationDouble guide cannulae (C235, Plastics One, Roanoke, VA) were implanted using a

stereotactic holder during 1.2% avertin anesthesia (0.02 ml/g, intraperitoneal) under aseptic conditionsas previously described (Blank et al., in press; Stiedl et al., 2000). Each double guide cannula withinserted dummy cannula and dust cap was fixed to the skull of the mouse with dental cement. Thecannulae were placed into both lateral brain ventricles, with anteroposterior (AP) coordinates zeroed atBregma AP 0 mm, lateral 1 mm, depth 3 mm (Franklin and Paxinos, 1997). The animals were allowedto recover for 4-5 d before the experiments started. On the day of the experiment, bilateral injections


121

were performed using an infusion pump (CMA/Microdialysis) at a constant rate of 0.33 µl/min (finalvolume: 0.25 µl per side). Cannula placement was verified post hoc in all mice by injection ofmethylene blue. For electrophysiological experiments double guide cannula placement was verified byunilateral methylene blue injection.

Fear ConditioningThe fear conditioning experiments were performed as previously described (Blank et al., in

press; Stiedl et al., 2000) using a computer-controlled fear conditioning system (TSE, Bad Homburg,Germany). Fear conditioning was performed in a Plexiglas cage (36 x 21 x 20 cm) within a fearconditioning box constantly illuminated (12 V, 10 W halogen lamp, 100-500 lux). In the conditioningbox, a high-frequency loudspeaker (Conrad, KT-25-DT, Hirschau, Germany) provided constantbackground noise [white noise, 68 dB sound pressure level (SPL)]. The training (conditioning)consisted of a single trial. The mouse was exposed to the conditioning context (180 sec) followed by atone (CS, 30 sec, 10 kHz, 75 dB SPL, pulsed 5 Hz). After termination of the tone, a footshock (US, 0.7mA, 2 sec, constant current) was delivered through a stainless steel grid floor. The mouse was removedfrom the fear conditioning box 30 sec after shock termination to avoid an aversive association with thehandling procedure. Under these conditions, the context served as background stimulus. Backgroundcontextual fear conditioning but not foreground contextual fear conditioning, where the tone is omittedduring training, has been shown to involve the hippocampus (Phillips and LeDoux, 1994). Memorytests were performed 24 hr after fear conditioning. Contextual memory was tested in the fearconditioning box for 180 sec without CS or US presentation (with background noise). Freezing, definedas a lack of movement except for respiration and heart beat, was assessed as the behavioral parameterof the defensive reaction of mice (Blanchard and Blanchard, 1969; Bolles and Riley, 1973; Fanselowand Bolles, 1979) by a time-sampling procedure every 10 sec throughout the memory test. In addition,activity-derived measures (inactivity, mean activity, and exploratory area) were recorded by a photobeam system (10 Hz detection rate) controlled by the fear conditioning system.

PKA and PKC AssayscAMP-dependent protein kinase (PKA) and protein kinase C (PKC) activities were assayed

using the PepTag Assay for non-radioactive detection of PKC or PKA (Promega Corp., Madison,WI) based on the phosphorylation of fluorescent-tagged PKC- or PKA-specific peptides. Afterincubation in either aCSF or 250 nM h/rCRF for 30 min, hippocampal slices were placed in ice-coldhomogenization buffer (20 mM Tris-HCl, pH 7.4, 2 mM EDTA, 2 mM EGTA, 48 mMmercaptoethanol, 0.32 M sucrose and freshly added protease inhibitor cocktail tablet (BoehringerMannheim). The tissue was homogenized with a teflon-plastic homogenizer and centrifuged at 100,000x g for 30 min in a Beckman XL-80 ultracentrifuge. The resulting supernatant contained the PKApreparation. The pellet was rehomogenized in homogenization buffer and sonicated 4 x 15 s, incubatedfor 30 min with Triton X-100 (0.2%) and centrifuged at 100,000 x g for 30 min. The supernatantcontained the membrane-bound PKC preparation, which was used for the PKC assay. Proteinconcentrations were determined with the Bradford assay (BioRad, München, Germany). The assay wasperformed as described in the manufacturer’s protocol. An aliquot of the PKA preparation wasincubated for 30 min at 30°C in PepTag PKA 5x reaction buffer (100 mM Tris-HCl, pH 7.4, 50 mMMgCl2, 5mM ATP) and 0.4 µg/µl of the PKA-specific peptide substrate PepTag A1 (L-R-R-A-S-L-G, Kemptide). The same procedure was used for the PKC preparations which were incubated inPepTag PKC reaction buffer (100 mM HEPES, pH 7.4, 6.5 mM CaCl2, 5 mM DTT, 50 mM MgCl2, 5mM ATP) containing 0.4 µg/µl of the PKC-specific peptide substrate PepTag C1 (P-L-S-R-T-L-S-V-A-A-K). The reaction was stopped by heating to 95°C for 10 min. Phosphorylation of the PKA- andPKC-specific substrates was used to measure kinase activity. Phosphorylated and unphosphorylatedPepTag peptides were separated on a 0.8% agarose gel by electrophoresis. The gel was photographedwith a transilluminator, and bands indicating substrate phosphorylated by PKA or PKC were quantifiedby densitometry (WinCam 2.2, Cybertech). The protein amount used for each assay was: PKA 4.5 µgand membrane-bound PKC 6.5 µg.

Western BlottingHippocampi of C57BL/6N or Balb/c mice were dissected out and homogenized in TBS (10

mM Tris, pH 7.6 and 150 mM NaCl), 10% sucrose and a protease inhibitor cocktail tablet (BoehringerMannheim). The homogenate was centrifuged at 20,000 x g for 30 min at 4°C. The supernatant was

Chapter 8

122

removed, and the membrane pellet was resuspended in a second identical wash step and centrifugedagain at 20,000 x g for 30 min at 4°C. The supernatant was removed and the membrane pellet wasresuspended in TBS, 1 mM EDTA, 1% sodium cholate and incubated for 60 min with constant mixingat 4°C. By centrifugation at 155,000 x g for 60 min (4°C) the supernatant containing soluble membraneproteins was obtained. Protein concentrations were determined with a Bradford assay (BioRad,München, Germany). Equal amounts of protein for each group were separated on a 10% SDS gel andtransferred to an Immobilon-P membrane using a semidry transfer apparatus. The blot was probedusing an anti-Gq/11α subunit antibody (1:4000; Calbiochem, San Diego, CA), an anti-Gsα subunitantibody (1:1000; NEN, Boston, MA) or an antibody directed against Gαi-1,2,3 protein (1:200;Calbiochem, San Diego, CA) and detected by secondary antibodies conjugated to alkaline phosphatase.As a chemiluminescence substrate CDP-Star ® (Tropix Inc., Bedford, MA) was used. Upondephosphorylation, the substrate decomposes, producing a prolonged emission of light that was imagedon photographic film (Fuji Super RX). The relative density of the bands was measured by densitometryusing the software WinCam 2.2 for Windows.

Preparation of Hippocampal MembranesMembranes were prepared as described (Grammatopoulos et al., 2001). Hippocampi of

C57BL/6N or Balb/c mice were homogenized in Dulbecco`s phosphate-buffered saline (extractionbuffer) containing 10 mM MgCl2, 2 mM EGTA, 1.5 g/l bovine serum albumin (BSA) (w/v), 0.15 mMbacitracin, 1 mM phenylmethylsulfonylfluoride (PMSF), pH 7.2 at 22°C. The homogenate wascentrifuged at 1,500 x g for 30 min at 4°C. The pellet was discarded and the supernatant spun at 45,000x g for 60 min at 4°C. The final pellet was resuspended in 10 ml of the described extraction bufferusing the homogenizer. The protein concentration of the membrane suspension was determined usingthe bicinchoninic acid method (Smith et al., 1985) with BSA as a standard.

Synthesis of 32P-GTP-γ-azidoanilide (32P-GTP-AA) and Photolabelling of Gα-subunits32P-GTP-AA was synthesized as previously described (Schwindinger et al., 1998). Mouse

hippocampal membranes were incubated in a darkroom with or without h/rCRF (100 nM) for 5 min at30°C before the addition of 5 µCi of 32P-GTP-AA in 120 µl of 50 mM HEPES buffer, pH 7.4,containing 30 mM KCl, 10 mM MgCl2, 1 mM benzamidine, 5 µM GDP, 0.1 mM EDTA. Afterincubation for 3 min at 30°C, membranes were collected by centrifugation and resuspended in 100 µlof the above buffer containing 2 mM glutathione, placed on ice and exposed to UV light (254 nm) at adistance of 5 cm for 5 min.

G Protein Immunoprecipitation32P-GTP-AA-labelled G proteins were precipitated by centrifugation and solubilized in 120 µl

of 2% SDS. Then, 360 µl of 10 mM Tris-HCl buffer, pH 7.4, containing 1% (v/v) Triton X-100, 1%(v/v) deoxycholate, 0.5% (w/v) SDS, 150 mM NaCl, 1 mM DTT, 1 mM EDTA, 0.2 mM PMSF, 10µg/ml aprotinin was added, and insoluble material was removed by centrifugation. Solubilizedmembranes were divided into 100 µl aliquots, and each aliquot was incubated with 10 µl of undilutedG protein antiserum at 4°C. Subsequently, 50 µl of protein A-Sepharose beads (10% w/v in the abovebuffer) was added, and the incubation was continued at 4°C overnight. The beads were collected bycentrifugation, washed twice and dried under vacuum. The immune complexes were dissociated fromprotein A by reconstitution in Laemmli`s buffer (100 µl) and boiling for 5 min. Samples were thensubjected to gel electrophoresis. The gels were stained with Coomassie Blue, dried and exposed to FujiX-ray film at –70°C for 2-5 d. The relative density of the bands was measured by optical densityscanning using the software Scion Image-Beta 3b for Windows (Scion Corporation, Frederick, MD).

StatisticsStatistical comparisons were made by using Student’s t-test and ANOVA. Data were

expressed as mean ± SEM. Significance was determined at the level of p < 0.05.


123

RESULTS

In both, C57BL/6N and Balb/c mice, stable intracellular recordings wereobtained from CA1 pyramidal neurons. The resting membrane potentials ofC57BL/6N mice (-68.4 ± 0.9 mV, n = 38) and Balb/c mice (-69.6 ± 1.2 mV, n = 33)did not differ significantly. Neither did the membrane input resistance of CA1 cellsof C57BL/6N mice (56.5 ± 3.5 MΩ, n = 38) and Balb/c mice (58.7 ± 3 MΩ, n = 33)differ from each other. Likewise, there were no significant differences between thespike amplitudes with values of 63.4 ± 1.2 mV (n = 19) found for C57BL/6N miceand 62.2 ± 1.3 mV (n = 13) for Balb/c mice.

When mouse CA1 pyramidal cells of either strain were excited by prolongeddepolarizing current pulses, they responded with prolonged spiking (Fig. 1A). Thedischarge rate was highest at the beginning of the current pulse (1 nA) and declinedto a steady rate during the course of the depolarizing pulse (Fig. 1B). Increasingstimulus intensities elicited enhanced neuronal spiking. In response to strongdepolarizing current pulses (1 nA, 600 ms), C57BL/6N and Balb/c mouse pyramidalcells fired 18.7 ± 2.5 (n = 12) and 18.6 ± 7.3 (n = 7) spikes, respectively (Fig. 1C).

Figure 1. A, Representative intracellular recordings from CA1 pyramidal neurons inhippocampal slices from C57BL/6N mice and Balb/c mice showing responses to 600ms depolarizing current pulses. B, Number of spikes elicited in 100 ms fragmentsduring a single depolarizing current pulse (600 ms, 1 nA). C, Plot of the number ofspikes elicited by a 600 ms depolarizing pulse vs. stimulus intensity.

C57BL/6N

1 nA

100 ms

10 mV

Balb/c

-65 mV

A

B

Time during depol. (s)0

0

2.5

5.0

7.5

10

0.2 0.4 0.6

Balb/c

C57BL/6NC

Stim. intensity (nA)0 0.2 0.6 1

0

510

15

20

25

Num

ber

of s

pike

s

Chapter 8

124

h/rCRF was added to mouse hippocampal slices to investigate the effects onthe neuronal spiking behavior. The number of spikes elicited by a 600 msdepolarizing current pulse was increased by 88 ± 24% (n = 7; p < 0.05) inC57BL/6N (data not shown) mice and by 87 ± 39% (n = 8; p < 0.05) in Balb/c micewhen compared to controls (Fig. 2A). In CA1 hippocampal neurons from bothmouse strains, the h/rCRF effect was antagonized by the CRF receptor antagonist[Glu 11,16]astressin (Fig. 2B).

Figure 2. Responses ofBalb/c mouse CA1pyramidal cells to 600m s d e p o l a r i z i n gcurrent pulses beforeand 20 min afterapplication of h/rCRF(250 nM) over aperiod of 10 min (A)and before and 20 minafter coapplication ofh/rCRF (250 nM) and[Glu 11,16]astressin (1µM) over a period of10 min (B ). Pulseintensity was keptconstant during eachexperiment; holdingpotential –65 mV.

In subsequent experiments, we investigated the underlying second-messenger pathways activated by h/rCRF to increase the neuronal excitability inmouse hippocampus. When slices were preincubated with the selective and cell-permeable PKA inhibitor H-89, the firing rate of hippocampal neurons fromC57BL/6N mice was not significantly changed by h/rCRF (n = 6; p = n.s.; Fig. 3A).In contrast, following the H-89 treatment h/rCRF still enhanced the neuronal activityof hippocampal neurons from Balb/c mice by 55 ± 10% (n = 5; p < 0.05; Fig. 3B).When hippocampal slices from Balb/c mice were preincubated with BIS-I, a highlyselective cell-permeable protein kinase C (PKC) inhibitor, subsequent h/rCRFapplication had no modulatory effect on the neuronal firing rate (n = 5; p = n.s.; Fig.3D). In contrast, following BIS-I treatment, h/rCRF application still enhancedneuronal spiking in CA1 cells from C57BL/6N mice by 52 ± 9% (n = 6; p < 0.05;Fig. 3C).

Under basal conditions, PKA activity, as measured by the phosphorylatedstate of a PKA-specific target peptide, was lower in hippocampal brain slices fromC57BL/6N mice in comparison to hippocampal brain slices from Balb/c mice.

Control h/r CRF

1 nA 10 mV

100 ms

[Glu 11,16]astressin +h/r CRF

ControlB


125

Following h/rCRF treatment, PKA activity in hippocampal slices fromC57BL/6N mice was increased, whereas it was decreased in hippocampal slicesfrom Balb/c mice when compared to the corresponding PKA activities in controlslices (Fig. 4A). Since membrane translocation of PKC is considered to be anindicator of PKC activation (Kraft and Anderson, 1983), we assayed PKC activity inthe membrane-bound fraction of hippocampal slice homogenates. After h/rCRFincubation of slices, PKC activity was only apparent in hippocampal slices of Balb/cmice (Fig. 4B) with no detectable PKC activity in hippocampal slices of C57BL/6Nmice.

Figure 3. Effect of the PKC inhibitor BIS-I and of the PKA inhibitor H-89 on h/rCRF-mediated modulation of excitability. Representative recordings in CA1 pyramidalcells from (A, C) C57BL/6N and (B, D) Balb/c mice showing the effect of 250 nMh/rCRF applied over a period of 20 min after preincubation with BIS-I (1.2 µM, 1 hr)or H-89 (10 µM, 3 hr). Pulse intensity was kept constant during each experiment.

The observed differences in the activation of second-messenger pathwaysfollowing h/rCRF application can be attributed to variations in the abundance of Gproteins. Thus, we examined the levels of Gs, Gi, and Gq/11 proteins in thehippocampus of both mouse inbred strains. Using immunoblots we did not observeany significant differences in the amount of G proteins (Fig. 5). To furtherinvestigate which G proteins were activated following h/rCRF application,hippocampal membrane suspensions were analyzed for the ability of h/rCRF tostimulate incorporation of 32P-GTP-AA into various G protein α-chains. Inhippocampal membranes of C57BL/6N mice, h/rCRF induced activation of Gs, Gi

and Gq/11 with an order of potency Gs>Gq/11>Gi, whereas in hippocampal membranesof Balb/c mice only stimulation of Gq/11 was detectable after h/rCRF treatment (Fig.6)

D Balb/c

1 nA 10 mV

100 ms-65 mV

BIS-I + h/rCRFBIS-IC

BIS-I + h/rCRF

C57BL/6N

BIS-I

H-89 + h/rCRF

C57BL/6N

H-89

A Balb/cH-89 + h/rCRFH-89

B

Chapter 8

126

Figure 4. PKA and PKCactivity in hippocampalslices of C57BL/6N andBalb/c mice. Hippocampalslices were incubated ineither 250 nM h/rCRF (30min) or aCSF (30 min ascontrol). Partially purifiedhomogenates of these slices(n = 11) from six animalswere tested for the ability tophosphorylate a PKA-specific (L-R-R-A-S-L-G,Kemptide) (A ) or a PKC-specific peptidic substrate(P-L-S-R-T-L-S-V-A-A-K)(B ) in a nonradioactiveassay. Identical amounts ofprotein were used for eachsample.

Figure 5. Expression of Gs, Gi and Gq/11

in hippocampal membrane fractions ofC57BL/6N and Balb/c mice. The bargraph summarizes Western blot data(mean ± SEM) of three independentexperiments each with five animals permouse strain.

To further delineate the impact of the observed different h/rCRF-mediatedsignaling pathways on learning and memory mice were subjected to contextual fearconditioning, a hippocampus-dependent associative learning paradigm (Kim andFanselow, 1992; Phillips and LeDoux, 1992; Phillips and LeDoux, 1994). WhenBalb/c mice received a bilateral h/rCRF-injection i.c.v. (n = 7; Fig. 7A) and weretrained 2 hr following the injection, contextual fear was significantly enhancedcompared to naive (P < 0.05; n = 9; Fig. 7A) and vehicle treated animals (P < 0.01; n= 30; Fig. 7A). This h/rCRF effect was prevented by either [Glu 11,16]astressin (n = 7)or BIS-I (n = 7). Both compounds had no effect when applied alone (Fig. 7A). InC57BL/6N mice freezing was not significantly changed when h/rCRF was injected 2hr (n = 15; p = n.s.; Fig. 7B) before the training session. However, injection of the

PhosphorylatedPKA-specific peptide

A

PhosphorylatedPKC-specific peptide

PKA Assay

B PKC Assay

Balb/c, control

Balb/c, h/rCRF

C57BL/6N, h/rCRF

C57BL/6N, control

Gi GsGq/110

20

40

60

80

100

120 Balb/cC57BL/6N


127

potent PKC-activator PDBu 2 hr before the training (n = 9) significantly enhancedcontextual fear compared to naive (p < 0.05; n = 9; Fig. 7B) and vehicle treatedanimals (p < 0.05; n = 27; Fig. 7B). There was no significant change of contextualfear after injection of the inactive isomer 4α−phorbol (p = n.s.; n = 4; Fig. 7B).

Figure 6. Autoradiograph of h/rCRF-induced photolabelling of G α-subunit subtypesfrom hippocampal membranes of C57BL/6N (n = 30) and Balb/c (n = 30) mice.Membranes were incubated with 32P-GTP-AA in the presence and absence of h/rCRF(100 nM), followed by UV crosslinking and immunoprecipitation of the G α-subunitsubtypes using specific antibodies. Proteins were resolved by SDS-PAGE, followed byautoradiographic visualization.

DISCUSSION

In this study, we provide evidence that signal processing of h/rCRF inmouse hippocampus can be mediated through two different signal transductionpathways. h/rCRF increased CA1 hippocampal neuronal activity via PKC in thehippocampus of Balb/c mice and via PKA in C57BL/6N mice. Hippocampus-dependent learning evaluated by context-dependent fear conditioning was onlyimproved in Balb/c mice after h/rCRF injection with no effect in C57BL/6N mice.Western blots from mouse hippocampal membrane proteins showed identicalamounts of the relevant G protein subunits in both mouse strains. However, in thehippocampus of Balb/c mice application of h/rCRF was accompanied by activationof Gq/11, whereas h/rCRF activated Gs, Gq/11 and Gi in the hippocampus of C57BL/6Nmice. All of the known effects of CRF in the rat hippocampus involve receptor-coupled activation of Gs, adenylate cyclase and an increase in cellular levels ofcAMP (Battaglia et al., 1987; Chen et al., 1986; Haug and Storm, 2000; Pihoker etal., 1992), which is in agreement with our findings in hippocampi of C57BL/6Nmice.

0

1

2

3

4

Gs Gq/11 Gi

fo

ld in

crea

seab

ove

basa

l OD

-uni

ts

C57BL/6NBalb/c

A BC57BL/6N Balb/c

Basal

h/rCRF

Basal

h/rCRF

Gs

Gq/11

Gi

Chapter 8

128

Figure 7. Effect of h/rCRF on context-dependent fear conditioning of (A) Balb/c and(B) C57BL/6N mice, injected with aCSF, h/rCRF, [Glu 11,16]astressin, PDBu or 4α-phorbol 2 hr before the training as indicated. For combined treatment [Glu11,16]astressin and BIS-I were given 15 min before h/rCRF application. Freezing wasmeasured in the retention test performed 24 hr after training. Statistically significantdifferences: * p < 0.05 vs. vehicle-injected animals and naives.

