Novel Cyclic Adenosine Monophosphate (Camp) Signalling

University of Groningen

Novel cyclic AMP signalling avenues in learning and memory Ostroveanu, A

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below. Document Version Publisher's PDF, also known as Version of record

Publication date: 2009

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA): Ostroveanu, A. (2009). Novel cyclic AMP signalling avenues in learning and memory. s.n.

Copyright Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

The publication may also be distributed here under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license. More information can be found on the University of Groningen website: https://www.rug.nl/library/open-access/self-archiving-pure/taverne- amendment.

Take-down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Download date: 10-10-2021

Novel cyclic AMP signalling avenues in learning and memory

A. Ostroveanu

The studies described in this thesis were performed at the Department of Molecular Neurobiology, Faculty of Mathematics and Natural Sciences, University of Groningen. Financial support from the graduate school of Behavioral and Cognitive Neuroscience and the University of Groningen for the publication of this thesis is gratefully acknowledged.

school of behavioral and cognitive nenuroscience

Cover photos: A. Ostroveanu Cover design: A. Ostroveanu Printed by: Ipskamp Drukkers

RIJKSUNIVERSITEIT GRONINGEN

Novel cyclic AMP signalling avenues in learning and memory

Proefschrift

ter verkrijging van het doctoraat in de Wiskunde en Natuurwetenschappen aan de Rijksuniversiteit Groningen op gezag van de Rector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op vrijdag 1 mei 2009 om 16.15 uur

door

Anghelus Ostroveanu geboren op 25 januari 1976 te Ciocanesti, Roemenie

Promotores: Prof. dr. P.G.M. Luiten Prof. dr. E.A. van der Zee Prof. dr. U.L.M. Eisel

Copromotor: Dr. I.M. Nijholt

Beoordelingscommissie: Prof. dr. M. Schmidt Prof. dr. M. Verhage Prof. dr. B. Roozendaal

ISBN: 978-90-367-3750-0

A man's real possession is his memory. In nothing else he is rich, in nothing else he is poor. Dreamthorp: Essays written in the Country Alexander Smith-Scottish essayist & poet (1830 - 1867)

to my dear Oana and the loved ones left home

Contents

Chapter 1 Introduction and aim of the thesis 9

Chapter 2 A-kinase anchoring protein 150 in the mouse brain is concentrated 39 in areas involved in learning and memory

Chapter 3 Both exposure to a novel context and associative learning induce an 63 upregulation of AKAP150 protein in mouse hippocampus

Chapter 4 Inhibition of PKA anchoring to A-kinase anchoring proteins impairs 75 consolidation and facilitates extinction of contextual fear memories

Chapter 5 Detailed analysis of mAKAP expression in the brain of young and 95 old mice

Chapter 6 Exchange protein activated by cyclic AMP 2 (Epac2) plays a specific 119 and time-limited role in memory retrieval

Chapter 7 General discussion 141

Summary of the thesis 171

Nederlandse samenvatting 175

Acknowledgements 179

List of publications 181

Curriculum vitae 183

Chapter 1

Introduction and aim of this thesis

Chapter 1

Contents:

1. cAMP signaling in learning and memory 2. Cyclic AMP dependent protein kinase 3. PKA in learning and memory 4. Compartmentalization of PKA signaling through AKAPs 5. A-kinase anchoring protein 150 6. Muscle AKAP 7. Role of AKAPs in learning and memory 8. Tools for AKAP research 9. Exchange protein directly activated by cAMP 10. Aim and outline of the thesis 11. References

10 Introduction and aim of this thesis

1. cAMP signaling in learning and memory

In the molecular mechanisms that underlie the formation of all memories from simple adaptive responses to complex declarative memories, nerve cells and their interacting functional networks are subject to a large variety of signaling molecules including neurotransmitters, hormones, cytokines and matrix molecules. All these messengers can exert their influence to the same nerve cell. At the same time each nerve cell according to the model of Hebb can participate in different memory traces and respond to a variety of temporal and spatial dynamics. An obvious question then coming up is how a single individual neuron can handle the myriads of stimuli necessary to build memories during life, or in other words: how do nerve cells organize the where and when in their extensive cellular domain?