In recent studies, however, it was reported that h/rCRF also activates thephospholipase C (PLC)/PKC-pathway in rat Leydig cells (Ulisse et al., 1990), incultured rat astrocytes (Takuma et al., 1994), in rat cerebellum (Miyata et al., 1999)and in rat cerebral cortex (Grammatopoulos et al., 2001). In addition, Malenka andcolleagues (1986) reported that activation of PKC markedly reduces accommodationof neuronal spiking in rat hippocampal pyramidal cells. In combination, both aspectssuggest that in Balb/c mice Gq/11–dependent PKC activation was responsible for theh/rCRF-mediated increase in neuronal activity. Surprisingly, PKA activity inhippocampal slices from Balb/c mice was reduced upon application of h/rCRF. Thiseffect might be initiated by Gq/11 stimulation, which is accompanied by an increase inG protein βγ-subunits. These subunits can inhibit type I adenylyl cyclase and thusPKA activity directly (Taussig et al., 1993). Aside from Gs and Gi activation, whichmay stimulate adenylyl cyclase via synergistic interaction (Federman et al., 1992),we also observed that h/rCRF activated Gq/11 in the hippocampus of C57BL/6N mice,allowing the stimulation of the PLC/PKC-pathway. However, in the membrane-bound fraction of hippocampal slice homogenates prepared from C57BL/6N miceno PKC activity was detected following h/rCRF application and we did not detectany significant contribution of PKC to the h/rCRF-mediated increase in neuronalspiking behavior of CA1 pyramidal cells.

This discrepancy might result from the fact that h/rCRF was applied tohippocampal membrane preparation when Gq/11 activation was observed. In contrast,

h/rCRF

4α−pho

rbol

h/rCRF

naive

PDBu

vehic

le

*C57BL/6NB

010203040

50607080

0

1020304050607080

Free

zing

Balb/c

astre

ssin

A

astr-

CRF

BIS -I

BIS-C

RFna

ive

vehic

le

*


129

for the PKC assay and the electrophysiological recordings h/rCRF was applied tointact hippocampal cells. Several lines of evidence suggest that due tocompartmentalization of G protein subtypes into distinct pools, receptors do nothave free access in intact cells to all G proteins with which they are capable ofcoupling in reconstituted systems (Neubig, 1994). Thus, complex interactions ofreceptors and G proteins with their effectors and the properties of the membrane,possibly including the resting membrane potential, appear to play an important rolein the receptor coupling characteristics. An additional possible explanation formainly detecting activation of PKA in the hippocampus of C57BL/6N mice uponapplication of h/rCRF is that receptors with dual signaling properties often stimulatedifferent pathways with different efficacies. A3 adenosine receptors, for example,interact with Gi proteins and, to a lesser extent, with Gq/11 proteins in CHO cells(Palmer et al., 1995). These receptors were shown to inhibit adenylyl cyclase in allcell types tested, whereas stimulation of PLC was cell type dependent. From thispoint of view it might be speculated that hippocampal CRF receptors in C57BL/6Nmice can in fact couple to the Gq/11/PLC/PKC-pathway in intact neurons. However,the potential stimulation of PKC by h/rCRF might not have been sufficient toenhance the conditioned fear response in C57BL/6N mice as observed after PKC-activation by PDBu. In the hippocampus of Balb/c mice h/rCRF stimulated theGq/11/PLC/PKC cascade and improved hippocampus-dependent learning. Similarresults were reported by Fordyce et al. (1995) who described that stimulation ofhippocampal PKC activity enhances contextual learning as determined by the fearconditioning task in DBA mice. In mouse hippocampus, a PKA-dependent periodfor contextual memory consolidation develops between 1 hr and 3 hr after training(Bourtchouladze et al., 1998). Considering the activation of the PKA system in thehippocampus of C57BL/6N mice upon h/rCRF application, it seems surprising thath/rCRF exhibited no facilitating effect on contextual fear conditioning in C57BL/6Nmice. In a recent study, the crucial temporal relationship between PKA inhibitionand training necessary to produce impairment of fear memory consolidation wasdemonstrated (Bourtchouladze et al., 1998). PKA-inhibitor-injected animals showsignificantly less freezing, when injected immediately after the training. However,inhibition of PKA has no effect on contextual memory when given at 1, 4, 6, 8, or23.5 hr after training. A similar narrow time window exists for PKA inhibitionbefore the training. When mice are PKA inhibitor-treated 20-30 min beforecontextual conditioning they show dramatic amnesia. However, inhibition of PKA 3hr before training does not affect retention at 24 hr after training. Thus, in thepresent study h/rCRF might have had no effect on long-term contextual memory inC57BL/6N mice because PKA was not activated within the decisive time window.

To summarize, in mouse hippocampus we have demonstrated that h/rCRFactivated at least two different signaling cascades, the PLC/PKC pathway (viainteraction with Gq/11) and the cAMP/PKA pathway (via interaction with Gs, Gq/11

and Gi). Future experiments will have to determine whether hippocampal CRFreceptors bear two independent G protein-specific interaction domains which allowthe coupling of one receptor to different G protein subunits. Alternatively, theobserved multi-signaling activity of h/rCRF might be caused by the activation ofdifferent hippocampal CRF receptors.

Chapter 8

130

In view of the contribution of the hippocampus to numerous forms oflearning (reviewed in: Kesner et al., 2000; Kim and Baxter, 2001; Maren, 2001) andthe fact that h/rCRF represents an early signal in the neuroendocrine response tostress (Koob and Bloom, 1985), our present findings may be an important steptoward understanding the cellular and molecular processes underlying interstrainvariability concerning the impact of stress on learning and memory (Brush et al.,1988; Francis et al., 1995; Palmer and Printz, 1999).

ACKNOWLEDGMENTS

We thank Dr. Klaus Eckart and Dr. Andreas Rühmann for the peptidesynthesis of [Glu 11,16]astressin and h/rCRF.

REFERENCES

Aldenhoff JB, Gruol DL, Rivier J, Vale W and Siggins GR (1983) Corticotropin releasing factordecreases postburst hyperpolarizations and excites hippocampal neurons. Science 221: 875-877.

Bale TL, Contarino A, Smith GW, Chan R, Gold LH, Sawchenko PE, Koob GF, Vale WW and Lee KF(2000) Mice deficient for corticotropin-releasing hormone receptor-2 display anxiety-like behaviourand are hypersensitive to stress. Nat. Genet. 24: 410-414.

Battaglia G, Webster EL and De Souza EB (1987) Characterization of corticotropin-releasing factorreceptor-mediated adenylate cyclase activity in the rat central nervous system. Synapse 1: 572-581.

Blank T, Nijholt I, Eckart K and Spiess J (2002) Priming of long-term potentiation in mousehippocampus by corticotropin-releasing factor and acute stress: implications for hippocampus-dependent learning. J. Neurosci. in press.

Beuzen A and Belzung C (1995) Link between emotional memory and anxiety states: A study byprincipal component analysis. Physiol. Behav. 58: 111-118.


Bolles RC and Riley AL (1973) Freezing as an avoidance response: another look at the operant-respondent distinction. Learn. Motiv 4: 268-275.

Bourtchouladze R, Abel T, Berman N, Gordon R, Lapidus K and Kandel ER (1998) Different trainingprocedures recruit either one or two critical periods for contextual memory consolidation, each ofwhich requires protein synthesis and PKA. Learn. Mem. 5: 365-374.

Brush FR, Del Paine SN, Pellegrino LJ, Rykaszewski IM, Dess NK and Collins PY (1988) CERsuppression, passive-avoidance learning, and stress-induced suppression of drinking in the Syracusehigh- and low-avoidance strains of rats (Rattus norvegicus). J. Comp. Psychol. 102: 337-349.


131

Chang CP, Pearse RV, O’Connell S and Rosenfeld MG (1993) Identification of a seven transmembranehelix receptor for corticotropin-releasing factor and sauvagine in mammalian brain. Neuron 11: 1187-1195.

Chen FM, Bilezikjian LM, Perrin MH, Rivier J and Vale W (1986) Corticotropin releasing factorreceptor-mediated stimulation of adenylate cyclase activity in the rat brain. Brain Res. 381: 49-57.

Chen R, Lewis KA, Perrin MH and Vale WW (1993) Expression cloning of a human corticotropin-releasing-factor receptor. Proc. Natl. Acad. Sci. USA 90: 8967–8971.

Coste SC, Kesterson RA, Heldwein KA, Stevens SL, Heard AD, Hollis JH, Murray SE, Hill JK,Pantely GA, Hohimer AR, Hatton DC, Phillips TJ, Finn DA, Low MJ, Rittenberg MB, Stenzel P andStenzel-Poore MP (2000) Abnormal adaptations to stress and impaired cardiovascular function in micelacking corticotropin-releasing hormone receptor-2. Nat. Genet. 24: 403-409.



Federman AD, Conklin BR, Schrader KA, Reed RR and Bourne HR (1992) Hormonal stimulation ofadenylyl cyclase through Gi-protein beta gamma subunits. Nature 356: 159-161.

Fordyce DE, Clark VJ, Paylor R and Wehner JM (1985) Enhancement of hippocampally-mediatedlearning and protein kinase C activity by oxiracetam in learning-impaired DBA/2 mice. Brain Res. 672:170-176.

Francis DD, Zaharia MD, Shanks N and Anisman H (1995) Stress-induced disturbances in Morriswater-maze performance: interstrain variability. Physiol. Behav. 58: 57-65.


Grammatopoulos DK, Randeva HS, Levine MA, Kanellopoulou KA and Hillhouse EW (2001) Ratcerebral cortex corticotropin-releasing hormone receptors: evidence for receptor coupling to multipleG-proteins. J. Neurochem. 76: 509-519.

Haug T and Storm JF (2000) Protein kinase A mediates the modulation of the slow Ca (2+)-dependentK(+) current, I(sAHP), by the neuropeptides CRF, VIP, and CGRP in hippocampal pyramidal neurons.J. Neurophysiol. 83: 2071-2079.

Hollrigel GS, Chen K, Baram TZ and Soltesz I (1998) The pro-convulsant actions of corticotropin-releasing hormone in the hippocampus of infant rats. Neuroscience 84: 71-79.

Kesner RP, Gilbert PE and Wallenstein GV (2000) Testing neural network models of memory withbehavioral experiments. Curr. Opin. Neurobiol. 10: 260-2655.

Kim JJ and Baxter MG (2001) Multiple brain-memory systems: the whole does not equal the sum of itsparts. Trends Neurosci. 24: 324-330.


Chapter 8

132

Kishimoto T, Radulovic J, Radulovic M, Lin CR, Schrick C, Hooshmand F, Hermanson O, RosenfeldMG and Spiess J (2000) Deletion of crhr2 reveals an anxiolytic role for corticotropin-releasinghormone receptor-2. Nat. Genet. 24: 415-419.

Koob GF and Bloom FE (1985) Corticotropin-releasing factor and behavior. Fed. Proc. 44: 259-263.

Kraft AS and Anderson WB (1983) Phorbol esters increase the amount of Ca2+, phospholipid-dependent protein kinase associated with plasma membrane. Nature 301: 621-623.

Lee EH, Lee CP, Wang HI and Lin WR (1993) Hippocampal CRF, NE, and NMDA systeminteractions in memory processing in the rat. Synapse 14: 144-153.

Lovenberg TW, Liaw CW, Grigoriadis DE, Clevenger W, Chalmers DT, De Souza EB and OltersdorfT (1995) Cloning and characterisation of a functionally distinct corticotropin-releasing factor receptorsubtype from rat brain. Proc. Natl. Acad. Sci. USA 92: 836-840.

Malenka RC, Madison DV, Andrade R and Nicoll RA (1986) Phorbol esters mimic some cholinergicactions in hippocampal pyramidal neurons. J. Neurosci. 6: 475-480.

Maren S (2001) Neurobiology of Pavlovian fear conditioning. Annu. Rev. Neurosci. 24: 897-931.

Merchenthaler I (1984) Corticotropin releasing factor (CRF)-like immunoreactivity in the rat centralnervous system. Extrahypothalamic distribution. Peptides 5: 53-69.

Miyata M, Okada D, Hashimoto K, Kano M and Ito M (1999) Corticotropin-releasing factor plays apermissive role in cerebellar long-term depression. Neuron 22: 763-775.

Neubig RR (1994) Membrane organization in G-protein mechanisms. FASEB J. 8: 939–946.

Nowakowski RS (1984) The mode of inheritance of a defect in lamination in the hippocampus ofBALB/c mice. J. Neurogenet. 1: 249-258.

Oliverio A, Eleftheriou RS and Bailey DW (1973) A gene influencing active avoidance performance inmice. Physiol. Behav. 11: 497-501.

Palmer TM, Gettys TW and Stiles GL (1995) Differential inter-action with and regulation of multipleG-proteins by the rat A3 adenosine receptor. J. Biol. Chem. 270: 16895–16902.

Palmer AA and Printz MP (1999) Strain differences in Fos expression following airpuff startle inSpontaneously Hypertensive and Wistar Kyoto rats. Neuroscience 89: 965-978.

Peeler DF and Nowakowski RS (1987) Genetic factors and the measurement of exploratory activity.Behav. Neural Biol. 48: 90-103.

Phillips RG and LeDoux JE (1992) Differential contribution of amygdala and hippocampus tocontextual and cued fear conditioning. Behav. Neurosci. 106: 274-285.


Pihoker C, Cain ST and Nemeroff CB (1992) Postnatal development of regional binding ofcorticotropin-releasing factor and adenylate cyclase activity in the rat brain. P r o g .Neuropsychopharmacol. Biol. Psychiatry 16: 581-586.

Radulovic J, Rühmann A, Liepold T and Spiess J (1999) Modulation of learning and anxiety bycorticotropin-releasing factor (CRF) and stress: differential roles of CRF receptors 1 and 2. J. Neurosci.19: 5016-5025.


133

Rühmann A, Köpke AKE, Dautzenberg FM and Spiess J (1996) Synthesis and characterization of aphotoactivatable analog of corticotropin-releasing factor for specific receptor labeling. Proc. Natl.Acad. Sci. USA 93: 10609-10613.

Schwindinger WF, Fredericks J, Watkins L, Robinson H, Bathon JM, Pines M, Suva LJ and LevineMA (1998) Coupling of the PTH/PTHrP receptor to multiple G-proteins. Direct demonstration ofreceptor activation of Gs, Gq/11, and Gi(1) by [alpha-32P]GTP-gamma-azidoanilide photoaffinitylabeling. Endocrine 8: 201-209.

Smith GW, Aubry JM, Dellu F, Contarino A, Bilezikjian LM, Gold LH, Chen R, Marchuk Y, HauserC, Bentley CA, Sawchenko PE, Koob GF, Vale W and Lee KF (1998) Corticotropin releasing factorreceptor 1-deficient mice display decreased anxiety, impaired stress response, and aberrantneuroendocrine development. Neuron 20: 1093-1102.

Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, GoekeNM, Olson BJ and Klenk DC (1985) Measurement of protein using bicinchoninic acid. Anal. Biochem.150: 76-85.



Swanson LW, Sawchenko PE, Rivier J and Vale WW (1983) Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study.Neuroendocrinology 36: 165-186.

Takuma K, Matsuda T, Yoshikawa T, Kitanaka J, Gotoh M, Hayata K and Baba A (1994)Corticotropin-releasing factor stimulates Ca2+ influx in cultured rat astrocytes. Biochem. Biophys. Res.Commun. 199: 1103-1107.

Taussig R, Quarmby LM and Gilman AG (1993) Regulation of purified type I and type IIadenylylcyclases by G protein beta gamma subunits. J. Biol. Chem. 268: 9-12.

Timpl P, Spanagel R, Sillaber I, Kresse A, Reul JM, Stalla GK, Blanquet V, Steckler T, Holsboer F andWurst W (1998) Impaired stress response and reduced anxiety in mice lacking a functionalcorticotropin-releasing hormone receptor 1. Nat. Genet. 19: 162-166.

Ulisse S, Fabbri A, Tinajero JC and Dufau ML (1990) A novel mechanism of action of corticotropinreleasing factor in rat Leydig cells. J. Biol. Chem. 265: 1964-1971.

Vale W, Spiess J, Rivier C and Rivier J (1981) Characterization of a 41-residue ovine hypothalamicpeptide that stimulates secretion of corticotropin and beta-endorphin. Science 213: 1394-1397.


Wang HL, Tsai LY and Lee EH (2000) Corticotropin-releasing factor produces a protein synthesis--dependent long-lasting potentiation in dentate gyrus neurons. J. Neurophysiol. 83: 343-349.

Wang HL, Wayner MJ, Chai CY and Lee EH (1998) Corticotropin-releasing factor produces a long-lasting enhancement of synaptic efficacy in the hippocampus. Eur. J. Neurosci. 10: 3428-3437.

SUMMARY AND GENERAL

DISCUSSION

Synaptic plasticity is a feature of the brain, which allows neuronal structureand function to accommodate to patterns of electrical activity. Synaptic plasticityproceeds through different stages in time: short-term effects are due to post-translational modifications like phosphorylation of existing proteins, whileconsolidation of the process leading to a long-lasting change in synaptic efficiencyand structural modifications requires and depends on a coordinated program ofchanges in gene transcription and synthesis of new proteins.

The magnitude and kinetics of calcium concentrations at the synapse arethought to be major determinants of short- and long-term effects on synapticefficacy (Abraham and Bear, 1996; Kauer et al., 1988; Lisman, 1989). The N-methyl-D-aspartate (NMDA) subtype of glutamate receptors in the mammalian brainis linked to many downstream signal transducing pathways in the neuron through itsCa2+ permeability. Although the NMDA receptor is widely distributed in the brain,it can be speculated that the control of its signaling characteristics differs in a brainregion-specific manner in order to fulfil diverse demands.

In this thesis we first addressed the intriguing question of how NMDAreceptor signaling can be modulated in a brain region-specific fashion and thuscontribute to long- and short-lasting episodes of synaptic plasticity.

Changes in synaptic plasticity have been suggested to be the basis of certaintypes of learning and memory. Therefore, in the final part of this thesis weinvestigated the relation between stress- and CRF-induced alterations in synapticstrength and changes in learning and memory.

1. Brain regional differences in the modulation of the NMDA receptors byphosphorylation events

In the first Chapters of this thesis we investigated how phosphorylationevents can modulate NMDA receptor function. Using the oocyte expression systemwe found that the function of NMDA receptors expressed from mRNA isolated fromvarious brain regions could be modulated by protein kinase C (PKC) activationwhereas NMDA receptor modulation by adenosine-3',5'-cyclic monophosphate(cAMP) dependent protein kinase (PKA) shows strong regional differences.Activation of PKC was shown to increase NMDA receptor function from total brain,hippocampus, striatum, cortex and hypothalamus, whereas PKA could only enhancethe function of NMDA receptors in striatum and hypothalamus. This modulation ofhypothalamic and striatal NMDA receptors by PKA was investigated in more detailin Chapters 2 and 3. It appeared that the increase in NMDA receptor signaling was

Chapter 9

136

not due to direct phosphorylation of the receptor by PKA but rather that PKA causedthe inhibition of protein phosphatases via activation of a third messenger. Inhibitionof protein phosphatases may result in a prolonged phosphorylation state of theNMDA receptor and thereby increase NMDA receptor function. In the striatum thisthird messenger could be identified as the phosphoprotein dopamine- and cAMP-regulated phosphoprotein with an apparent Mr 32000 (DARPP-32). Thus, activationof PKA leads to the phosphorylation and thereby activation of DARPP-32. DARPP-32 acts as an inhibitor of protein phosphatase 1 (PP1) upon activation. Consideringthe important role of NMDA receptor function in synaptic plasticity, these regionaldifferences in NMDA receptor modulation by protein kinases and phosphatases mayhave a strong impact on regional differences in the contribution of NMDA receptorsignaling to synaptic plasticity.

1.1 DARPP-32In view of the central role of DARPP-32 in NMDA receptor function,

region-specific distribution of phosphoproteins like DARPP-32 might underlie thebrain regional differences in NMDA receptor signaling. Indeed, DARPP-32 wasfound at particularly high levels in the substantia nigra, olfactory tubercle, bednucleus of the stria terminalis, globus pallidus, caudato putamen, nucleusaccumbens, portions of the amygdaloid complex, and corticothalamic neurons(Ouimet et al., 1984). In these brain areas DARPP-32 can participate in the controlof the phosphorylation-state of NMDA receptors thereby controlling the excitabilityof the neurons and the efficacy of synaptic transmission.

On the other hand, stimulation of NMDA receptors in rat striatal slices hasbeen shown to induce dephosphorylation of phosphorylated DARPP-32, presumablyby the calcium-dependent phosphatase calcineurin (Halpain et al., 1990). Thismechanism might reflect a negative feedback-loop protecting striatal cells fromNMDA receptor-mediated toxicity in the presence of high amounts ofphosphorylated DARPP-32 (Colwell et al., 1996).

The inhibition of PP1 by DARPP-32 does not only modulate thephosphorylation state of the NMDA receptor but does also affect thephosphorylation state of various other proteins, including voltage-dependent Na+channels, L-, N-, P-type voltage-dependent Ca2+ channels, the electrogenic ionpump Na+, K+-ATP-ase, and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionicacid (AMPA) and γ-aminobutyric acid (GABA) type A receptors (Flores-Hernandezet al., 2000; Greengard et al., 1998; Yan et al., 1999b). Thus DARPP-32 may play ageneral role in coordinating the excitability of the cells. A major step inunderstanding the role of DARPP-32 in these processes has been achieved by thecharacterization of mice with DARPP-32 deficiency (Fienberg et al., 1998). Theresults obtained with these mice demonstrate an obligatory role for DARPP-32 indopaminergic neurotransmission (Fienberg and Greengard, 2000).

Besides DARPP-32, several other phosphoproteins with apparent equivalentfunctions are present in the mammalian brain. Similar to DARPP-32, inhibitor-1 canalso inhibit PP1 upon activation by PKA however inhibitor-1 shows a moreubiquitous distribution in the brain (Gustafson et al., 1991; Hemmings et al., 1992;MacDougall et al., 1989). This raises the question whether these phosphoproteins

Summary and general discussion

137

play different roles in the regulation of neuronal activity in the brain. Besides thedistinct regional distribution in some specific neuronal structures like neocortex,hippocampus and amygdala (Barbas et al., 1993; Gustafson et al., 1991), DARPP-32and inhibitor-1 may be expressed in different types of neurons in those areas whereboth are present. In support of this suggestion is the finding that in the cerebellumDARPP-32 is mainly found in the Purkinje cells and inhibitor-1 in granule cells(Alder and Barbas, 1995). Furthermore, a distinctive feature of the primary structureof DARPP-32 is the existence of several clusters of acidic residues (Williams et al.,1986). Therefore, DARPP-32 can be phosphorylated by casein kinase I and II incontrast to inhibitor-1, which can not be modulated by casein kinases (Desdouits etal., 1995; Girault et al., 1989). Phosphorylation at the casein kinase I site inhibitsdephosphorylation of DARPP-32 by calcineurin (Desdouits et al., 1995) presentingan additional way to modulate DARPP-32.

Even in those areas where either inhibitor-1 or DARPP-32 is present, thecontribution of the phosphoprotein to synaptic plasticity may be synapse-selective.A striking example was found in mice lacking inhibitor-1 (Allen et al., 2000). Inwildtype animals long-term-potentiation (LTP) could be induced in both the lateralperforant pathway (lpp) which makes excitatory synaptic contact onto dentategranule cell dendrites in the outer third, and the medial perforant pathway (mpp)which forms synapses in the middle third of the dentate molecular layer. Inhibitor-1is present in both pathways. However, in the inhibitor-1 deficient mice there was aprominent difference in the ability to produce LTP in these two pathways. LTPproduction was normal in the mpp but LTP in the lpp was significantly reduced. Onecan conclude that inhibitor-1 is not required in the mpp but is important in the lpp.Thus, even within a particular brain area there may be heterogeneous molecularmechanisms leading to changes in synaptic plasticity.

This heterogeneity of underlying molecular mechanisms is supported by ourstudy. NMDA responses recorded from oocytes injected with rat hippocampalpoly(A+) mRNA were not affected by stimulation of PKA, indicating that theamount of inhibitor-1, which is present in hippocampus, was not sufficientlyabundant to mediate the effect of forskolin, a PKA activator, on NMDA responses.However, when oocytes were injected with hippocampal poly(A+) mRNA pluscomplementary RNA (cRNA) coding for DARPP-32 (resulting in a greatly elevatedlevel of this PP1 inhibitor), NMDA responses were potentiated after stimulation ofPKA.

In conclusion, we demonstrated that DARPP-32 is a crucial element in theregion-specific modulation of synaptic plasticity. A more complete understanding ofthe functional role of DARPP-32 in the forebrain is important while thedopaminergic system is a key element in the control of movement (Nestler, 1994), inthe neuronal circuits that control motivation (Salamone, 1991) and the regulation ofmemory and attention (Williams and Goldman-Rakic, 1995). In addition,imbalances in dopaminergic and glutamatergic synaptic transmission have beenimplicated in several neurological disorders, including Parkinson’s disease,Huntington’s disease, addictive behavior and schizophrenia (Albin et al., 1989; Beal,1992; Davis et al., 1991; Nestler, 1994).

Chapter 9

138

1.2 Possible physiological relevance of PKA-mediated NMDA receptor modulationin the hypothalamus

Interestingly, in parallel to our finding that NMDA receptor function in thehypothalamus was enhanced upon phosphorylation by PKA, behavioral studiesreport that either activation of NMDA receptors containing NR2A and/or NR2B, oractivation of PKA within the lateral and perifornical hypothalamus stimulates eatingbehavior in rats (Stanley, 1996; Stanley et al., 1993a,b, 1997; Gillard et al., 1997,1998a,b; Khan et al., 1999). On the basis of these and our findings, it may bespeculated that phosphorylation of the NMDA receptor by PKA may contribute toalterations in both food intake and body weight.

1.3 Direct versus indirect phosphorylation of the NMDA receptorIn Chapter 3 it was described that NMDA-induced currents recorded from

Xenopus oocytes injected with cRNA encoding the NMDA receptor subunits NR1,NR2A, and/or NR2B, were not affected by PKA activation but were increased byPKC stimulation. NMDA receptors contain multiple consensus sites forphosphorylation by PKA and PKC (Hollmann and Heinemann 1994). NR1, NR2Aand NR2B can be directly phosphorylated by PKA and various PKC isoforms invitro (Leonard and Hell, 1997; Tingley et al. 1993, 1997). Interestingly, our findingthat PKA did not modulate NMDA receptor activity in oocytes injected with cRNAencoding NMDA receptor subunits, suggests that direct phosphorylation of thereceptor subunits by PKA may not be responsible for elevated NMDA receptoractivity. Direct phosphorylation of the receptor by PKA could instead play a role inother processes like subunit assembly and receptor targeting (Crump et al., 2001).

PKC has been shown to substantially modulate physiological properties ofthe NMDA receptor (Chapters 2 and 3; Chen and Huang 1992; Xiong et al., 1998).However, the relation of PKC-induced phosphorylation of the NR1 and NR2subunits to PKC-induced potentiation of NMDA receptor activity remains unclear.PKC was shown to affect currents through the NMDA receptor channels by directaction on phosphorylation sites in the NR2B C-terminus (Liao et al., 2001).However, several other studies report that PKC-mediated potentiation of NMDAreceptor activity is not necessarily due to direction phosphorylation of the receptor(Yamakura et al., 1993; Sigel et al., 1994; Zheng et al., 1999). Experimentsinvolving various truncation mutants revealed the unexpected finding that NMDAreceptors assembled from subunits lacking all known sites of PKC phosphorylationcan still show PKC potentiation (Zheng et al., 1999). These results indicate thatPKC-induced potentiation of NMDA receptor activity may not only occur by directphosphorylation of the receptor protein but may also imply associated targeting,anchoring, or signaling protein(s) to increase open probability and/or insertion ofnew channels in the plasma membrane. Like PKA, phosphorylation of receptors byPKC could also play a role in processes such as subcellular distribution of receptorsubunits. In studies where NR1 was expressed in fibroblasts, phosphorylation ofNR1 by PKC at serine-889 (Ser889) and Ser890 disrupted the subcellulardistribution of NR1 protein, suggesting that phosphorylation may play an importantrole in receptor clustering (Ehlers et al., 1995).


139

To summarize, a protein kinase can respond to multiple signals and regulatenumerous cellular functions. Although the mechanisms by which specificity isachieved are still not completely understood, the use of multiple isoforms withdifferent regulation and distribution patterns in the brain provides one possibleimplement. Similarly, different distribution and regulation of regulatory subunits ofprotein phosphatases modulates substrate specificity and targeting. Together, proteinkinases and phosphatases can regulate the activity of third messengers like thephosphoproteins inhibitor-1 and DARPP-32, which also show a strong regionaldistribution.

The described regional differences in distribution of protein kinases,phosphatases and phosphoproteins may not only underlie the brain regionaldifferences in the modulation of NMDA receptor signaling but can also be involvedin the regulation of several other neurotransmitter systems.

2. Brain regional differences in gene expression

The brain regional differences found in the modulation of NMDA receptorfunction by PKA (Chapters 2 and 3) suggest that dopamine type 1 (D1) and D2receptors, which are positively and negatively coupled to adenylate cyclase,respectively (Stoof and Kebabian, 1981; Dal Toso et al., 1989) may modulateNMDA receptor function in a brain region-specific manner. Dopamine and NMDAactivate multiple signaling pathways that are able to regulate gene expression. InChapter 4 and 5 dopamine and NMDA interactions in the regulation of immediateearly genes (IEG) were investigated and compared in the basal ganglia and thelimbic system monitoring phosphorylation of the transcription factor cAMPresponsive element binding protein (CREB) (Chapter 5) and production of FOSprotein (Chapter 4). Dopaminergic and/or NMDA receptor systems were stimulatedby intraperitoneal injection of agonist and/or antagonists. Indeed, strong regionaldifferences in the coupling of D1 and NMDA receptor stimulation with CREBphosphorylation and FOS production were observed. The D1 receptor agonist wasfound to elicit CREB phosphorylation and FOS production in the caudate nucleus.The NMDA receptor agonist also triggered CREB phosphorylation in the caudatenucleus however FOS production was not affected. Both drugs induced FOSproduction in the dentate gyrus, but only the NMDA receptor agonist also increasedlevels in the hippocampal CA1-3 region. Levels of phosphorylated CREB (pCREB)in the hippocampus were not affected by any of the two drugs.

Interestingly, in those brain regions where the D1 receptor agonist inducedCREB phosphorylation or FOS production, this rise in IEG expression could beblocked by the D1 receptor antagonist and also by the NMDA receptor antagonist.Similarly, the increase in pCREB or FOS levels after NMDA receptor agonistapplication was prevented by a NMDA receptor antagonist and a D1 receptorantagonist. Thus, interactions between the cyclic AMP and NMDA receptorpathways at the cellular level are required for dopamine- and NMDA-regulated geneexpression in the brain. These results are in agreement with earlier studies whichfound that dopamine receptor-stimulated phosphorylation of CREB and FOS

Chapter 9

140

production can be prevented by NMDA receptor antagonists (Keefe et al., 1998;Konradi et al., 1996; Nakazato, 1998; Tomitaka et al., 1995).

An additional brain region-specific regulation of gene responses bydopamine and NMDA interactions was achieved by D2 receptors, as revealed by theability of a D2 receptor antagonist to increase CREB phosphorylation and FOSproduction only in the caudate nucleus and to inhibit (in the amygdala and dentategyrus) or enhance (in the caudate nucleus) D1 receptor agonist-induced FOSproduction.

The observed, highly differentiated dopamine and NMDA receptorinteraction in the regulation of pCREB and FOS may enable the brain to integratedopaminergic and glutamatergic signaling to meet multiple diverse demands in abrain region-specific manner.

2.1 Specificity in signaling pathwaysThe intracellular mechanisms allowing for this triggering or permissive role

of D1 and NMDA receptors in pCREB or FOS production in a brain region-specificmanner are unknown at this time, but there are several explanations possible for theobserved regional differences in CREB and FOS expression upon activation of thedopaminergic and NMDA signaling system.

CREB phosphorylation and FOS production can be induced by multiplesecond messenger pathways (see Introduction). As addressed above, there may bedifferences in second messenger pathways in various brain regions, which are theresult of tissue specific distribution of phosphoproteins, protein kinases andphosphatase subtypes.

For example, DARPP-32 may be an important third messenger in theinteraction between dopaminergic and NMDA receptor signaling in the caudatenucleus. In Chapter 3 it was shown that activation of PKA (e.g. by dopaminereceptor activation) can lead to enhanced NMDA responses via the activation ofDARPP-32. Furthermore, glutamate, acting through NMDA receptors has beenshown to inactivate phosphorylated DARPP-32 through calcineurin (Halpain et al.,1990). The involvement of DARPP-32 in FOS production was also indicated by thefinding that DARPP-32-deficient mice have lower striatal FOS production afterapplication of the D1/D2 receptor agonists than wild-type controls (Fienberg et al.,1998; Svenningsson et al., 2000). Yan and colleagues (1999a) also used theDARPP-32 deficient mouse to show that DARPP-32 is an important component ofthe signaling cascades that mediate the D2 receptor-induced phosphorylation ofCREB. Thus, DARPP-32 may represent an intracellular target for dopamine andNMDA interactions. Indeed, activation of D1 produces FOS exclusively in striatalneurons expressing DARPP-32 (Beretta et al., 1992).

In addition, differences in second messenger pathway coupling of thedopamine receptors may also contribute to the regional differences in IEGexpression. A striking example is the D1 receptor, which is coupled to adenylatecyclase type V in striatum and thus activates cAMP (Glatt and Snyder, 1993),whereas the amygdaloid D1 receptors were found to be positively coupled to thephosphoinositol pathway (Undie and Friedman,1990). This would result in differentpathways underlying the FOS production in these two areas.


141

Although the promotor region of c-fos contains the Ca2+/ cAMP responsiveelement (CRE), where phosphorylated CREB can bind to induce gene expression, itis interesting to note that CREB phosphorylation did not always lead to FOSproduction. We found that in the caudate nucleus D1 receptor activation increasedCREB phosphorylation and subsequent FOS production whereas NMDA inducedCREB phosphorylation but not FOS production. It is unlikely that levels of CREBphosphorylation were not high enough in response to NMDA injection to promotegene expression since comparable levels of CREB phosphorylation have beenobserved in our study (Chapter 5, Fig 2) and several other studies after stimulationof cAMP and non-cAMP signals (Brindle et al., 1995; Seternes et al., 1999). Thus,there can be a difference in the ability of a CREB target gene like the c-fos gene torespond to one signal but not the other, despite comparable phosphorylation ofCREB. Similar to our finding that CREB phosphorylation may not always lead toFOS production, NMDA injection was shown to increase FOS production in thehippocampus although there was no rise in CREB phosphorylation observed.

The finding that CREB phosphorylation leads to FOS production only in thecaudate nucleus upon D1 receptor stimulation and not following NMDA receptoractivation, is in agreement with the recent findings that phosphorylation of CREB issufficient to induce cellular gene expression in response to cAMP. However,additional promotor-bound factors are required for target gene activation by CREBin response to factors other than cAMP, indicating the presence of cellularcomponents in the CREB pathway that discriminate between different inputs (Mayret al., 2001). The structure of CREB complexed with its coactivator CREB bindingprotein (CBP) has provided new insights into potential mechanisms underlying thissignal discrimination. It appeared that, in addition to mediating Ser133phosphorylation of CREB, PKA regulates additional proteins that are required forCREB-CBP complex formation and thus recruitment of the transcriptional apparatusto cAMP-responsive genes (Brindle et al., 1995). In addition to PKA activatedfactors other transactivators may be necessary for phosphorylated CREB to inducegene expression (Nakajima et al., 1996, Ginty, 1997). Thus, many extracellularstimuli can induce CREB phosphorylation at Ser133, yet, only some of these stimulican induce CREB-mediated transcription in the absence of other transactivators.

Whether phosphorylation of CREB may lead to c-fos gene expression, mayalso depend on the state of phosphorylation of CREB at other sites. For example,phosphorylation of CREB at Ser142 by casein kinase II and CaMKII has beenshown to block activation of target genes and formation of the CREB-CBP complexin vitro (Parker et al., 1998; Sun et al., 1994). However, substantial phosphorylationof this site has not yet been observed in vivo.

Differences in CREB-mediated gene expression activity after treatment withcAMP versus non-cAMP signals were also traced to the promotor level of the targetgene. A single consensus CRE is sufficient for target-gene activation through CREBin response to cAMP and calcium signals. But cellular activation in response togrowth factors requires additional promotor-bound factors such as serum responsefactor which synergize with CREB in a phosphorylated Ser133 dependent manner(Bonni et al., 1995). Many other studies state the importance of the occupation ofthe several cis-elements on the promotor of the target gene. The integrity of all ofthe major cis-regulatory elements was shown to be required for proper calcium

Chapter 9

142

sensitive activation of c-fos transcription in transgenic animals (Robertson et al.,1995). Although D1-triggered FOS production was shown to be mediated bypCREB which acts on the CRE (Das et al., 1997), Sharp and colleagues (1995)observed that the activation of CRE in striatum might not be sufficient to triggerFOS expression and that Ca2+ entry through NMDA receptor may be required toprovide an additional activation of the serum responsive element (SRE). This couldexplain the permissive role of NMDA in the D1 receptor agonist-induced increasein FOS production in the caudate nucleus (Chapter 4). From our results it is alsoevident that NMDA receptor activation by itself was not sufficient to induce FOSproduction in the caudate nucleus. This was in contrast to the findings in thehippocampus where both NMDA and SKF-38393 enhanced FOS production withouta significant change in CREB phosphorylation. It might be speculated, that in thehippocampus NMDA and SKF-38393 enhanced FOS production mainly via SRE.

In addition, control of the duration of CREB phosphorylation may be acritical regulator of CRE-mediated gene expression by dopamine and calcium (Liuand Graybiel, 1996). Interestingly, Ca2+, depending on its mode of entry intoneurons either via NMDA receptor or L-type Ca2+ channels, can activate twodistinct signaling pathways (Ginty, 1997). Calcium influx through L-type Ca2+channels leads to sustained CREB Ser-133 phosphorylation whereas the entry ofcalcium through NMDA receptors, except in very young neurons, triggers onlytransient phosphorylation of CREB (Hardingham et al., 1999; Sala et al., 2000).Recently, it has been demonstrated that calcium entry through the NMDA receptoractivates a phosphatase that dephosphorylates CREB. This might explain thetransient nature of the phosphorylation (Sala et al., 2000). PP1 has been shown toassociate with the NMDA receptor (Westphal et al., 1999), however, it remains to bedetermined whether the local tethering of this phosphatase near the receptor isrequired for the rapid dephosphorylation of CREB at Ser133.

The use of intraperitoneal drug administration to investigate in vivo IEGexpression has been a broadly accepted approach (e.g. Konradi et al., 1994, 1996, LeMoine et al., 1997). In vivo studies provide insight at the level of the complexcircuitry control of neuronal functions. This is important because the regulation ofpCREB and FOS may not occur on the basis of simple convergence of intracellularsignaling, but may rather rely on tissue-specific and circuitry-related interactions. Astriking example of the importance to investigate the regulation of CREBphosphorylation and FOS production in vivo is given by the finding that, althoughNMDA agonists are capable of triggering c-fos mRNA and FOS production instriatal cultures (Condorelli et al., 1994; Lerea et al., 1993), we did not find any FOSproduction in the striatum after in vivo administration of NMDA (Chapter 4). Thereare several explanations possible for this difference in c-fos expression in the in vitroand in vivo approach. It is possible that in vivo the Mg2+ block of NMDA receptorsprevented FOS production in the striatum, since striatal neurons are morehyperpolarized relative to their limbic counterparts (Hernandez-Lopez et al., 1997).Another explanation for the inability of NMDA to trigger FOS production can befound in the tonic inhibition of NMDA receptor function in this brain region by D2receptors (Chapter 4).


143

3. Synaptic plasticity and learning and memory

In the last part of this thesis, the relation between stress- and CRF-inducedalterations in synaptic strength and changes in learning and memory wasinvestigated. Electrophysiological recordings from the CA1 area in hippocampalslices were combined with experiments assessing learning using context-dependentfear conditioning which represents a hippocampus-dependent behavior test (Kim andFanselow, 1992; Phillips and LeDoux, 1994).

3.1 MetaplasticityAlterations in the strength of synaptic connections have been proposed to

underlie the cellular mechanisms that are involved in learning and memory. Withinthis general hypothesis, the potential for bidirectional changes in the strength ofsynaptic connections (LTP and long-term depression (LTD)) is crucial forpreserving the capability for change at a given synapse. In addition to bidirectionalmodification of synaptic strength, there is growing recognition of the existence ofmodifications of synaptic plasticity based on previous activation of the synapse(Christie and Abraham, 1992; Huang et al., 1992). In our studies, stress was used asa stimulus to induce changes in synaptic plasticity, and the effect of this change onsubsequent LTP induction and hippocampus-dependent learning was investigated.The acute behavioral stress consisted of one-hour immobilization.

It has long been known that hippocampal plasticity is susceptible tomodulation by stress. Diamond and colleagues showed that primed burstpotentiation, which is a long-term increase in CA1 population spike amplitudeproduced by brief physiologically patterned electrical stimulation of thehippocampal commissure is blocked after a stressful exposure to a novelenvironment (Diamond et al., 1994). Several other studies demonstrated that theinduction of hippocampal LTP by high frequency stimulation (HFS) was inhibitedafter stress (Foy et al., 1987; Kim et al., 1996). In our study, we also found reducedLTP when hippocampal slices from animals, that were decapitated 2 hr after stress,were stimulated with a standard 100 Hz HFS protocol producing saturating LTP inhippocampal slices of control animals. Interestingly, when slices from these animalswere stimulated with a weak theta burst stimulation (TBS; 5 X 100 Hz) protocol,which does not produce LTP in control animals, LTP persistence was facilitated.Our present results are the first to show that stress primes LTP persistence (Chapter7). The involvement of corticotropin-releasing factor (CRF) in this effect wasindicated by the finding that priming could be prevented by the CRF receptorantagonist [Glu 11,16]astressin. Furthermore, the same priming effect was seen inhippocampal slices preincubated with human/rat CRF (h/rCRF).