So far, a large number of signalling proteins have been shown to be involved in learning and memory processes. One of the most important molecules involved is cyclic adenosine monophosphate (cAMP, cyclic AMP or 3'-5'-cyclic adenosine monophosphate), a second messenger ubiquitously expressed and shown to be involved in many other biological processes besides learning and memory. This is e.g. evidenced by the fact that there are more than 90.000 scientific publications concerning cAMP since its discovery as a diffusible intracellular second messenger produced in response to hormone action (Sutherland and Rall, 1958). In the brain, activation of cAMP signaling occurs after stimulation of adenylyl cyclases by stimulatory G-proteins after binding of an extracellular ligand (e.g. adrenaline) to a GPCR and by Ca2+ through the Ca2+-binding protein calmodulin (Wang and Storm, 2003). Initially, the importance of cAMP signaling in learning and memory was demonstrated in genetic and pharmacological studies in invertebrate models for associative learning such as Aplysia (Byrne and Kandel, 1996), Drosophila (Waddell et al., 2000), and in honeybees (Apis mellifera) (Hammer and Menzel, 1995). Soon after, evidence for the role of cAMP in learning and memory in mammals was published. Genetic manipulations of several components of the cAMP pathway, such as augmentation of adenylyl cyclase activity (Pineda et al., 2004) or altering the expression of G-proteins (Bourtchouladze et al., 2006),

11 Chapter 1 led to severe memory impairments in mice. Although cAMP was acknowledged as crucial molecule in learning and memory, the precise role of cAMP in these processes remained unclear due to contradictory results from other pharmacological and genetic studies. While the elevation of cAMP levels in the hippocampus by forskolin, an activator of adenylyl cyclase, improved memory for passive avoidance tasks in rats (Bernabeu et al., 1997), pharmacological activation of the cAMP pathway in the prefrontal cortex was shown to impair working memory performance (Taylor et al., 1999). Very recently, it was shown that transgenic mice in which cAMP levels are spatially and temporally elevated in forebrain regions, exhibit enhanced memory consolidation and retrieval for contextual fear conditioning (Isiegas et al., 2008). Altogether, these results suggest that cAMP signals are mediated by different intracellular mechanisms and their actions dependent on the brain region involved. Thus cAMP appears to play a complex role in distinct cognitive processes in different brain regions.

In this thesis we will investigate the role of two, recently discovered, cAMP signaling avenues, A-kinase anchoring proteins (AKAPs) and exchange protein activated by cAMP (Epac), in memory processes.

2. Cyclic AMP dependent protein kinase

One of the main targets of cAMP is cAMP-dependent protein kinase (PKA). Mammalian PKA includes four regulatory (RIα, RIβ, RIIα, RIIβ) and three catalytic (Cα, Cβ, Cγ) subunits, each encoded by a separate gene (McKnight et al., 1988; Doskeland et al., 1993). Two major types of mammalian PKA, type I (PKA I, with RIα and RIβ dimers) and type II (PKA II, with RIIα and RIIβ dimers), were initially described by their pattern of elution from diethylaminoethylcellulose (DEAE) cellulose columns (Tasken et al., 1993, Francis and Corbin, 1999). Both type I and type II PKA are expressed in the mammalian brain (Cadd and McKnight, 1989). In the absence of cAMP, PKA consists of an inactive heterotetramer of two catalytic subunits bound to two regulatory subunits (Taylor et al, 1990). Each regulatory subunit contains two tandem cAMP binding sites, a high affinity