This phenomenon where prior synaptic activity can influence a subsequentchange in synaptic plasticity in the brain is called ‘metaplasticity’ (Abraham,1996).In our study, metaplasticity was induced by stress or h/rCRF application. It is notnecessarily expressed as a change in the efficacy of baseline synaptic transmissionbut it is manifested as a change in the threshold for LTP induction.

Modulation of NMDA receptor activation is a likely target for metaplasticityexpression. Several studies reported that prior low level activation of NMDAreceptors in the CA1 area inhibits subsequent induction of LTP (Coan et al., 1989;

Chapter 9

144

Fujii et al., 1991; Huang et al., 1992; Izumi et al., 1992; O’Dell and Kandel, 1994;but see Moody et al., 1999). Similarly, the NMDA receptor was shown to beinvolved in the detrimental effects of stress on HFS-induced LTP since the effect onLTP was prevented in animals that were given a competitive NMDA receptorantagonist before experiencing stress (Kim et al., 1996). The inhibition of LTP inthese experiments was however not absolute and could be overcome by strongertetanic stimulation (Huang et al., 1992).

The priming effect observed in hippocampal slices from stressed animalsstimulated with the 5 X 100 Hz TBS protocol is similar to the facilitation of LTPpersistence observed after activation of group I metabotropic glutamate receptors(mGluRs) or muscarinic receptor activation (Bortolotto et al., 1994; Christie et al.,1995; Cohen and Abraham, 1996; Cohen et al., 1998, 1999; Raymond et al., 2000).Both mGluR and muscarinic receptors are positively coupled to phosphoinositideturnover and thus capable to stimulate PKC activation. Bortolotto and Collingridge(2000) reported that PKC is involved in a form of metaplasticity that regulates theinduction of LTP at CA1 synapses. PKC activation was also involved in the primingeffect in our study. In Balb/c mice, h/rCRF application increased neuronalexcitability via the PKC pathway (Chapter 8) and a PKC inhibitor (Chapter 7) couldblock the stress- and h/rCRF-induced priming effect. To date, there is no relationknown between NMDA receptor activation and LTP facilitation. Activation ofmGluR enhances NMDA receptor function but this persists only for the period ofmGluR activation (Aniksztejn et al., 1992; Challis et al., 1994; Harvey andCollingridge, 1993). Cohen and Abraham (1996) propose that this form ofmetaplasticity, where prior activity leads to facilitation of LTP, is NMDA receptorindependent since the facilitation of LTP outlasted the enhancement of NMDAreceptor-mediated responses.

Besides the CRF receptor system, several other stress-related hormones arepotent regulators of LTP and LTD (Diamond et al., 1992; Kim et al., 1996) andthese hormones could thus play an important role in the stress-induced changes insynaptic plasticity (Chapter 7). Stress leads to the release of corticosterone from theadrenals during stress. Circulating corticosterone exerts negative feedback ontospecific brain regions including the hippocampus to inhibit further release. Thehippocampus is enriched with both the high affinity (type I) mineralocorticoid andthe lower-affinity (type II) glucocorticioid receptors. Binding of corticosterone tomineralocorticoid receptors seems to be a prerequisite for optimal LTP, whileactivation of glucocorticoid receptors promotes LTD induction (Diamond et al.,1992; Joels, 2001; Joels and de Kloet, 1992, Pavlides et al., 1993). This suggests thatlow levels of corticosterone enhance potentiation through stimulation of themineralocorticoid receptors and during stressful situations glucocorticoid receptorsbecome important because levels of of corticosterone become high to activate thisreceptor (McEwen and Sapolsky, 1995). Direct evidence for a role of theglucocorticoid receptor in the control of synaptic plasticity by stress came from Xuand colleagues (Xu et al., 1998). They showed that the glucocorticoid receptorantagonist RU 38486 blocked the effects of stress on LTD.


145

In summary, it can be speculated that LTP induction in CA1 under normalconditions varies as an inverted U-shaped form with applied stimulus strength (Fig1). HFS induced strong LTP in our control animals whereas TBS stimulation wassubthreshold to induce LTP. Stress may change the state of the synapses in such away that the whole LTP induction curve is shifted to the left, so that weakstimulation is more likely and strong stimulation is less likely to produce LTP. Insome experiments, the same stimulation patterns that subsequently facilitate LTDinduction were also found to be effective at inhibiting LTP induction (Christie andAbraham, 1992; Christie et al., 1995). Thus, inhibition of LTP may favor LTDinduction. Similarly, stress paradigms that inhibited LTP induction enhanced LTD(Kim et al., 1996; Xu et al., 1997).

In spite of the wealth of data described to date, we are far from being able toassemble a coherent molecular picture of metaplasticity. In principle, stress orh/rCRF alter the environment (pre- and postsynaptic) of the synapse by activatingkinases and phosphatases in such a way that subsequent stimulation of the synapsewill have different effects on synaptic plasticity depending on the phosphorylationstatus or several target molecules at the time the signal is received. Depending on theintensity of the subsequent stimulation signal, synaptic plasticity is either inhibitedor enhanced.

Figure 1. Model for activity-dependent variation in the threshold for LTP (adaptedfrom the Bienenstock-Coope-Munro (BCM) theory) The relation between inducingstimulus level and changes in synaptic strength is not fixed, but shifts according to thehistory of prior activity (modified from Kim and Yoon, 1996).

Cha

nges

in s

ynap

tic s

tren

gth

LTP

LTD

0

Inducing stimulus level

? ?

controlstress

100Hz HFS5x100 HzTBS

Chapter 9

146

3.2 Correlation between synaptic plasticity and learningStress and h/rCRF induced a metaplastic change in the state of synapses that

lasted for at least two hours. The persistence of such changes allows for anintegration of synaptic events that occurs widely in time, which is important fornormal learning and memory functions (Abraham, 1996).

Former studies reported that stressful events block the induction ofhippocampal HFS-induced LTP and suggested that deficits in spatial learningfollowing stress may be related to this suppression of LTP-like phenomena in thehippocampus (Diamond and Rose, 1994; Garcia, 2001). However, in our study, 2 hrafter immobilization LTP induction was facilitated after 5 X 100 Hz TBSstimulation. This was paralleled by increased fear conditioning in animals that weretrained 2 hr after the stress session. Similar to the priming effect, this learningimprovement could also be blocked by the antagonist [Glu 11,16]astressin andbisindolylmaleimide (BIS), a PKC inhibitor (Chapter 7). The finding, that LTPinduced by electrical stimulation modeled after endogenous theta-rhythm activitypatterns bears more relevance to behavior than LTP induced by arbitrary tetanictrains, parallels a study from Stäubli and colleagues (1999). They found that thefacilitatory effects on memory induced by GABA type B receptor antagonistscorresponded to those on LTP induced by TBS but not HFS.

To investigate the possible role of CRF receptors in the changes in synapticplasticity and learning and memory immediately after the end of the stress session,animals were injected with the CRFR antagonist CP-154,526. Systemic injection ofthis CRF receptor antagonist 15 min before the stress turned the stress into alearning-enhancing stimulus. CP-154,526 pretreatment had no significant effects onvarious forms of LTP but produced an increase in paired-pulse facilitation (PPF).PPF, in which double pulses were applied with differing intervals, is a measure forshort-lived plasticity and is thought to be mediated by a presynaptic mechanism(Kamiya and Zucker, 1994). Several other studies also investigated the relationbetween PPF and learning. However, a clear role for PPF in learning and memorycan still not be determined. For example, behavioral experiments were performed infour lines of mutant mice lacking key presynaptic proteins and thus showedabnormal PPF but apparently normal LTP (Silva et al., 1996). It was shown thatmutant mice which have decreased PPF, have also profound impairments in learningtasks including context-dependent fear conditioning, whereas behavioral analysisrevealed that mutant mice that had increased PPF performed equally well aswildtype animals in learning tasks (Silva et al., 1996). Furthermore, Sacchetti andcolleagues (2001) found a clear-cut correlation between the increase in excitabilityof the Schaffer collateral-CA1 dendrite synapses and freezing responseconsolidation but this did not appear to imply presynaptic transmitter releasemodifications, because they were not accompanied by PPF modifications.

It remains rather difficult to conclude, how the observed hippocampalfunctional changes are related to memory. Further investigations will be needed toascertain whether these modifications are due to a change in hippocampal activityduring stress, or whether they are, by themselves, a prerequisite of learning.


147

3.2.1 Possible role of calcium/calmodulin-dependent protein kinase IIA main factor in setting the direction of synaptic activity after afferent

stimulation is the intracellular Ca2+ concentration. Low concentrations ofintracellular Ca2+ may activate protein phosphatases required for LTD (Mulkey etal., 1994) whereas high concentrations activate protein kinases required for LTP(Malenka et al., 1989; Malinow et al., 1989). One consequence of increased Ca2+levels is activation of calcium/calmodulin-dependent protein kinase II (CaMKII).Interestingly, Lisman (1985, 1989) has developed a hypothesis involving CaMKII asthe key molecule for the storage of memory at individual synapses. Calcium-dependent activation of this kinase leads to autophosphorylation upon which itbecomes calcium-independent, and thus gives a degree of permanence to the signal,which could be important for memory. Therefore, we investigated the importance ofthe activated form of this kinase in our stress-induced effects on synaptic plasticityand learning and memory. Stress induced the activation of CaMKII with a maximum2 hr after the end of the stress session, which paralleled the learning improvementfound when animals were trained 2 hr after immobilization. The CaMKII inhibitorKN-62 antagonized this stress-mediated learning enhancement (Chapter 7).

Interestingly, application of KN-62 before stress did not block LTPpersistence after weak TBS stimulation. Thus, in our study, CaMKII may not haveplayed a role in the metaplastic effect where prior synaptic activity facilitatessubsequent persistence of LTP. This finding is rather unexpected with respect to therole of CaMKII as a molecular switch. If prior synaptic events set a molecularswitch so that CaMKII becomes persistently autophosphorylated, a subsequentincoming signal may be less effective in inducing LTP. Important is the observationfrom Mayford and colleagues (1995, 1996) who showed that a point mutation ofCaMKII, that mimics the effect of autophosphorylation and makes the kinase lessCa2+-dependent, both raises the threshold for LTP induction and makes LTD morelikely after low-frequency stimulation. Work by Giese and colleagues (1998)complemented these studies. They introduced a point mutation into the geneencoding CaMKII to block autophosphorylation. CA1 LTP could not be elicited inthe mutant mice across a range of stimulation frequencies. These mice also exhibitedprofound deficits in spatial learning in the water maze. Interestingly, we alsoobserved an impairment of HFS-induced LTP 2 hr after stress and found highestlevels of activated CaMKII at this timepoint (Chapter 8). High levels of activeCaMKII may phosphorylate ion channels important for LTP induction to such adegree that they may not respond to a second burst of transmitter so effectively(Lledo et al., 1995; Pettit et al., 1994). Activated CaMKII may serve as a sink forfree Ca2+ and calmodulin (Meyer et al., 1992), preventing the Ca2+/calmodulincomplex from activating enzymes involved in the induction of LTP. A reduction infree Ca2+ levels may result in a preferential activation of calcineurin. An alternativebiophysical hypothesis for the involvement of activated CaMKII in metaplasticitywas offered by Tompa and Friedrich (1998) who suggested that, because of the highabundance of this kinase in the hippocampus, autophosphorylation of CaMKIImolecules in the postsynaptic density can result in a significant localhyperpolarization of the postsynaptic membrane which possibly affects thesubsequent induction of LTP.

Chapter 9

148

Although one can thus speculate that CaMKII activation after stress changesthe possibility to induce LTP, the finding that stress-induced LTP persistence afterweak TBS was unaffected by KN-62 suggest that the involvement of CaMKII inLTP induced after stress, depends on the stimulus protocol used. In support of thissuggestion, it was found that the transgenic animals with the mutation that mimicsautophosphorylation, only showed impaired LTP in response to stimuli in the rangeof 5 to 10 Hz whereas LTP in response to 100 Hz-tetanus was unaltered (Mayford etal., 1995, 1996).

It is also important to note that our measurements do not differentiatebetween presynaptic and postsynaptic activated CaMKII. The CaMKII mechanismsdescribed above concentrate on postsynaptic mechanisms of action. However, bothpre- and postsynaptic changes can contribute substantially to changes in synapticplasticity in the hippocampal CA1 area and besides its role in LTP, CaMKII wasalso shown to play a crucial role in presynaptic function (Chapman et al., 1995). InChapter 6 increase in learning induced by the CRF receptor antagonist CP-14,526was paralleled by a rise in active CaMKII. The selective CaMKII inhibitor KN-62could block enhanced learning. The increased PPF after CP-154,526 pretreatmentcould also be blocked by KN-62. Because a presynaptic mechanism is thought tounderlie PPF, it can be speculated that a modification in presynaptic Ca2+homeostasis may have caused the changes in synaptic activity. Activation of proteinkinases in presynaptic terminals, particularly CaMKII was shown to correlate withneurotransmitter release (Chapman et al., 1995; Nichols et al., 1990). Our data are inagreement with the finding that mice heterozygous for a αCaMKII mutation havedecreased PPF but normal LTP in the CA1 region and demonstrate highly reducedcontextual fear conditioning (Silva et al., 1996).

When it comes to making the connection between synaptic plasticity andmemory, it is clearly to simplistic to associate changes in short-term and long-termplasticity with hippocampus-dependent memory formation. The key observations wemade, are, that LTP priming appeared to be essential for the enhancement oflearning 2 hr after stress in view of the observation that inhibition of LTP primingby [Glu 11,16]astressin and the PKC inhibitor BIS was associated with impairedlearning when [Glu 11,16]astressin or BIS were injected before stress. On the otherhand, it was demonstrated that inhibition of CaMKII activity reduced contextual fearconditioning without affecting LTP priming. This observation suggests that primingof LTP and activation of CaMKII represent two essential mechanisms, which maycontribute independently to long-term memory. Furthermore, the CP-154,526-induced increase in learning immediately after stress did not seem to be associatedwith a change in LTP induction but rather with an enhancement of short-termpotentiation (STP). CaMKII appeared to play a profound role in this effect.

3.2.2 Comparison between two mouse inbred strains Balb/c and C57BL/6NIn pilot studies Balb/c and C57BL/6N mice showed differences in context-

dependent fear conditioning following acute stress. Since CRF is an importantmediator of the stress response, we investigated possible differences of CRFreceptor activation in the hippocampus of both mouse strains that could account forthis difference.


149

Using intracellular recordings from hippocampal slices, h/rCRF was shownto enhance the excitability of hippocampal CA1 neurons as was observed by anincrease in neuronal firing after h/rCRF application in both Balb/c and C57BL/6Nmice (Chapter 8). However, the underlying second messenger pathways seemed todiffer in the two mouse strains. PKA was involved in the h/rCRF-induced increasein excitability in C57BL/6N mice whereas in Balb/c mice PKC appeared to play themain role. In addition, h/rCRF application did not influence learning in C57BL/6Nmice whereas it increased learning in Balb/c mice (Chapter 8).

Our finding that the PKC pathway was involved in the increasedhippocampal excitability in Balb/c mice after h/rCRF application (Chapter 8) andthat priming could be induced in Balb/c mice after h/rCRF application (Chapter 7) isin agreement with Cohen and colleagues (1998) who showed the importance of PKCin priming. One question arising from the above interpretation is whether C57BL/6Nmice, which show increased hippocampal excitability after h/rCRF application viathe PKA pathway, may be able to show increased LTP persistence after h/rCRFapplication. Several studies report that PKA is involved in learning and theinduction of long-lasting synaptic potentiation in the hippocampal CA1 area. Theinvolvement of PKA in LTP may depend on the stimulus protocol used. LTPinduced by multiple bursts of 100 Hz stimulation is long-lasting and criticallydependent on PKA whereas LTP induced by a single 100 Hz burst of stimulation isrelatively less robust and less dependent on PKA activation (Blitzer et al., 1995;Huang and Kandel, 1994).

Differences in second messenger coupling of the CRF receptor in thehippocampus of both mouse strains may at least in part explain why h/rCRFenhanced hippocampus-dependent learning in Balb/c mice, with no effect inC57BL/6N mice.

4. Concluding remarks

Neurons have the remarkable ability to undergo changes in the strength andpatterns of their signaling to other neurons. Several recent studies have suggestedthat long-lasting changes in synaptic structure and function depend in part on genetranscription and new protein synthesis. In this thesis we focused on the brainregion-specific phosphorylation mechanisms (e.g. DARPP-32) that modulateneuronal signals, such as the function of NMDA receptor channels, which maysubsequently stimulate the activation of particular transcription factors within thenucleus such as CREB and c-fos.

The relationship between the activation of these pathways and a particularbiological response in various brain areas is complex, and we observed considerablecross-talk between these cascades. Thus, an important unanswered question is howdifferent synaptic signals can activate the same signal transduction cascades butelicit distinct biological responses. Although our understanding of signalingspecificity in the nervous system remains rudimentary, it appears that temporal andspatial features of the synaptic signal itself may be critical for determining whichsignal transduction pathways are activated and lead to changes in gene expression aswe have demonstrated in the case of NMDA receptor- and dopamine receptor-

Chapter 9

150

signaling. Besides gene transcription, synaptic stimulation can also alter synapticstrength through the modification of existing proteins like PKA, PKC and CaMKII.It is a fast developing view that these second messengers play a role in neuronalplasticity, and hence in learning and memory. To figure out the exact nature of suchmechanisms, and their contributions to higher cognitive processes is a task for thefuture. However, in the short term, it is very well possible to elucidate therelationships that exist in the metabolic pathways with which second messengersinteract. In addition to their reported interaction with IEGs, we found that activationof the second messenger PKC by CRF or acute stress plays a key role in changingthe threshold for LTP induction which correlated with changes in associativelearning. Non-linear properties, such as threshold, play a key role in the informationprocessing of biological neural networks and, based on our findings, it appearsespecially interesting to further investigate the meaning of changing thresholds forseveral kinds of learning.

REFERENCES

Abraham WC (1996) Activity-dependent regulation of synaptic plasticity (metaplasticity) in thehippocampus. In: The hippocampus: functions and clinical relevance, pp. 15-26. ED. N. Kato. ElsevierScience B.V: Amsterdam.

Abraham WC and Bear MF (1996) Metaplasticity: the plasticity of synaptic plasticity. TrendsNeurosci. 19: 126-130.

Albin RL, Young AB and Penney JB (1989) The functional anatomy of basal ganglia disorders. TrendsNeurosci. 12: 366-375.

Alder R and Barbas H (1995) Complementary distribution of the phosphoproteins DARPP-32 and I-1in the cerebellar system. Neuroreport 6: 2368-2372.

Allen PB, Hvalby O, Jensen V, Errington ML, Ramsay M, Chaudhry FA, Bliss TV, Storm-Mathisen J,Morris RG, Andersen P and Greengard P (2000) Protein phosphatase-1 regulation in the induction oflong-term potentiation: heterogeneous molecular mechanisms. J. Neurosci. 20: 3537-3543.

Aniksztejn L, Otani S and Ben-Ari Y (1992) Quisqualate metabotropic receptors modulate NMDAcurrents and facilitate induction of long-term potentiation through protein kinase C. Eur. J. Neurosci. 4:500-505.

Barbas H, Gustafson EL and Greengard P (1993) Comparison of the immunocytochemical localizationof DARPP-32 and I-1 in the amygdala and hippocampus of the rhesus monkey. J. Comp. Neurol. 334:1-18.

Beal MF (1992) Mechanisms of excitotoxicity in neurologic diseases. FASEB J. 6: 3338-3344.

Beretta S, Robertson HA and Graybiel AM (1992) Dopamine and glutamate agonists stimulate neuron-specific expression of Fos-like protein in the striatum. J. Neurophysiol. 68: 767-777.

Blitzer RD, Wong T, Nouranifar R, Iyengar R and Landau EM (1995) Postsynaptic cAMP pathwaygates early LTP in hippocampal CA1 region. Neuron 15: 1403-1414.


151

Bonni A, Ginty DD, Dudek H and Greenberg ME (1995) Serine 133-phosphorylated CREB inducestranscription via a cooperative mechanism that may confer specificity to neurotrophin signals. Mol.Cell Neurosci. 6: 168-183.

Bortolotto ZA, Bashir ZI, Davies CH and Collingridge GL (1994) A molecular switch activated bymetabotropic glutamate receptors regulates induction of long-term potentiation. Nature 368: 740-743.

Bortolotto ZA and Collingridge GL (2000) A role for protein kinase C in a form of metaplasticity thatregulates the induction of long-term potentiation at CA1 synapses of the adult rat hippocampus. Eur. J.Neurosci. 12: 4055-4062.

Brindle P, Nakajima T and Montminy M (1995) Multiple protein kinase A-regulated events arerequired for transcriptional induction by cAMP. Proc. Natl. Acad. Sci. USA 92: 10521-10525.

Challis RA, Mistry R, Gray DW and Nahorski SR (1994) Modulation of muscarinic cholinoceptor-stimulated inositol 1,4,5-trisphosphate accumulation by N-methyl-D-aspartate in neonatal rat cerebralcortex. Neuropharmacology 33: 15-25.

Chapman PF, Frenguelli BG, Smith A, Chen CM and Silva AJ (1995) The alpha-Ca2+/calmodulinkinase II: a bidirectional modulator of presynaptic plasticity. Neuron 14: 591-597.

Chapman CA, Perez Y and Lacaille JC (1998) Effects of GABA(A) inhibition on the expression oflong-term potentiation in CA1 pyramidal cells are dependent on tetanization parameters. Hippocampus8: 289-298.

Chen L and Huang LY (1992) Protein kinase C reduces Mg2+ block of NMDA-receptor channels as amechanism of modulation. Nature 356: 521-523.

Christie BR and Abraham WC (1992) NMDA-dependent heterosynaptic long-term depression in thedentate gyrus of anaesthetized rats. Synapse 10: 1-6.

Christie BR, Stellwagen D and Abraham WC (1995) Reduction of the threshold for long-termpotentiation by prior theta-frequency synaptic activity. Hippocampus 5: 52-59.

Coan EJ, Irving AJ and Collingridge GL (1989) Low-frequency activation of the NMDA receptorsystem can prevent the induction of LTP. Neurosci. Lett. 105: 205-210.

Cohen AS and Abraham WC (1996) Facilitation of long-term potentiation by prior activation ofmetabotropic glutamate receptors. J. Neurophysiol. 76: 953-962.

Cohen AS, Coussens CM, Raymond CR and Abraham WC (1999) Long-lasting increase in cellularexcitability associated with the priming of LTP induction in rat hippocampus. J. Neurophysiol. 82:3139-3148.

Cohen AS, Raymond CR and Abraham WC (1998) Priming of long-term potentiation induced byactivation of metabotropic glutamate receptors coupled to phospholipase C. Hippocampus 8: 160-170.

Colwell CS, Altemus KL, Cepeda C and Levine MS (1996) Regulation of N-methyl-D-aspartate-induced toxicity in the neostriatum: a role for metabotropic glutamate receptors? Proc. Natl. Acad. Sci.USA 93: 1200-1204.

Condorelli DF, Dell'Albani P, Amico C, Lukasiuk K, Kaczmarek L and Giuffrida-Stella AM (1994)Glutamate receptor-driven activation of transcription factors in primary neuronal cultures. Neurochem.Res. 19: 489-499.

Crump FT, Dillman KS and Craig AM (2001) cAMP-dependent protein kinase mediates activity-regulated synaptic targeting of NMDA receptors. J. Neurosci. 21: 5079-5088.

Chapter 9

152

Dal Toso R, Sommer B, Ewert M, Herb A, Pritchett DB, Bach A, Shivers BD and Seeburg PH (1989)The dopamine D2 receptor: two molecular forms generated by alternative splicing. EMBO J. 8: 4025-4034.

Das S, Grunert M, Williams L and Vincent SR (1997) NMDA and D1 receptors regulate thephosphorylation of CREB and the induction of c-fos in striatal neurons in primary culture. Synapse 25:227-233.

Davis KL, Kahn RS, Ko G and Davidson M (1991) Dopamine in schizophrenia: a review andreconceptualization. Am. J. Psychiatry 148: 1474-1486.

Desdouits F, Siciliano JC, Greengard P and Girault JA (1995) Dopamine- and cAMP-regulatedphosphoprotein DARPP-32: phosphorylation of Ser-137 by casein kinase I inhibits dephosphorylationof Thr-34 by calcineurin. Proc. Natl. Acad. Sci. USA 92: 2682-2685.

Diamond DM, Bennett MC, Fleshner M and Rose GM (1992) Inverted-U relationship between thelevel of peripheral corticosterone and the magnitude of hippocampal primed burst potentiation.Hippocampus 2: 421-430.

Diamond DM, Fleshner M, Ingersoll N and Rose GM (1996) Psychological stress impairs spatialworking memory: relevance to electrophysiological studies of hippocampal function. Behav. Neurosci.110: 661-672.

Diamond DM and Rose GM (1994) Stress impairs LTP and hippocampal-dependent memory. Ann. N.Y. Acad. Sci. 746: 411-414.

Ehlers MD, Tingley WG and Huganir RL (1995) Regulated subcellular distribution of the NR1 subunitof the NMDA receptor. Science 269: 1734-1737

Fienberg AA and Greengard P (2000) The DARPP-32 knockout mouse. Brain Res. Rev. 31: 313-319.

Fienberg AA, Hiroi N, Mermelstein PG, Song W-J, Snyder GL, Nishi A, Cheramy A, O’Callaghan JP,Miller DB, Cole DG, Corbett R, Haile CN, Cooper DC, Onn SP, Grace AA, Ouimet CC, White FJ,Hyman SE, Surmeier DJ, Girault JA, Nestler EJ and Greengard P (1998) DARPP-32: Regulator of theefficacy of dopaminergic neurotransmission. Science 281: 838-842.

Flores-Hernandez J, Hernandez S, Snyder GL, Yan Z, Fienberg AA, Moss SJ, Greengard P andSurmeier DJ (2000) D(1) dopamine receptor activation reduces GABA(A) receptor currents inneostriatal neurons through a PKA/DARPP-32/PP1 signaling cascade. J. Neurophysiol. 83: 2996-3004.

Foy MR, Stanton ME, Levine S and Thompson RF (1987) Behavioral stress impairs long-termpotentiation in rodent hippocampus. Behav. Neural. Biol. 48: 138-149.

Frey U, Frey S, Schollmeier F and Krug M (1996) Influence of actinomycin D, a RNA synthesisinhibitor, on long-term potentiation in rat hippocampal neurons in vivo and in vitro. J. Physiol. 490:703-711.

Fujii S, Saito K, Miyakawa H, Ito K and Kato H (1991) Reversal of long-term potentiation(depotentiation) induced by tetanus stimulation of the input to CA1 neurons of guinea pig hippocampalslices. Brain Res. 555: 112-122.

Garcia R (2001) Stress, hippocampal plasticity, and spatial learning. Synapse 40: 180-183.

Giese KP, Fedorov NB, Filipkowski RK and Silva AJ (1998) Autophosphorylation at Thr286 of thealpha calcium-calmodulin kinase II in LTP and learning. Science 279: 870-873.


153

Gillard ER, Khan AM, Grewal RS, Mouradi B, Wolfsohn SD and Stanley BG (1998a) The secondmessenger cAMP elicits eating by an anatomically specific action in the perifornical hypothalamus. J.Neurosci. 18: 2646-2652.

Gillard ER, Khan AM, Haq AU, Grewal RS, Mouradi B and Stanley BG (1997) Stimulation of eatingby the second messenger cAMP in the perifornical and lateral hypothalamus. Am. J. Physiol. 273:R107-112.

Gillard ER, Khan AM, Mouradi B, Nalamwar O and Stanley BG (1998b) Eating induced byperifornical cAMP is behaviorally selective and involves protein kinase activity. Am. J. Physiol. 275:R647-653.

Ginty DD (1997) Calcium regulation of gene expression: isn't that spatial? Neuron 18: 183-186.

Girault JA, Hemmings HC Jr, Williams KR, Nairn AC and Greengard P (1989) Phosphorylation ofDARPP-32, a dopamine- and cAMP-regulated phosphoprotein, by casein kinase II. J. Biol. Chem. 264:21748-21759.

Glatt CE and Snyder SH (1993) Cloning and expression of an adenylyl cyclase localized to the corpusstriatum. Nature 361: 536-538.

Greengard P, Nairn AC, Girault JA, Ouimet CC, Snyder GL, Fisone G, Allen PB, Fienberg A andNishi A (1998) The DARPP-32/protein phosphatase-1 cascade: a model for signal integration. BrainRes. Rev. 26: 274-284.

Gustafson EL, Girault JA, Hemmings HC Jr, Nairn AC and Greengard P (1991) Immunocytochemicallocalization of phosphatase inhibitor-1 in rat brain. J. Comp. Neurol. 310: 170-188.

Halpain S, Girault JA and Greengard P (1990) Activation of NMDA receptors inducesdephosphorylationof DARPP-32 in rat striatal slices. Nature 343: 369-372.

Hardingham GE, Chawla S, Cruzalegui FH and Bading H (1999) Control of recruitment andtranscription-activating function of CBP determines gene regulation by NMDA receptors and L-typecalcium channels. Neuron 22: 789-798.

Harvey J and Collingridge GL (1993) Signal transduction pathways involved in the acute potentiationof NMDA responses by 1S,3R-ACPD in rat hippocampal slices. Br. J. Pharmacol. 109: 1085-1090.


Hernandez-Lopez S, Bargas J, Surmeier DJ, Reyes A and Galarraga E (1997) D1 receptor activationenhances evoked discharge in neostriatal medium spiny neurons by modulating an L-type Ca2+conductance. J. Neurosci. 17: 3334-3342.


Hsu KS, Ho WC, Huang CC and Tsai JJ (1999) Prior short-term synaptic disinhibition facilitates long-term potentiation and suppresses long-term depression at CA1 hippocampal synapses. Eur. J. Neurosci.11: 4059-4069.

Huang YY, Colino A, Selig DK and Malenka RC (1992) The influence of prior synaptic activity on theinduction of long-term potentiation. Science 255: 730-733.

Chapter 9

154

Huang YY and Kandel ER (1994) Recruitment of long-lasting and protein kinase A-dependent long-term potentiation in the CA1 region of hippocampus requires repeated tetanization. Learn. Mem. 1: 74-82.

Izumi Y, Clifford DB and Zorumski CF (1992) Inhibition of long-term potentiation by NMDA-mediated nitric oxide release. Science 257: 1273-1276.Joels M (2001) Corticosteroid actions in the hippocampus. J. Neuroendocrinol. 13: 657-669.

Joels M and de Kloet ER (1992) Control of neuronal excitability by corticosteroid hormones. TrendsNeurosci. 15: 25-30.

Kamiya H and Zucker RS (1994) Residual Ca2+ and short-term synaptic plasticity. Nature 371: 603-606.

Kauer JA, Malenka RC and Nicoll RA (1988) NMDA application potentiates synaptic transmission inthe hippocampus. Nature 334: 250-252.

Keefe KA and Ganguly A (1998) Effects of NMDA receptor antagonists on D1 dopamine receptor-mediated changes in striatal immediate early gene expression: Evidence for involvement ofpharmacologically distal NMDA receptors? Dev. Neurosci. 20: 216-228.

Khan AM, Curras MC, Dao J, Jamal FA, Turkowski CA, Goel RK, Gillard ER, Wolfsohn SD andStanley BG (1999) Lateral hypothalamic NMDA receptor subunits NR2A and/or NR2B mediateeating: immunochemical/behavioral evidence. Am. J. Physiol. 276: R880-891.



Kim JJ and Yoon KS (1998) Stress: metaplastic effects in the hippocampus. Trends Neurosci. 21: 505-509.

Konradi C, Cole RL, Heckers S and Hyman SE (1994) Amphetamine regulates gene expression in ratstriatum via transcription factor CREB. J. Neurosci. 14: 5623-5634.

Konradi C, Leveque JC and Hyman SE (1996) Amphetamine and dopamine-induced immediate earlygene expression in striatal neurons depends on postsynaptic NMDA receptors and calcium. J. Neurosci.16: 4231-4239.

Lledo PM, Hjelmstad GO, Mukherji S, Soderling TR, Malenka RC and Nicoll RA (1995)Calcium/calmodulin-dependent kinase II and long-term potentiation enhance synaptic transmission bythe same mechanism. Proc. Natl. Acad. Sci. USA 92: 11175-11179.

Le Moine C, Svenningsson P, Fredholm BB and Bloch B (1997) Dopamine-adenosine interactions inthe striatum and the globus pallidus: inhibition of striatopallidal neurons through either D2 or A2Areceptors enhances D1 receptor-mediated effects on c-fos expression. J. Neurosci. 17: 8038-8048.


Lerea LS and McNamara JO (1993) Ionotropic glutamate receptor subtypes activate c-fos transcriptionby distinct calcium-requiring intracellular signaling pathways. Neuron 10: 31-41.

Liao GY, Wagner DA, Hsu MH and Leonard JP (2001) Evidence for direct protein kinase-C mediatedmodulation of N-methyl-D-aspartate receptor current. Mol. Pharmacol. 59: 960-964.


155

Lisman JE (1985) A mechanism for memory storage insensitive to molecular turnover: a bistableautophosphorylating kinase. Proc. Natl. Acad. Sci. USA 82: 3055-3057.

Lisman JE (1989) A mechanism for the Hebb and the anti-Hebb processes underlying learning andmemory. Proc. Natl. Acad. Sci. USA 86: 9574-9578.

Liu FC and Graybiel AM (1996) Spatiotemporal dynamics of CREB phosphorylation: transient versussustained phosphorylation in the developing striatum. Neuron 17: 1133-1144.

MacDougall LK, Campbell DG, Hubbard MJ and Cohen P (1989) Partial structure and hormonalregulation of rabbit liver inhibitor-1; distribution of inhibitor-1 and inhibitor-2 in rabbit and rat tissues.Biochim. Biophys. Acta. 1010: 218-226.

Malenka RC, Kauer JA, Perkel DJ, Mauk MD, Kelly PT, Nicoll RA and Waxham MN (1989) Anessential role for postsynaptic calmodulin and protein kinase activity in long-term potentiation. Nature340: 554-557.

Malinow R, Schulman H and Tsien RW (1989) Inhibition of postsynaptic PKC or CaMKII blocksinduction but not expression of LTP. Science 245: 862-866.

Mayford M, Bach ME, Huang YY, Wang L, Hawkins RD and Kandel ER (1996) Control of memoryformation through regulated expression of a CaMKII transgene. Science 274: 1678-1683.

Mayford M, Wang J, Kandel ER and O'Dell TJ (1995) CaMKII regulates the frequency-responsefunction of hippocampal synapses for the production of both LTD and LTP. Cell 81: 891-904.

Mayr BM, Canettieri G and Montminy MR (2001) Distinct effects of cAMP and mitogenic signals onCREB-binding protein recruitment impart specificity to target gene activation via CREB. Proc. Natl.Acad. Sci. USA 98: 10936-10941.

McEwen BS and Sapolsky RM (1995) Stress and cognitive function. Curr. Opin. Neurobiol. 5: 205-216.

Meyer T, Hanson PI, Stryer L and Schulman H (1992) Calmodulin trapping by calcium-calmodulin-dependent protein kinase. Science 256: 1199-1202.

Moody TD, Carlisle HJ and O'Dell TJ (1999) A nitric oxide-independent and beta-adrenergic receptor-sensitive form of metaplasticity limits theta-frequency stimulation-induced LTP in the hippocampalCA1 region. Learn. Mem. 6: 619-633.

Mulkey RM, Endo S, Shenolikar S and Malenka RC (1994) Involvement of a calcineurin/inhibitor-1phosphatase cascade in hippocampal long-term depression. Nature 369: 486-488.

Nakajima T, Fukamizu A, Takahashi J, Gage FH, Fisher T, Blenis J and Montminy MR (1996) Thesignal-dependent coactivator CBP is a nuclear target for pp90RSK. Cell 86: 465-474.

Nakazato E, Ohno M and Watanabe S (1998) MK-801 reverses Fos expression induced by the fulldopamine D1 receptor agonist SKF-82958 in the rat striatum. Eur. J. Pharmacol. 342: 209-212.

Nestler EJ (1994) Hard target: understanding dopaminergic neurotransmission. Cell 79: 923-926.

Ngezahayo A, Schachner M and Artola A (2000) Synaptic activity modulates the induction ofbidirectional synaptic changes in adult mouse hippocampus. J. Neurosci. 20: 2451-2458.

Nguyen PV, Abel T and Kandel ER (1994) Requirement of a critical period of transcription forinduction of a late phase of LTP. Science 265: 1104-1107.

Chapter 9

156

Nichols RA, Sihra TS, Czernik AJ, Nairn AC and Greengard P (1990) Calcium/calmodulin-dependentprotein kinase II increases glutamate and noradrenaline release from synaptosomes. Nature 343: 647-651.

O'Dell TJ and Kandel ER (1994) Low-frequency stimulation erases LTP through an NMDA receptor-mediated activation of protein phosphatases. Learn. Mem. 1: 129-139.


Parker D, Jhala US, Radhakrishnan I, Yaffe MB, Reyes C, Shulman AI, Cantley LC, Wright PE andMontminy M (1998) Analysis of an activator:coactivator complex reveals an essential role forsecondary structure in transcriptional activation. Mol. Cell 2: 353-359.

Pavlides C, Watanabe Y and McEwen BS (1993) Effects of glucocorticoids on hippocampal long-termpotentiation. Hippocampus 3: 183-192.

Pettit DL, Perlman S and Malinow R (1994) Potentiated transmission and prevention of further LTP byincreased CaMKII activity in postsynaptic hippocampal slice neurons. Science 266: 1881-1885.


Raymond CR, Thompson V, Tate WP and Abraham WC (2000) Metabotropic glutamate receptorstrigger homosynaptic protein synthesis to prolong long-term potentiation. J. Neurosci. 20: 969-976.


Sacchetti B, Lorenzini CA, Baldi E, Bucherelli C, Roberto M, Tassoni G and Brunelli M (2001) Long-lasting hippocampal potentiation and contextual memory consolidation. Eur. J. Neurosci. 13: 2291-2298.

Sala C, Rudolph-Correia S and Sheng M (2000) Developmentally regulated NMDA receptor-dependent dephosphorylation of cAMP response element-binding protein (CREB) in hippocampalneurons. J. Neurosci. 20: 3529-3536.

Salamone JD (1991) Behavioral pharmacology of dopamine systems: a new synthesis. In Willner P &Scheel-Krieger J (Eds). The mesolimbic dopamine system: from motivation to action. John Wiley andSons, NY, pp 599-613.

Seternes OM, Johansen B and Moens U (1999) A dominant role for the Raf-MEK pathway inforskolin, 12-O-tetradecanoyl-phorbol acetate, and platelet-derived growth factor-induced CREB(cAMP-responsive element-binding protein) activation, uncoupled from serine 133 phosphorylation inNIH 3T3 cells. Mol. Endocrinol. 13: 1071-1083.

Sharp FR, Liu J, Nickolenko J and Bontempi B (1995) NMDA and D1 receptors mediate induction ofc-fos and junB genes in striatum following morphine administration: implications for studies ofmemory. Behav. Brain Res. 66: 225-230.

Sigel E, Baur R and Malherbe P (1994) Protein kinase C transiently activated heteromeric N-methyl-D-aspartate receptor channels independent of the phosphorylatable C-terminal splice domain and ofconsensus phosphorylation sites. J. Biol. Chem. 269: 8204-8208.


157

Silva AJ, Rosahl TW, Chapman PF, Marowitz Z, Friedman E, Frankland PW, Cestari V, Cioffi D,Sudhof TC and Bourtchuladze R (1996) Impaired learning in mice with abnormal short-lived plasticity.Curr. Biol. 6: 1509-1518.

Stanley BG, Butterfield BS and Grewal RS (1997) NMDA receptor coagonist glycine site: evidence fora role in lateral hypothalamic stimulation of feeding. Am. J. Physiol. 273: R790-796.

Stanley BG, Ha LH, Spears LC and Dee MG 2nd (1993a) Lateral hypothalamic injections of glutamate,kainic acid, D,L-alpha-amino-3-hydroxy-5-methyl-isoxazole propionic acid or N-methyl-D-asparticacid rapidly elicit intense transient eating in rats. Brain Res. 613: 88-95.

Stanley BG, Willett VL 3rd, Donias HW and Dee MG 2nd, Duva MA (1996) Lateral hypothalamicNMDA receptors and glutamate as physiological mediators of eating and weight control. Am. J.Physiol. 270: R443-449.

Stanley BG, Willett VL 3rd, Donias HW, Ha LH and Spears LC (1993b) The lateral hypothalamus: aprimary site mediating excitatory amino acid-elicited eating. Brain Res. 630: 41-49.

Stanton PK (1995) Transient protein kinase C activation primes long-term depression and suppresseslong-term potentiation of synaptic transmission in hippocampus. Proc. Natl. Acad. Sci. USA 92: 1724-1728.

Stäubli U, Scafidi J and Chun D (1999) GABAB receptor antagonism: facilitatory effects on memoryparallel those on LTP induced by TBS but not HFS. J. Neurosci. 19: 4609-4615.

Stoof JC and Kebabian JW (1981) Opposing roles for D-1 and D-2 dopamine receptors in efflux ofcyclic AMP from rat neostriatum. Nature 294: 366-368.

Sun P, Enslen H, Myung PS and Maurer RA (1994) Differential activation of CREB byCa2+/calmodulin-dependent protein kinases type II and type IV involves phosphorylation of a site thatnegatively regulates activity. Genes Dev. 8: 2527-2539.

Svenningsson P, Fienberg AA, Allen PB, Moine CL, Lindskog M, Fisone G, Greengard P andFredholm BB (2000) Dopamine D(1) receptor-induced gene transcription is modulated by DARPP-32.J. Neurochem. 75: 248-257.

Tingley WG, Ehlers MD, Kameyama K, Doherty C, Ptak JB, Riley CT and Huganir RL (1997)Characterization of protein kinase A and protein kinase C phosphorylation of the N-methyl-D-aspartatereceptor NR1 subunit using phosphorylation site-specific antibodies. J. Biol. Chem. 272: 5157-5166.

Tingley WG, Roche KW, Thompson AK and Huganir RL (1993) Regulation of NMDA receptorphosphorylation by alternative splicing of the C-terminal domain. Nature 364: 70-73.

Tomitaka S-I, Hashimoto K, Narita N, Minabe Y and Tamura A (1995) Amantadine induces c-fos inrat striatum: reversal with dopamine D1 and NMDA receptor antagonists. Eur. J. Pharmacol. 285: 207-211.

Tompa P and Friedrich P (1998) Synaptic metaplasticity and the local charge effect in postsynapticdensities. Trends Neurosci. 21: 97-102.

Undie AS and Friedman E (1990) Stimulation of a dopamine D1 receptor enhances inositol phosphatesformation in rat brain. J. Pharmacol. Exp. Ther. 253: 987-992.

Westphal RS, Tavalin SJ, Lin JW, Alto NM, Fraser ID, Langeberg LK, Sheng M and Scott JD (1999)Regulation of NMDA receptors by an associated phosphatase-kinase signaling complex. Science 285:93-96.

Chapter 9

158

White IM, Christensen JR, Flory GS, Miller DW and Rebec GV (1995) Amphetamine, cocaine, anddizocilpine enhance performance on a lever-release conditioned avoidance response task in rats.Psychopharmacology 118: 324-331.

Williams GV and Goldman-Rakic PS (1995) Modulation of memory fields by dopamine D1 receptorsin prefrontal cortex. Nature 376: 572-575.

Williams KR, Hemmings HC Jr, LoPresti MB, Konigsberg WH and Greengard P (1986) DARPP-32, adopamine- and cyclic AMP-regulated neuronal phosphoprotein. Primary structure and homology withprotein phosphatase inhibitor-1. J. Biol. Chem. 261: 1890-1903.

Xiong ZG, Raouf R, Lu WY, Wang LY, Orser BA, Dudek EM, Browning MD and MacDonald JF(1998) Regulation of N-methyl-D-aspartate receptor function by constitutively active protein kinase C.Mol. Pharmacol. 54: 1055-1063.


Xu L, Holscher C, Anwyl R and Rowan MJ (1998) Glucocorticoid receptor and protein/RNAsynthesis-dependent mechanisms underlie the control of synaptic plasticity by stress. Proc Natl AcadSci USA 95: 3204-3208.

Yamakura T, Mori H, Shimoji K and Mishina M (1993) Phosphorylation of the carboxyl-terminaldomain of the zeta 1 subunit is not responsible for potentiation by TPA of the NMDA receptor channel.Biochem. Biophys. Res. Commun. 196: 1537-1544.

Yan Z, Feng J, Fienberg AA and Greengard P (1999a) D(2) dopamine receptors induce mitogen-activated protein kinase and cAMP response element-binding protein phosphorylation in neurons. Proc.Natl. Acad. Sci. USA 96: 11607-11612.

Yan Z, Hsieh-Wilson L, Feng J, Tomizawa K, Allen PB, Fienberg AA, Nairn AC and Greengard P(1999b) Protein phosphatase 1 modulation of neostriatal AMPA channels: regulation by DARPP-32and spinophilin. Nat. Neurosci. 2: 13-17.

Zheng X, Zhang L, Wang AP, Bennett MV and Zukin RS (1999) Protein kinase C potentiation of N-methyl-D-aspartate receptor activity is not mediated by phosphorylation of N-methyl-D-aspartatereceptor subunits. Proc. Natl. Acad. Sci. USA 96: 15262-15267.

NEDERLANDSE SAMENVATTING

Wat verandert er in het zenuwstelsel als een dier iets leert? Hoe worden

herinneringen opgeslagen in de hersenen? Hieraan ten grondslag ligt het vermogen

van het zenuwstelsel om zich aan wisselende omstandigheden aan te passen, of, met

andere woorden, plastisch te zijn. Zoals elk orgaan zijn onze hersenen opgebouwd

uit afzonderlijke cellen, de hersencellen ofwel neuronen. De werking van de

hersenen berust op de communicatie tussen de neuronen. De communicatie vindt

plaats op plaatsen waar de membranen van neuronen heel dicht bij elkaar liggen, de

zogenaamde synapsen. Neuronen en de verbindingen daartussen zijn geen

onveranderlijke structuren. De synapsen vertonen plasticiteit. Hiermee wordt

bedoeld dat de elektrische prikkelbaarheid van neuronen verandert als gevolg van

hun eigen activiteit. Veranderingen in de synapsen tussen neuronen worden vaak

beschouwd als processen die essentieel zijn voor leren, geheugen en aanpassing.

Aan synaptische plasticiteit liggen zowel korte termijn als ook lange termijn

veranderingen in neuronale transmissie ten grondslag. Korte termijn veranderingen

bestaan voornamelijk uit post-translationele wijzigingen zoals fosforylatie van

bestaande proteïnen door proteïne kinases of defosforylatie door proteïne fosfatases,

terwijl lange termijn veranderingen ontstaan door een gecoördineerd programma van

veranderingen in gen transcriptie en de synthese van nieuwe proteïnen.

In deze dissertatie wordt geprobeerd meer inzicht te krijgen in de

moleculaire mechanismen die ten grondslag liggen aan veranderingen in synaptische

plasticiteit in het centrale zenuwstelsel en de rol van deze veranderingen bij leer- en

geheugenprocessen. Omdat dit een veelomvattend gebied is dat onmogelijk in zijn

totaliteit onderzocht kan worden, hebben we ons geconcentreerd op twee centrale

onderzoeksvragen.

Veranderingen in synaptische plasticiteit kunnen tot stand komen door

veranderingen in inhiberende of exciterende neurotransmissie of beide. Een van de

belangrijkste exciterende neurotransmitters in de hersenen is glutamaat. Glutamaat

Nederlandse samenvatting

160

kan binden aan een veelvoud van receptoren die benoemd zijn op basis van hun

elektrofysiologische en farmacologische eigenschappen. Meerdere studies hebben

aangetoond dat met name de fosforylatie status van de zogenaamde N-methyl-D-

aspartaat (NMDA) receptor belangrijk is bij veranderingen in synaptische

plasticiteit.

Als eerste wordt in de hoofdstukken 2 tot en met 5 gekeken of er regionale

verschillen zijn in de modulatie van de NMDA receptor door twee van de

voornaamste proteïne kinases, genaamd proteïne kinase C (PKC) en cyclisch AMP-

afhankelijke proteïne kinase (PKA). Het blijkt dat activatie van PKC leidt tot

vergroting van NMDA geïnduceerde stromen in alle onderzochte hersengebieden

(cortex, hippocampus, striatum en hypothalamus), terwijl de activatie van PKA maar

in enkele specifieke hersengebieden (striatum en hypothalamus) leidt tot vergroting

van NMDA geïnduceerde stromen.

In hoofdstuk 2 hebben we de modulatie van de NMDA receptor uit de

hypothalamus door PKA nader onderzocht. De experimenten wijzen erop dat het

functioneren van de NMDA receptor niet gemoduleerd wordt door directe

fosforylatie van de receptor door PKA. Eerder zorgt activatie van PKA ervoor dat

proteïne fosfatases geremd worden, waardoor de defosforylatie van de receptor

verminderd wordt.

In hoofdstuk 3 wordt het mechanisme waardoor de activatie van PKA leidt

tot verhoogd functioneren van de NMDA receptor in het striatum, beschreven. In het

striatum wordt het fosfoproteïne dopamine- en cAMP-gereguleerd fosfoproteïne

(DARPP-32) door PKA gefosforyleerd waarna het werkt als een remmer van

proteïne fosfatases. DARPP-32 laat een sterke hersengebied-specifieke verdeling in

de hersenen zien. Dit zou een verklaring kunnen zijn voor het feit dat activatie van

PKA niet in alle hersengebieden een vergroting van NMDA geïnduceerde stromen

tot gevolg heeft. Gezien de belangrijke rol van NMDA receptoren in synaptische

plasticiteit, zouden deze regionale verschillen in NMDA receptor modulatie door

proteïne kinases en fosfatases een sterke invloed kunnen uitoefenen op regionale

verschillen in NMDA receptorgekoppelde signalen die leiden tot veranderingen in

synaptische plasticiteit.


161

De regionale verschillen in de modulatie van de NMDA receptor door PKA

wijzen erop dat receptoren die gekoppeld zijn aan PKA ook het functioneren van de

NMDA receptor zouden kunnen beïnvloeden op een hersengebied-specifieke

manier. Een van de belangrijkste neurotransmitter receptoren in het striatum, die,

afhankelijk van het receptor subtype, zowel positief als negatief gekoppeld kan zijn

aan PKA, is de dopamine receptor. Daarom hebben we activatie van het

dopaminerge systeem gebruikt om modulatie van de NMDA receptor in vivo te

onderzoeken.

In veel modellen van synaptische plasticiteit worden veranderingen in de

activatie van de transcriptie factor cAMP-respons element bindend proteïne (CREB)

en c-fos gen expressie beschouwd als cruciale factoren voor de inductie en

handhaving van langdurige veranderingen in synaptische plasticiteit. Dus

veranderingen in de aanwezigheid van deze ‚immediate early genes‘ zouden inzicht

kunnen geven in de intracellulaire mechanismen die ten grondslag liggen aan

veranderingen in synaptische plasticiteit. Een veelvoud aan signaaltransductie-

wegen worden door NMDA en dopamine geactiveerd die ‚immediate early gene‘

expressie kunnen reguleren.

In de hoofdstukken 4 en 5 wordt onderzoek naar de interactie tussen NMDA

en dopamine receptor systemen in de regulatie van de immediate early genes FOS

(hoofdstuk 4) en CREB (hoofdstuk 5) in het striatum en het limbische systeem in

vivo beschreven.

In hoofdstuk 4 wordt beschreven hoe de expressie van FOS proteïne

verandert na intraperitoneale injectie van dopamine of NMDA receptor agonisten

en/of antagonisten. Inderdaad vinden we sterke regionale verschillen in de expressie

van het FOS proteïne na injectie van de agonisten. NMDA injectie verhoogt FOS

proteïne expressie in de hippocampus en de centrale amygdala, terwijl de dopamine

receptor subtype 1 (D1) receptor agonist FOS proteïne expressie verhoogt in het

striatum en de basomediale, corticale en mediale amygdala. In alle onderzochte

gebieden kan de NMDA receptor antagonist zowel de NMDA als ook de D1

receptor-geïnduceerde verhoging van het FOS proteïne blokkeren. Evenzo kan de

D1 receptor antagonist de D1 receptor en NMDA receptor-geïnduceerde verhoging


162

blokkeren. Dus NMDA en dopamine receptoren reguleren FOS produktie via

regionaal gedifferentieerde mechanismen maar het functioneren van de ene receptor

is wel onontbeerlijk voor het functioneren van de ander.

De promotor van het c-fos gen heeft een bindingsplaats voor de geactiveerde

vorm van CREB (gefosforyleerd CREB, pCREB). CREB kan geactiveerd worden

door PKA en door verhoogde intracellulaire calcium concentraties. Daarom is het

interessant te kijken of de regionale verschillen in FOS productie na injectie van

dopamine en NMDA receptor agonisten correleren met veranderingen in de

expressie van gefosforyleerd CREB. Hoewel ook de distributie van pCREB na

agonist injectie sterke regionale verschillen vertoont (hoofdstuk 5), correleert de

pCREB distributie niet in alle hersengebieden met FOS expressie. Er is dus een

duidelijk gedifferentieerde interactie tussen dopamine en NMDA receptor systemen

in de regulatie van pCREB en FOS in de verschillende hersengebieden.

De tweede centrale onderzoeksvraag in dit proefschrift luidt: Is er een relatie

tussen stress geïnduceerde veranderingen in synaptische plasticiteit en leren en

geheugen? Deze vraag wordt behandeld in de hoofdstukken 6, 7 en 8. Stress in de

vorm van een één uur durende immobilisatie wordt in deze dissertatie gebruikt als

stimulus om veranderingen in synaptische plasticiteit te induceren. Het effect van

deze veranderingen in synaptische plasticiteit op korte en lange termijn

hippocampale plasticiteit en het presteren in een hippocampus-afhankelijke leertaak

wordt onderzocht.

De veranderingen in hippocampale plasticiteit en leervermogen meteen na

einde van de stress sessie worden besproken in hoofdstuk 6. De nadruk wordt

daarbij gelegd op de rol van ‚corticotropin-releasing factor (CRF)‘ één van de

belangrijkste neuropeptides betrokken bij de gedrags, autonome en neuroendocriene

responsen op stress. Intraperitoneale injectie van een CRF receptor antagonist kort

voor immobilisatie blijkt ervoor te kunnen zorgen dat dieren die getraind worden in

de leertaak meteen na einde van de stress sessie, significant beter leren. Deze

verbetering in leerprestatie gaat gepaard met veranderingen in korte termijn

plasticiteit in de hippocampus maar niet met veranderingen in lange termijn

plasticiteit.


163

Training twee uur na einde van de stress sessie leidt ook tot een verbetering

van het leervermogen (hoofdstuk 7). Electrofysiologische metingen in het CA1

gebied van de hippocampus laten zien dat twee uur na immobilisatie lange termijn

potentiatie (LTP) geïnduceerd kan worden na een stimulatie die in controle dieren

niet leidt tot LTP. De rol van CRF in dit effect valt af te leiden uit het feit dat

incubatie van CRF evenzo resulteert in LTP.

In beide studies (hoofdstuk 6 en 7) blijkt het niveau van de actieve vorm van

de proteïne kinase calcium/calmodulin-afhankelijke proteïne kinase te correleren

met het leervermogen.

Nu de rol van CRF in de door stress veroorzaakte veranderingen in

plasticiteit en leervermogen vastgesteld is, onderzoeken we in hoofdstuk 8 het effect

van CRF op neuronale exciteerbaarheid en leervermogen in twee verschillenden

muizenstammen namelijk Balb/c en C57BL/6N in meer detail. Hoewel CRF in beide

muizenstammen leidt tot verhoogde exciteerbaarheid in de hippocampus, zijn de

signaaltransductie-wegen die hiertoe leiden verschillend in beide

muizenstammen.Verder beïnvloedt CRF in Balb/c muizen wel het leervermogen,

terwijl het leervermogen in C57BL/6N onveranderd blijft.

In de algemene discussie van dit proefschrift (hoofdstuk 9) wordt een

samenvatting van de resultaten gegeven en wordt besproken hoe deze resultaten

zouden kunnen bijdragen aan een beter inzicht in de mechanismen van synaptische

plasticiteit. Zo wordt er nader ingegaan op de mechanismen die ten grondslag

zouden kunnen liggen aan de hersengebied-specifieke modulatie van NMDA

receptoren en de regionale verschillen in ‚immediate early gene‘ expressie. Naar

aanleiding van de tweede centrale onderzoeksvraag wordt gespeculeerd over een

mogelijke relatie tussen veranderingen in synaptische plasticiteit en leervermogen.

SUPPLEMENT

NS-257, a novel competitive AMPA receptorantagonist, interacts with kainate and NMDA

receptors

Brain Research (1999) 821: 374-382

ABSTRACT

In this study, we examined the effects of a novel, water-soluble, putativecompetitive AMPA receptor antagonist, 1,2,3,6,7,8-hexahydro-3-(hydroxyimino)-N,N,7-trimethyl-2-oxobenzo[2,1-b:3,4-c’]dipyrrole-5-sulfonamide (NS-257) onAMPA, kainate and NMDA receptors using the two-electrode voltage-clamptechnique in Xenopus oocytes. All glutamate receptor subtypes were inhibited byNS-257 in a voltage-independent way. When kainate was applied to oocytes injectedwith total mouse brain mRNA, mainly AMPA receptors were activated. Theantagonistic effects of NS-257 on these kainate-induced currents wereconcentration-dependent and competitive. In the same way, NS-257 blockedkainate-induced currents recorded from oocytes expressing homomeric GluR-1receptors. In our experiments higher concentrations (> 1 µM) of NS-257 alsoproduced inhibitory effects on kainate and to a lesser extent on NMDA receptorfunction as indicated by recordings from GluR-6 or NR-1b/2A cRNA injectedoocytes. While NMDA receptor function was inhibited in a competitive fashion,kainate responses recorded from homomeric GluR-6 receptors were blocked in amixed competitive-noncompetitive manner. This mixed antagonistic action of NS-257 might have been caused by preincubating oocytes with concanavalin A, whichblocks desensitization of kainate receptors. Although NS-257 appeared to be a lesspotent AMPA receptor antagonist then other known antagonists like NBQX, itsmain advantage over all other reported compounds so far is its higher aqueoussolubility which still represents the major weakness of the other AMPA receptorantagonists, especially for clinical use.

Supplement

166

INTRODUCTION

Glutamate, the principal excitatory neurotransmitter in the brain, acts on twoclasses of ionotropic glutamate receptors, namely NMDA and non-NMDAreceptors. The non-NMDA receptor group is further divided into AMPA and kainatereceptor subtypes. Although AMPA receptors can be activated by both AMPA andkainate, they can be distinguished from kainate receptors by the different molecularstructure of their subunits. Molecular biological studies have determined that AMPAreceptors are composed of subunits GluR1-4 that can form functional channels. ThecDNAs encoding five different kainate receptor subunits have been cloned andclassified as either high affinity (KA-1 and KA-2) or low affinity (GluR5-7) kainatereceptor subunits (Hollmann and Heinemann, 1994). Besides studying thesedifferences on the molecular level, it is important to investigate the physiologicalfunctions of these two receptor subtypes. It is well established that AMPA receptorsmediate fast synaptic responses at most excitatory synapses in the mammalian brain(Collingridge and Lester, 1989). However, the physiological functions of the kainatereceptor subtype are still largely unknown (Lerma, 1997; Lerma et al., 1997).

Despite the widespread distribution of both [3H]kainate-binding sites(Honoré et al., 1986; London and Coyle, 1979) and mRNAs coding for kainatereceptors in brain (Wisden and Seeburg, 1993), it has been difficult to demonstratekainate receptor-mediated currents in neurons. This may be due to the rapiddesensitization of kainate receptors after exposure to kainate as demonstrated indorsal root ganglia and cultured hippocampal neurons (Huettner, 1990; Lerma et al.,1993) and the ‘masking’ of kainate responses by the large nondesensitizing responsethat kainate induces at AMPA receptors (Paternain et al., 1995). Recently, a firststep towards the elucidation of kainate receptor function was made when thissubclass of glutamate receptor was shown to be synaptically activated in mossyfiber/CA3 neuron contacts (Castillo et al., 1997; Vignes and Collingridge, 1997).Rodriquez-Moreno et al. demonstrated that activation of kainate receptors down-regulates GABAergic inhibition in hippocampal CA1 pyramidal neurons involvingmetabotropic receptors (Rodriquez-Moreno et al., 1997; Rodriquez-Moreno andLerma, 1998).

Figure 1. Structure of NS-257.

N

NO

N OHMe2NO2S

H

H3C

NS-257 interacts with kainate and NMDA receptors

167

The lack of specific pharmacological tools to discriminate betweenresponses of AMPA and kainate receptors has impeded the analysis of the synapticfunction of kainate receptors. Therefore, specific AMPA receptor antagonists orkainate receptor agonists are required to discriminate between the two subtypes ofnon-NMDA receptors. Recently, a potent and competitive non-NMDA antagonist,NS-257 1,2,3,6,7,8-hexahydro-3-(hydroxyimino)-N,N,7-trimethyl-2-oxobenzo[2,1-b:3,4-c’]dipyrrole-5-sulfonamide (Fig. 1), was described (Nielsen et al., 1995;Porter and Greenamyre, 1994; Wätjen et al., 1994). In binding experiments using ratcortical membrane tissue, NS-257 showed an 18-fold higher affinity for [3H]AMPA-binding sites than for [3H]kainate-binding sites. These results suggested selectivityfor AMPA receptors (Nielsen et al., 1995). Using quantitative receptorautoradiography, NS-257 was shown to compete for [3H]AMPA binding sites inseveral rat forebrain regions and cerebellar cortex (Porter and Greenamyre, 1994). Ithas been assumed that NS-257 is capable of crossing the blood-brain barrier on thebasis of its anticonvulsant activity after systemic administration (Wätjen et al.,1994). Its highly aqueous solubility made it an interesting candidate for clinicalconsideration to treat neuronal degenerative conditions depending upon glutamatetoxicity such as stroke, Huntington’s disease, amyotrophic lateral sclerosis andAlzheimer’s disease (Lipton et al., 1994; Meldrum and Garthwaite, 1990).

In this study, we investigated the potency and selectivity of this newcompound, NS-257, on different glutamate receptor subtypes using the two-electrode voltage-clamp technique in Xenopus oocytes injected with either totalmouse brain mRNA or cRNA encoding for GluR-1, NR-1b/2A or GluR-6.


RNA Preparation and Expression in Xenopus OocytesTotal RNA was isolated from the whole brain of adult male C57BL/6J mice by extraction of

fresh tissue with guanidine thiocyanate and precipitation with LiCl (Cathala et al., 1983). Poly (A)+

mRNA was purified by oligo(dT)-cellulose chromatography (Pharmacia mRNA Purification Kit) anddissolved in RNase-free water at a concentration of 0.5 µg/µl. Plasmids containing the cDNA clonesencoding the NR-1b, NR-2A, GluR-1 and GluR-6 receptor subunits were linearized by digestion withPvuI (NR-1b), XhoI (GluR-1) or NotI (NR-2A and GluR-6) and used as transcription templates. In vitrotranscription was carried out by T7 or T3 RNA polymerase in the presence of the capping reagentm7G(5')ppp(5')G. RNA transcripts were precipitated with ethanol and the precipitate was dissolved inDEPC-water. Oocytes were collected from anaesthetized specimens of Xenopus laevis as described(Kushner et al., 1988). Follicular cell layers were removed manually after incubation for 2 h at roomtemperature in Ca2+-free modified Barth's solution containing 3.2 mg/ml collagenase, type II. Stage Vand stage VI oocytes were pressure-injected with 50 ng poly (A)+ total brain mRNA/oocyte within 24 hafter harvesting. In some experiments, oocytes were injected with 10 ng GluR-1 or GluR-6cRNA/oocyte. The NR-1b cRNA was coinjected with NR-2A cRNA in the ratio 1:2 (10 ng: 20 ng).Oocytes were kept at 18°C in modified Barth's solution.

Electrophysiological RecordingsFour to seven days after injection, oocytes were placed in a recording chamber (volume ± 40

µl) and continuously superfused (1.5 ml/min) with recording solution at room temperature (20-24°C).Oocytes were voltage-clamped by a conventional two-microelectrode voltage-clamp technique(Stühmer, 1992). The membrane potential of the oocytes was held at -80 mV using a Turbo Clamp Tec01C amplifier (N.P.I. Electronic, Tamm, Germany). The two microelectrodes were filled with 3 M KCl

Supplement

168

and had resistances between 1-3 MΩ. Drugs were applied to the depression where the oocyte was heldwith a micropipette (200 µl) after stopping the superfusion. The depression where the oocyte was heldhad a volume of ± 5 µl, therefore any dilution or superfusion artefacts could be neglected. In someexperiments, oocytes were preincubated with a solution containing either concanavalin A (300 µg/ml;type IV; 10 min before concentration response curve establishment) or cyclothiazide (100 µM; 1 minbefore recording each response). Following cyclothiazide preincubation, cyclothiazide was also addedto the agonist solution. NS-257 and 6-nitro-7-sulphamoyl-benzo(f)quinoxaline-2-3-dione (NBQX) wereprepared as a 1 mM stock solution in bidistilled water and DMSO, respectively. Cyclothiazide wasdissolved in DMSO to a concentration of 20 mM. The final DMSO concentration never exceeded 0.5%(v/v), which, upon application, induced no change in membrane current on its own and which had noeffect on kainate-induced currents. Stock solutions of all other substances were prepared in recordingsolution. The pH of the kainate stock solution was adjusted to 7.2. Stock solutions were diluted withrecording solution to the final concentrations. Concentration response curves of kainate and NMDA,starting with the lowest concentration, were recorded after at least 2-3 identical responses could beelicited. Glycine (10 µM) was routinely added to the NMDA solution. Each response was followed bya 2 min wash-out period with recording solution before the next response was elicited. This wash-outprocedure was sufficient to prevent interference of subsequent treatments. Concentration responsecurves before and after treatment were always measured in the same oocyte. The stability of responseswas tested by applying 100 µM agonist before and after establishment of concentration responsecurves. Both response amplitudes never significantly differed from each other excluding that a decreaseof responses was simply due to a time-dependent “run-down” of response amplitudes upon agonistapplication. The voltage dependence of the blocking activity of NS-257 was determined by stepping themembrane potential through a series of voltage commands (from -120 to +30 mV). Leakage currentswere estimated by applying the same series of voltage steps in the absence of agonist. The differencebetween these two currents at each potential was taken as the agonist-induced current. The sameprocedure was followed for the agonist in the presence of NS-257. The concentration of NS-257 usedin these experiments was the IC50 concentration of NS-257 as calculated from the appropriate inhibitioncurves. For all experiments at least two different batches of oocytes were used. Oocytes were notfurther used if reproducible responses could not be recorded within 30 min. Current signals were low-pass filtered at 30 Hz employing a four pole Bessel filter and digitized by an ITC 16-MAC interface(Instrutech, Great Neck, NY, U.S.A.). Data were sampled at 100 Hz and stored on a Macintosh(7100/66) computer using data acquisition software (Pulse 7.40, HEKA electronic, Lambrecht/Pfalz,Germany).

Data AnalysisThe maximal response amplitude, the agonist concentration causing half maximal response

(EC50), and the slope coefficient (n) were determined with a non-linear least square curve-fittingprogram by fitting the data to the logistic equation:amplitude = maximal amplitude / [1+(EC50/[agonist])n].

For each oocyte, absolute response amplitude values were first fitted to the above equation.The absolute values were then normalized relative to the extrapolated maximal response amplitudecalculated. The curves shown were obtained by reapplying the equation to the normalized averagevalues. The reduction in maximal amplitude was obtained by subtracting the maximal amplitude, ascalculated from the equation, in the presence of NS-257 from the maximal amplitude in the absence ofNS-257. Statistical comparisons were made using the Student's t test. Results were expressed as mean ±SEM. p-values ≥ 0.05 were regarded as not significant.

SolutionsModified Barth's solution contained (in mM): 88 NaCl, 1 KCl, 2.4 NaHCO3, 15 Hepes, 0.8

MgSO4, 0.3 Ca(NO3)2 and 0.6 CaCl2, and 50 U/ml penicillin and 50 µg/ml streptomycin (adjusted topH 7.6 with NaOH). Ca2+-free modified Barth's solution contained (in mM): 88 NaCl, 1 KCl, 2.4NaHCO3, 15 Hepes, 0.8 MgSO4, and 50 U/ml penicillin and 50 µg/ml streptomycin (adjusted to pH 7.6with NaOH). Electrophysiological recordings were carried out in recording solution containing (inmM): 115 NaCl, 2.5 KCl, 1.8 Ba Cl2, 10 Hepes (pH adjusted to 7.2 with NaOH). Ba2+ ions werechosen as a substitute for Ca2+ ions in order to minimize secondary activation of Ca2+-dependent Cl-

currents.


169

MaterialsMature female specimens of X. laevis were obtained from Kähler (Hamburg, Germany).

Collagenase, type II, was from Worthington Biochemicals (Freehold, NJ, U.S.A.), mRNA PurificationKit from Pharmacia LKB Biotechnology (Uppsala, Sweden). Penicillin and streptomycin were fromGibco (Paisley, U.K.). NS-257 hydrochloride was obtained from RBI (Natick, MA, U.S.A.) and NBQXfrom Tocris (Bristol, U.K.). All other drugs and salts were purchased from Sigma (St. Louis, MO,U.S.A.).

RESULTS

Properties of kainate-induced currents recorded from oocytes injected withtotal mouse brain mRNA

Xenopus oocytes injected with total mouse brain mRNA were voltage-clamped at a holding potential of -80 mV. From the kainate concentration responsecurve a slope coefficient of 1.51 ± 0.03 and an EC50 concentration of 155.9 ± 6.2 µMwere calculated. When oocytes were preincubated with concanavalin A (con A),which is known to prevent kainate receptor desensitization (Partin et al., 1993;Wong and Mayer, 1993), kainate-induced currents were not affected (EC50

concentration 167.3 ± 8.8, slope coefficient 1.47 ± 0.09; n = 4; Fig. 2).

Figure 2. Cyclothiazide but not concanavalin A (con A) strongly affects responsesrecorded from oocytes expressing total mouse brain mRNA. A, Representative inwardcurrents evoked by 100 µM kainate after 10 min incubation with con A (300 µg/ml;10 min) or before and during 100 µM cyclothiazide application. B, Concentrationresponse relationships for kainate with or without preincubation with eithercyclothiazide or con A. Each point represents the mean ± SEM of four to twelveoocytes. The smooth curves are the result of nonlinear least square fit of the data tothe logistic equation and are plotted relative to the extrapolated maximal current.

In contrast, responses were markedly potentiated by cyclothiazide, acompound that is known to block glutamate-induced desensitization of the AMPAreceptor subtype (Fig. 2A) (Partin et al., 1994). The data indicated much greater

0

20

40

60

80

100120

100 101 102 103 104

Kainate concentration (µM)0

2 s200 nA

Con A

Kainate

Cyclothiazide

Control

Kainate

Control

KainateKainateA B

Frac

tion

of m

axim

al r

espo

nse

(%)

ControlConcanavalin ACyclothiazide

Supplement

170

cyclothiazide-mediated potentiation of kainate responses at low compared to highdoses of agonist (data not shown). Cyclothiazide produced a leftward shift of thekainate concentration response curve (Fig. 2B). Data analysis revealed a decrease ofthe EC50 concentration for kainate from a control concentration of 155.9 ± 6.2 µM to92.8 ± 5.7 µM and a change in the slope coefficient from 1.47 ± 0.09 to 1.22 ± 0.04(n = 8).

Antagonism of kainate-induced currents by NS-257The potency of NS-257 to block these kainate-induced currents recorded

from oocytes injected with total mouse brain mRNA was examined. Bathapplication of kainate (1-1000 µM) induced inward currents with a larger amplitudeas concentration increased. Co-application of NS-257 (0.5 or 1 µM) resulted in areversible inhibition of the response to kainate. NS-257 reduced the kainate-inducedcurrent competitively, with a increase in the EC50 for kainate (control: 154.0 ± 17.1µM; + 0.5 µM NS-257: 253.5 ± 16.7 µM; + 1.0 µM NS-257: 371.4 ± 15.6 µM; n =4) and no significant effect on the slope coefficient (control: 1.37 ± 0.05; + 0.5 µMNS-257: 1.36 ± 0.03; + 1.0 µM NS-257: 1.37 ± 0.03) or maximal response (Fig. 3).

Figure 3. Effect of NS-257 onkainate-induced currents recordedfrom total mouse brain mRNAinjected oocytes. A, Representativecurrents evoked by 100 µMkainate with and without 1.0 µMNS-257 . B , Concentrationresponse curves show theinhibition of kainate-inducedcurrents by 0.5 µM and 1.0 µMNS-257. Values were normalizedto the extrapolated maximalagonist-induced current in theabsence of NS-257. The curvesshown were obtained by applyingthe logistic equation to the relativeaverage values. Points representmean ± SEM of four oocytes.

Antagonistic activity of NS-257 on recombinant AMPA receptorsKainate-induced currents were recorded from oocytes injected with GluR-1

cRNA. In the absence of NS-257, the kainate EC50 concentration was 72.7 ± 6.6 µMwhereas in the presence of 1.0 µM NS-257 the kainate concentration response curveshowed a shift to the right giving an average EC50 concentration of 196.2 ± 12.6 µM(n = 9; Fig. 4). NS-257 did neither change the maximal response nor the slopecoefficient (control: 1.02 ± 0.03; + 1.0 µM NS-257: 1.07 ± 0.02) compared tocontrol conditions.

4 s40 nA

Kainate Kainate + NS-257

0

20

40

60

80

100

120 Control+ NS-257 (0.5 µM)+ NS-257 (1.0 µM)

100 101 102 103 104


B

Frac

tion

of m

axim

al r

espo

nse

(%)

A


171

Figure 4. NS-257 antagonizescur ren t s r ecorded f romrecombinant AMPA receptors. A,Responses evoked by 100 µmkainate in the presence andabsence of 1.0 µM NS-257 fromoocytes injected with GluR-1cRNA. B, Concentration responsecurve of kainate-induced currentswith and without 1.0 µM NS-257.Data are presented as describedfor Fig. 3. Points represent mean ±SEM of four oocytes.

Antagonistic effect of NS-257 on kainate-induced currents recorded fromoocytes injected with GluR-6 cRNA

The pharmacological properties of NS-257 on kainate receptors wereinvestigated in Xenopus oocytes injected with cRNA coding for the GluR-6 subunit.GluR-6 was used on the basis of previous studies demonstrating that GluR-6subunits form functional homomeric channels, which behave like native kainatereceptors (Lerma et al., 1993; Patneau et al., 1994).

Figure 5. Mixed competitive-noncompetitive block of kainatereceptors by NS-257. A , Inwardcurrent activated by 100 µMkainate before and after a 10 minincubation with 300 µg/ml con A.NS-257 reduced the kainte-inducedcurrent. B, After con A treatmentGluR-6 cRNA injected oocyteswere exposed to increasing kainateconcentrations with and without 50µM NS-257. Points represent mean± SEM of eight oocytes. Data areshown as described for Fig. 3

4 s100 nA

Kainate Kainate + NS-257A

B

0

20

40

60

80

100

120 Control+ NS-257 (1 µM)

10 -1 100 101 102 10 3 10 4


Frac

tion

of m

axim

al r

espo

nse

(%)

4 s100 nA

Kainate Kainate + NS-257KainateConcanavalin A

020

40

60

80

100

120 Control+ NS-257 (50 µM)

10-2 10-1 100 101 102 103


A

B

Frac

tion

of m

axim

al r

espo

nse

(%)

Supplement

172

Receptor desensitization was routinely prevented by preincubation with conA for 10 min. From the kainate concentration response curve, an EC50 concentrationof 1.28 ± 0.21 µM and a slope coefficient of 1.17 ± 0.09 were deduced. Fifty µMNS-257 produced a reversible inhibition of kainate-induced currents in a mixedcompetitive-noncompetitive manner (EC50 concentration 34.49 ± 2.74 µM; slopecoefficient 1.46 ± 0.10; n = 8; Fig. 5). The maximal response was reduced by 14 ±3%.

Antagonistic effect of NS-257 on NMDA-induced currentsFinally, the effect of NS-257 on the NMDA receptor subtype was tested.

NMDA responses were elicited in oocytes injected with NR-1b/2A cRNA byapplication of NMDA (1-1000 µM) and its coagonist glycine (10 µM).Concentration response curves generated for NMDA in 4 oocytes showed an EC50

concentration of 75.1 ± 9.8 µM and a slope coefficient of 1.45 ± 0.04. Whensubsequently a constant concentration of NS-257 (30 µM) was combined withincreasing concentrations of NMDA, it was evident that NS-257 shifted the NMDAconcentration response curve to the right (EC50 concentration 134.1 ± 16.5 µM; Fig.6). There was no significant change in the maximal response but the slopecoefficient changed from 1.45 ± 0.04 to 1.82 ± 0.09.

Figure 6. Effect of NS-257 onrecombinant NMDA receptors.A, Representative current evokedby 100 µM NMDA and itscoagonist glycine (10 µM) in theabsence and presence of 30 µMNS-257. B , The concentrationresponse relationships forNMDA with and without 30 µMNS-257 in NR-1b/2A cRNAinjected oocytes is shown.Glycine (10 µM) was routinelyadded to the NMDA solution.Points represent mean ± SEM offour oocytes. Data are presentedas described for Fig. 3.

Potency and selectivity of NS-257The potency of NS-257 (0-100 µM) to block kainate-induced currents which

were recorded from oocytes injected with total mouse brain mRNA, was comparedto the potency of NBQX (0-50 µM), a known potent and selective competitiveAMPA receptor antagonist. Kainate was used at a concentration producing a halfmaximal response (EC50 concentration 155 µM) as determined from theconcentration response curves. The IC50 concentrations estimated from inhibition

4 s

40 nA

NMDA NMDA + NS-257

0

20

40

60

80

100

120

+ NS-257 (30 µM)Control

100 101 102 103

NMDA concentration (µM)0

A

B

Frac

tion

of m

axim

al r

espo

nse

(%)


173

curves were 1.05 ± 0.04 µM for NS-257 (n = 5) and 0.08 ± 0.01 µM for NBQX (n =4; Fig. 7A). The slope coefficients did not significantly differ, -1.31 ± 0.02 and -1.40± 0.12 for NS-257 and NBQX, respectively.

To determine the selectivity of NS-257 for recombinant AMPA, kainate orNMDA receptors, a range of concentrations (0-100 µM) was tested for their abilityto block currents evoked by the EC50 agonist concentration as calculated from theappropriate concentration response curves. Increasing concentrations of NS-257(from 0 to 100 µM) potently blocked kainate-induced currents recorded from GluR-1 receptors (IC50 concentration: 0.97 ± 0.08 µM; slope coefficient: -1.28 ± 0.05; n =5). Interestingly, NS-257 blocked kainate-induced currents from recombinantAMPA receptors in the same way as kainate-induced currents recorded from oocytesinjected with total mouse brain mRNA, as reflected by overlapping inhibition curves(Fig. 7B). NS-257 was less potent in blocking responses recorded from GluR-6 orNR-1b/2A receptors (Fig. 7B). The IC50 concentrations were 2.30 ± 0.05 µM (n = 5)and 28.3 ± 4.4 µM (n = 3), respectively. The slope coefficients, however, did notsignificantly differ from GluR-1 receptors (GluR-6: -1.26 ± 0.03 and NR-1b/2A: -1.38 ± 0.12).

Figure 7. Potency of NS-257. A,Concentration-dependent effect ofNS-257 and NBQX on kainate-induced currents recorded fromtotal mouse brain mRNA injectedoocytes. NS-257 or NBQX wasadded to the EC50 concentration ofkainate (155 µM) as calculated fromthe concentration response curve. B,Comparison of the blocking activityof NS-257 on the different glutamatereceptor subtypes as recorded fromoocytes injected with GluR-1, GluR-6, NR-1b/2a cRNA or total mousebrain mRNA. NS-257 (0-100 µM)was added to the EC50 concentrationof agonist . Responses arenormalized relative to the currentevoked by the EC50 concentration ofagonist in the absence of theantagonist. Points represent mean ±SEM of four to five oocytes.

0

20

40

60

80

'AMPAR'

GluR-1GluR-6NR-1b/2A

10 -2 10-1 100 101 102 10NS-257 concentration (µM)

0

B

3

100

0

20

40

60

80

100

NS-257NBQX

10-2 10 -1 100 101 102

Antagonist concentration (µM)0

A

% o

f co

ntro

l%

of

cont

rol

Supplement

174

Voltage dependence of the NS-257 blockAn examination of the I-V relationship was used to assess whether the

inhibition induced by NS-257 was voltage-dependent. The I-V curves for kainate-induced currents (100 µM) recorded from oocytes injected with GluR-1 over amembrane potential range from -120 to +30 mV, exhibited a strong inwardrectification. This characteristic inward rectification was unaffected by NS-257 (0.97µM; n = 7; Fig. 8A). Similar results were obtained by using kainate (1 µM) asreceptor agonist for homomeric GluR-6 receptors. This I-V curve also displayed astrong inward rectification, which was not changed by NS-257 (2.3 µM; n = 6; Fig.8B). The I-V relationship for the response to 100 µM NMDA and its coagonistglycine (10 µM) recorded from NR-1b/2A cRNA injected oocytes was linear andhad a reversal potential close to 0 mV. Both linearity of the I-V curve and reversalpotential for NMDA-induced currents were unchanged in the presence of 28 µMNS-257 (n = 6; Fig. 8C).

DISCUSSION

In recent years, numerous attempts were made to discriminatepharmacologically between AMPA and kainate receptors. The lack of specificantagonists has precluded the analysis of kainate receptor function withoutinterference from activated AMPA receptors. In this electrophysiological study weexamined the inhibitory action of NS-257, a novel AMPA receptor antagonist(Nielsen et al., 1995; Porter and Greenamyre, 1994; Wätjen et al., 1994), on NMDAand non-NMDA glutamate receptors using the Xenopus oocyte expression system.

Radioligand binding studies suggested that NS-257 binds preferentially toAMPA receptors with limited affinity for NMDA and kainate receptors (Wätjen etal., 1994). In agreement, our electrophysiological experiments exhibited a selectivityof NS-257 for AMPA receptors with confined inhibitory effects on kainate and to alesser extent on NMDA receptors. The blocking activity of NS-257 was voltage-independent for both kainate-activated currents recorded from homomeric GluR-1and GluR-6 receptors (both displayed inward-rectifying I-V relationships) as well asfor NMDA-activated currents recorded from heteromeric NR-1b/2A receptors(linear I-V relationship). This lack of voltage dependence pointed that NS-257 wasbinding in all cases at the receptor complex largely outside the transmembraneelectric field of the channel pore region. The two compounds cyclothiazide andconcanavalin A (con A) were used to characterize the kainate-induced currentsrecorded from total mouse brain mRNA injected oocytes. Previous studies ofrecombinant glutamate receptors expressed in Xenopus oocytes and humanembryonic kidney (HEK 293) cells demonstrated that cyclothiazide-mediatedpotentiation is highly selective for AMPA receptor function. In contrast, con A isonly effective at kainate but not at AMPA receptors (Partin et al., 1993; Wong andMayer, 1993).


175

Figure 8. Voltage-dependence of the NS-257 blocking activity.I-V relationships determined from GluR-1 (A), GluR-6 (B) or NR-1b/2A (C) cRNAinjected oocytes before and after the application of NS-257 (IC50 concentration ascalculated from the appropriate inhibition curves).

Since in our experimental paradigm cyclothiazide, but not con A, affectedkainate-induced currents recorded from oocytes expressing total mouse brainmRNA, it was concluded that these currents were mainly caused by the activation ofAMPA receptors. The EC50 concentration and slope coefficient calculated from theobtained kainate concentration response curve were in the range reported for oocytesinjected with mRNA isolated from rat cortex (Okada et al., 1996) and hippocampalneurons (Patneau et al., 1993).

Interestingly, cyclothiazide appeared to enhance responses more potently atlow than at high kainate concentrations, which resulted in a distorted concentrationresponse curve for low kainate concentrations in the presence of cyclothiazide.These functional properties might be explained by the fact that heteromeric AMPAreceptors contain subunit splice variants with different sensitivity to cyclothiazideand/or kainate (Partin et al., 1994).

The kainate concentration response curve recorded from total mouse brainmRNA injected oocytes revealed a parallel shift to the right without change in themaximal response, when NS-257 was added to the agonist. This was consistent withthe suggestion that NS-257 acted as a competitive antagonist at kainate recognitionsites of AMPA receptors.

A similar mode of antagonism was found when kainate-induced responseswere recorded from GluR-1 receptors. Kainate concentration response curves fromtotal mouse brain mRNA injected oocytes and GluR-1 cRNA injected oocytesshowed different EC50 concentrations. This might be explained by a difference in

GluR-1

NR-1b/2A

-150 -100 -50 50

-100

-50

50

NMDANMDA + NS-257

V (mV)

-100 -50 50

-150

-100

-50

50

KainateKainate + NS-257

V (mV)-150 -150 -100 -50 50

-3000

-2000

-1000

1000

KainateKainate + NS-257

V (mV)A

C

BI (nA

)

I (nA)

I (nA)

GluR-6

Supplement

176

receptor subunit composition. However, receptor subunit composition did notinfluence NS-257 block as evidenced by the two overlapping concentration responsecurves showing the inhibition of kainate-induced currents by different NS-257concentrations. The observation that the inhibition curves were overlapping, alsoindicated that kainate-induced responses from total mouse brain mRNA injectedoocytes were mainly caused by AMPA receptor activation.

Concentration response curves established from oocytes expressing totalbrain mRNA, GluR-1 or NR-1b/2A cRNA all exhibited a competitive antagonism ofNS-257. In contrast, concentration response curves for GluR-6 receptors showed amixed competitive-noncompetitive nature of NS-257 antagonism. Although thisfinding might suggest that NS-257 was binding at two different kainate receptorbinding sites it can not be excluded that the preincubation with con A was at leastpartially responsible for this mixed character of NS-257 inhibition.

When the potency of NS-257 to block kainate-induced currents recordedfrom total mouse brain mRNA injected oocytes was compared to the potency ofNBQX, a well-characterized AMPA receptor antagonist, it was shown that NBQXwas ~13-fold more potent. This was in agreement with former binding studies,which showed that NS-257 is a less potent AMPA receptor antagonist compared toNBQX (Porter and Greenamyre, 1994; Wätjen et al., 1994).

Receptor antagonists are not only interesting as experimental tools forinvestigating fundamental central nervous mechanisms, but also as potential drugsfor CNS disorders. Various competitive and noncompetitive AMPA receptorantagonists have been shown to be neuroprotective in animal models of ischemicinjury or other neurodegenerative conditions (Gill, 1994). But the use of mostAMPA/kainate antagonists like NBQX was hampered by solubility problems, whichresulted in nephrotoxicity following intravenous administration (Xue et al., 1994).The higher aqueous solubility of NS-257 compared to other potent AMPA receptorantagonists and the finding that systemic administration of NS-257 protected againstseizures (Wätjen et al., 1994) makes it a highly interesting compound for clinicaluse. We showed that, besides blocking AMPA receptor function, NS-257 exhibiteddistinct inhibition of responses recorded from kainate and NMDA receptor channels.Therefore, to further consider the use of NS-257 for experimental or therapeuticpurposes a concentration around 1µM is recommendable to suppress AMPAreceptor-mediated currents with minor side effects on kainate and NMDA receptorfunction.

ACKNOWLEDGMENTS

The authors would like to thank Dr. M. Hollmann for kindly providing theGluR-1, GluR-6, NR-1b and NR-2A cDNA and Dr. J. Radulovic for providing themouse brain tissue.


177

REFERENCES

Castillo PE, Malenka RC and Nicoll RA (1997) Kainate receptors mediate a slow postsynaptic currentin hippocampal CA3 neurons. Nature 388: 182-186.

Cathala G, Savouret JF, Mendez B, West BL, Karin M, Martial JA and Baxter JD (1983) A method forisolation of intact, translationally active ribonucleic acid. DNA 2: 329-335.

Collingridge GL and Lester RAJ (1989) Excitatory amino acid receptors in vertebrate central nervoussystem. Pharmacol. Rev. 40: 143-210.

Gill R (1994) The pharmacology of α -amino-3-hydroxy-5-methyl-4-isoxazole propionate(AMPA)/kainate antagonists and their role in cerebral ischaemia. Cerebrovasc. Brain Metab. Rev. 6:225-256.


Honoré T, Drejer J and Nielsen M (1986) Calcium discriminates two [3H]kainate binding sites withdifferent molecular target sizes. Neurosci. Lett. 65: 47-52.

Huettner JE (1990) Glutamate receptor channels in rat DRG neurons: activation by kainate andquisqualate and blockade of desensitization by Con A. Neuron 5: 255-266.

Kushner L, Lerma J, Zukin RS and Bennett MVL (1988) Coexpression of N-methyl-D-aspartate andphencyclidine receptors in Xenopus oocytes injected with rat brain mRNA. Proc. Natl. Acad. Sci. USA85: 3250-3254.

Lerma J, Paternain AV, Naranjo JR and Mellström B (1993) Functional kainate-selective glutamatereceptors in cultured hippocampal neurons. Proc. Natl. Acad. Sci. USA 90: 11688-11692.

Lerma J (1997) Kainate reveals its targets. Neuron 19: 1155-1158.

Lerma J, Morales M, Vicente MA and Herreras O (1997) Glutamate receptors of the kainate type andsynaptic transmission. Trends Neurosci. 20: 9-12.

Lipton SA and Rosenberg PA (1994) Excitatory amino acids as a final common pathway for neurologicdisorders. New Engl. J. Med. 330: 613-622.

London E and Coyle J (1979) Specific binding of [3H]kainic acid to receptor sites in rat brain. Molec.Pharmacol. 15: 492-505.

Meldrum B and Garthwaite J (1990) Excitatory amino acid neurotoxicity and neurodegenerativedisease. Trends Pharmacol. Sci. 11: 379-387.

Nielsen EO, Johansen TH, Wätjen F and Drejer J (1995) Characterization of the binding of [3H]NS-257, a novel competitive AMPA receptor antagonist, to rat brain membranes and brain sections. J.Neurochem. 65: 1264-1273.

Okada M, Kohara A and Yamaguchi T (1996) Characterization of YM90K, a selective and potentantagonist of AMPA receptors, in rat cortical mRNA-injected Xenopus oocytes. Eur. J. Pharmacol. 30:299-306.

Partin KM, Patneau DK and Mayer ML (1994) Cyclothiazide differentially modulates desensitizationof α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor splice variants. Molec. Pharmacol.46: 129-138.

Supplement

178

Partin KM, Patneau DK, Winters CA, Mayer ML and Buonanno A (1993) Selective modulation ofdesensitization at AMPA versus kainate receptors by cyclothiazide and concanavalin A. Neuron 11:1069-1082.

Paternain AV, Morales M and Lerma J (1995) Selective antagonism of AMPA receptors unmasks thekainate receptor-mediated responses in hippocampal neurons. Neuron 14: 185-189.

Patneau DK, Vyklicky L Jr and Mayer ML (1993) Hippocampal neurons exhibit cyclothiazide-sensitive rapidly desensitizing responses to kainate. J. Neurosci. 13: 3496-3509.

Patneau DK, Wright PW, Winters C, Mayer ML and Gallo V (1994) Glial cells of the oligodendrocytelineage express both kainate- and AMPA-preferring subtypes of glutamate receptor. Neuron 12: 357-371.

Porter RHP and Greenamyre JT (1994) Regional variations in the pharmacology of AMPA receptors asrevealed by receptor autoradiography. Brain Res. 644: 202-206.

Rodriquez-Moreno A, Herreras O and Lerma J (1997) Kainate receptors presynaptically downregulateGABAergic inhibition in the rat hippocampus. Neuron 19: 893-901.

Rodriquez-Moreno A and Lerma J (1998) Kainate receptor modiulation of GABA release involves ametabotropic function. Neuron 20: 1211-1218.

Stühmer W (1992) Electrophysiological recording from Xenopus oocytes. Meth. Enzymol. 207: 319-339.

Vignes M and Collingridge GL (1997) The synaptic activation of kainate receptors. Nature 388: 179-182.

Wätjen F, Bigge CF, Jensen LH, Boxer PA, Lescosky LJ, Nielsen EO, Malone TC, Campbell GW,Coughenour LL, Rock DM, Drejer J and Marcoux FW (1994) NS-257 (1,2,3,6,7,8-hexahydro-3(hydroxyimino-N,N,7-trimethyl-2-oxobenzo(2,1-b:3,4-c’)dipyrrole-5-sulfonamide) is a potent,systemically active AMPA receptor antagonist. Bioorg. Med. Chem. Lett. 4: 371-376.

Wisden W and Seeburg P (1993) A complex mosaic of high-affinity kainate receptors exists in ratbrain. J. Neurosci. 13: 3582-3598.

Wong LA and Mayer ML (1993) Differential modulation by cyclothiazide and concanavalin A ofdesensitization at native α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid- and kainate-preferring glutamate receptors. Molec. Pharmacol. 44: 504-510.

Xue D, Haung Z-G, Barnes K, Lesiuk HJ, Smith KE and Buchan AM (1994) Delayed treatment withAMPA, but not NMDA, antagonists reduces neocortical infarction. J. Cereb. Blood Flow Metab. 14:251-261.

LIST OF PUBLICATIONS

Full papers:

1. Blank T, Zwart R, Nijholt I and Spiess J (1996) Serotonin 5-HT2 receptoractivation potentiates N-methyl-D-aspartate receptor-mediated ion currents by aprotein kinase C-dependent mechanism. J. Neurosci. Res. 45: 153-160.

2. Blank T, Nijholt I, Teichert U, Kuegler H, Behrsing H, Fienberg A, Greengard Pand Spiess J (1997) The phosphoprotein DARPP-32 mediates cAMP-dependentpotentiation of striatal N-methyl-D-aspartate responses. Proc. Natl. Acad. Sci. USA94: 14859-14864.

3. Nijholt I, Blank T, Grafelmann B, Cepok S, Kuegler H and Spiess J (1999) NS-257, a novel competitive AMPA receptor antagonist, interacts with kainate andNMDA receptors. Brain Res. 821: 374-382.

4. Radulovic J, Blank T, Nijholt I, Kammermeier J and Spiess J (2000) In vivoNMDA/dopamine interaction indicated by FOS production in the limbic system andbasal ganglia of the mouse brain. Molec. Brain Res. 75: 271-280.

5. Nijholt I, Blank T, Liu A, Kuegler H and Spiess J (2000) Modulation ofhypothalamic NMDA receptor function by cAMP-dependent protein kinase andphosphatases. J. Neurochem. 75:749-754.

6. Nijholt I, Blank T, Ahi J and Spiess J (2002) In vivo CREB phosphorylationmediated by dopamine and NMDA receptor activation in mouse hippocampus andcaudate nucleus. Brain Res. Gene Expression Patterns 1: 101-106.

7. Blank T, Nijholt I, Vollstaedt S and Spiess J (2002) The corticotropin-releasingfactor receptor 1 antagonist CP-154,526 markedly enhances associative learning andpaired- pulse facilitation immediately after a stressful experience. Submitted.

8. Nijholt I, Blank T, Eckart K and Spiess J (2002) Priming of long-termpotentiation in mouse hippocampus by corticotropin-releasing factor and acutestress: implications for hippocampus-dependent learning. J. Neurosci. 22: 3788-3794.

9. Blank T, Nijholt I, Grammatopoulos DK, Randeva HS, Hillhouse EW and SpiessJ (2002) Modulation of neuronal excitability and associative learning by differentcorticotropin-releasing factor-signaling cascades in mouse hippocampus. Submitted.

List of publications

180

Abstracts:

1. Blank T, Nijholt I, Behrsing H and Spiess J (1996) Modulation of NMDAreceptors by PKC and PKA systems in rat striatum and hippocampus. In:Proceedings of the 24th Göttingen Neurobiology Conference ; II, 629.

2 Blank T, Nijholt I, Behrsing H and Spiess J (1996) Modulation of NMDA receptorfunction in rat striatum-interplay of kinases and phosphatases. Soc. Neurosci. Abstr.22: 153.10.

3. Blank T, Nijholt I and Spiess J (1997) Inhibition of protein phosphatases isresponsible for cAMP-dependent potentiation of striatal NMDA responses. In:Proceedings of the 25th Göttingen Neurobiology Conference: II, 810.

4. Blank T, Nijholt I, Teichert U, Kuegler H and Spiess J (1997) PKA-dependentmodulation of NMDA responses: putative role of the phosphoprotein DARPP-32.Soc. Neurosci. Abstr. 23: 279.16.

5. Blank T, Nijholt I, Teichert U, Kuegler H and Spiess J (1997) Role of thephosphoprotein DARPP-32 in PKA-dependent modulation of striatal NMDAreceptor function. Satellite symposium of the International Congress ofBiochemistry, Cellular Regulation by Protein Phosphorylation, Seattle (abstract nrP7-2).

6. Radulovic J, Blank T, Kammermeier J, Nijholt I and Spiess J (1998) Modulationof NMDA-induced Fos production by dopaminergic mechanisms in mousehippocampus, amygdala and striatum. Soc. Neurosci. Abstr. 24: 823.10.

7. Blank T, Radulovic J, Nijholt I, Kammermeier J and Spiess J (1998) Modulationof dopamine-induced Fos production by NMDA, D1 and D2 receptor antagonists.Soc. Neurosci. Abstr. 24: 823.9.

8. Nijholt I, Blank T and Spiess J (1998) Convergent action of forskolin andcalyculin A on hypothalamic NMDA responses. Soc. Neurosci. Abstr. 24: 242.16.

9. Blank T, Nijholt I and Spiess J (1999) Hippocampal CA1 pyramidal cells fromC57BL/6N mice reveal high neuronal excitability. Soc. Neurosci. Abstr. 25: 399.17.

10. Nijholt I, Blank T, Ahi J and Spiess J (1999) CREB phosphorylation mediatedby dopamine D1 receptor in mouse hippocampus and caudate nucleus. Soc.Neurosci. Abstr. 25: 168.6.

11. Blank T, Nijholt I and Spiess J (2000) CRF enhances neuronal excitability ofmouse CA1 pyramidal neurons through CRF receptor 1. Soc. Neurosci. Abstr. 26:808.1.

List of publications

181

12. Nijholt I, Blank T and Spiess J (2000) Long-lasting efects of acute stress onCRFR1, pCREB and FOS in the mouse hippocampus. Soc. Neurosci. Abstr. 26:851.2.

13. Blank T, Nijholt I and Spiess J (2000) Acute Stress enhances neuronalexcitability of mouse CA1 Pyramidal Neurons. XXX 1st ISPNE meetingMelbourne, Australia.

14. Blank T, Nijholt I and Spiess J (2001) Facilitation of LTP by acute stress andcorticotropin-releasing factor (CRF) in mouse hippocampus: Implications forhippocampal-dependent memory. Soc. Neurosci. Abstr. 27: 919.5.

15. Nijholt I, Blank T and Spiess J (2001) The nonpeptide CRF receptor antagonistCP-154,526 enhances hippocampus-dependent learning and short-lived plasticityimmediately after exposure to stressful stimulus. Soc. Neurosci. Abstr. 27: 919.4.

DANKWOORDNa een fantastisch jaar als doctoraal studente op de afdeling Moleculaire

Neuroendocrinologie van het Max Planck Instituut voor ExperimenteleGeneeskunde in Göttingen, werd mij gevraagd of ik er voor voelde om mijnpromotie onderzoek op deze afdeling te doen. De beslissing was niet moeilijk en ikheb positief op dit aanbod gereageerd. Nu is het eindresultaat er. Ik wil met nameonderstaande mensen bedanken voor hun hulp, steun en vertrouwen.

Allereerst natuurlijk mijn oorspronkelijke promotor, wijlen professor BelaBohus. Zijn enthousiasme en onbegrensde wetenschappelijke kennis zijn ookvandaag nog voor mij een grote bron van inspiratie en motivatie. Na zijn plotselingoverlijden was professor Jaap Koolhaas direkt bereid de taak als promotor over tenemen. Daarvoor mijn dank.

An dieser Stelle gilt vor allem auch mein Dank Herrn Professor JoachimSpiess, daß er mir ermöglicht hat, die vorliegende Arbeit in seiner Abteilunganzufertigen.

De leescommissie bestaande uit professor Paul Luiten, professor MarianJoëls en professor Erik Boddeke, dank ik voor het beoordelen van mijn werk.

Ganz besonders danke ich Dr. Thomas Blank. Zusammen sind wir überhohe Berge und durch tiefe Täler gegangen.

Es wäre nicht möglich gewesen alle in dieser Arbeit beschriebenenExperimente auszuführen ohne die Hilfe von den technischen AssistentinnenStefanie Vollstädt, Heidemarie Kügler und Markus Hartmann und den studentischenHilfskräften, Jens Kammermeier, Allen Liu, Sabine Cepok, Eleanor Chen und BirgitGrafelmann.

Auch bei den anderen Mitgliedern der Abteilung 400 möchte ich michherzlich bedanken. Vor allem bei Karin Birkenfeld, Stefanie Vollstädt, BirgitGrafelmann, Sonja Kriks, Almuth Burgdorff, Janak Ahi, Anja Ronnenberg undMelanie Koch. Besonders erwähnen möchte ich auch Dr. Oliver Stiedl und Dr.Jelena Radulovic, die mir stets für hilfreiche Diskussionen zur Verfügung standen.

Wanneer je lang in het buitenland woont en werkt is het goed te weten, dater aan het thuisfront altijd mensen zijn die voor je klaar staan. Hoewel we somsweken of maanden geen kontakt hadden, waren jullie er op de belangrijke momentenaltijd weer voor me. Bedankt, Jeannine, Bea, Annette, Ticia en Sigrid!

Tenslotte denk ik dat dit het juiste moment is om mijn familie te bedanken.Zij zijn al die jaren, bij alles wat ik deed een grote steun geweest.

CURRICULUM VITAE

Ingrid M. Nijholt was born on June 19, 1973 in Drachten, the Netherlands.After attending primary and secondary school (Drachtster Lyceum in Drachten), shewent to the University of Groningen, where she studied biology. After finishing thecore courses she continued in the master’s program with a research project on thephysiological role of neurosteroids under supervision of Dr. Hans Beldhuis and Prof.dr. Bela Bohus at the Department of Animal Physiology.

During a one year stay at the Department of Molecular Neuroendocrinology,Max Planck Intitute in Goettingen she worked on the modulation of NMDAchannels by protein kinases and phosphatases with Dr. Thomas Blank and Prof. dr.Joachim Spiess. Ingrid received her Master’s degree cum laude in December 1996with the main subject Behavior and Neurosciences. In Februar 1997 she started withthe research project leading to this thesis at the Department of MolecularNeuroendocrinology, Max Planck Intitute, Goettingen. The work was supervised byProf. dr. Bela Bohus and Prof. dr. Jaap Koolhaas of the University of Groningen andDr. Thomas Blank and Prof. dr. Joachim Spiess of the Max Planck Institute inGoettingen.

Currently, Ingrid is working as a post-doc fellow in the same department.Her project is a follow-up of this thesis.

STELLINGENbehorende bij het proefschrift

MOLECULAR MECHANISMS OF SYNAPTIC PLASTICITY:Implications for immediate early gene expression and learning

Ingrid Nijholt

1. De regionale verschillen in de verdeling van zogenaamde ‚third messengers‘maken het mogelijk dat één en dezelfde receptor de synaptische plasticiteit inverscheidene hersengebieden verschillend beïnvloedt (dit proefschrift).

2. De activatie van ‚immediate early genes‘ is hersengebied-specifiekgereguleerd (dit proefschrift).

3. Verhoogde neuronale exciteerbaarheid correleert niet per definitie metverbeterd leervermogen (dit proefschrift).

4. Stress kan het leervermogen zowel verminderen als verbeteren afhankelijkvan het tijdsinterval tussen de stress en het aanleren van de taak (ditproefschrift).

5. Gezien de verschillen die er al bestaan tussen twee verschillende inbredmuizenstammen, moet men uiterst voorzichtig zijn met het generaliseren vanresultaten en uiteindelijk het overdragen daarvan naar de mens (n.a.v. ditproefschrift).

6. Everybody knows what stress is and nobody knows what it is (Hans Selye,1973).

7. In het kader van een verenigd Europa zou het zinvol zijn in alle landen eenzelfde procedure voor het verlenen van de doktorgraad in te stellen.

8. Het is te betreuren dat in de wetenschap tegenwoordig vaak de individueleambitie prevaleert boven het gezamenlijk bereiken van een gemeenschappelijkdoel.

9. De Nederlandse termen proefschrift en promotieonderzoek doenonvoldoende recht aan zowel de inhoud als de totstandkoming van hetvoorliggende product. De Duitse termen Doktorarbeit en WissenschaftlicheArbeit doen dit veel meer.

10. Hoe meer talen je beheerst, hoe meer je moet zoeken naar woorden.

11. Zelfs in Nederland kan men mountainbiken.

Documents

MOLECULAR MECHANISMS OF SYNAPTIC PLASTICITY · 2016. 3. 7. · Synaptic plasticity is a highly regulated process, emerging from complex interactions not only at the synapse itself