12 Introduction and aim of this thesis site and a low affinity site (Taylor et al., 1990). Activation of PKA occurs upon increasing cAMP levels. Accordingly, cAMP binding to these two sites located on the R subunits, results in the dissociation of the holoenzyme and the release of the monomeric C subunits which in turn can phosphorylate serine and threonine residues on targeted proteins (Taylor et al., 1990, Wang et al., 1991, Gibbs et al., 1992). The C subunits are broad-spectrum serine/threonine kinases that could potentially target numerous proteins, therefore signalling specificity for PKA is required. A basic specificity of PKA action is supported by the fact that there is a specific subcellular localization of PKA isoforms. For example, RI is found throughout the cytoplasm, whereas RII is localized to nuclei, nucleoli, the microtubule organizing center, the Golgi apparatus, and the plasma membrane (Griffiths et al., 1990, Li et al., 1988). Interestingly, even within the same subcellular compartment the PKA isoforms can have a differential localisation. For example RIIα and RIIβ have been demonstrated to localize differently in the Golgi-centrosomal area (Keryer et al., 1999). The specificity of PKA action can also be achieved by its affinity to cAMP. The PKA isoforms become activated at different concentrations of cAMP: type Iα is activated when the cAMP concentration reaches 10 nM, type Iβ at 100 nM, type IIα at 200 nM and type IIβ at 600 nM (Taylor et al, 1990). However, a more fine-tuned specificity for PKA action is provided by the family of A-kinase anchoring proteins (AKAP) which specifically tether PKA via the R subunits to unique subcellular sites within the cell. It is now well established that PKA regulates many vital processes through its reversible phosphorylation of proteins. These processes include cellular metabolism, gene expression, cell and tissue development, morphogenesis, ion channel conductivity, synaptic transmission, and cell motility (Sutherland, 1972, Scott, 119, Taylor et al., 1990).

3. PKA in learning and memory

Research into the signal transduction cascades involved in learning and memory has given us a more detailed understanding of how these processes take place in the brain and it became clear that protein kinases play a crucial role. Among these kinases, PKA emerged as an important molecule in both synaptic and behavioral changes necessary for learning

13 Chapter 1 and memory. Pioneering studies conducted in Kandel’s laboratory showed for the first time that stimulation of PKA was necessary for the consolidation of long term memory in Aplysia (Schacher et al., 1988). The involvement of PKA in learning and memory was demonstrated later on also in Drosophila. In a molecular-genetic approach, inducible transgene expression of a gene encoding a peptide inhibitor of PKA (an N-terminal regulatory subunit fragment containing a pseudosubstrate inhibitory domain, and a wild- type catalytic subunit) impaired associative learning in transgenic flies (Drain et al., 1991). Subsequently and not surprisingly, a large number of studies reported the involvement of PKA in learning and memory in various species. For example, Zhao and colleagues showed that inhibitors of PKA impair long-term memory formation in chicks (Zhao et al., 1995). Moreover, studies by Romano and colleagues showed that PKA plays a key role in long- term memory storage in long term habituation (a gradual decrease in the response to a repeated irrelevant stimulus) in the Chasmagnathus crab (Romano et al., 1996). Overall, these experiments have laid much of the conceptual foundations for the subsequent research on the role of PKA in learning and memory in the mammalian brain. The most compelling evidence for the role of PKA in learning and memory in rodents came from the lab of T. Abel. Using a genetic approach, Abel and colleagues generated transgenic mice that express a dominant negative form of the regulatory subunit of PKA in the postnatal excitatory neurons within the forebrain (Abel et al., 1997). These mice exhibited normal initial learning in the hidden platform version of the water maze, where mice learn to locate a submerged platform during repeated trials, but showed deficits in memory during the retrieval test (Abel et al., 1997). While these experiments clearly showed the involvement of PKA in learning, the repeated training trials do not allow a discrimination between short- term and long-term memory. However, behavioral tests such as contextual fear conditioning, a single trial learning task in which animals learn to associate fear with a particular neutral context (e.g. conditioning box), allow to discriminate between short and long-term memory. In fear conditioning, the regulatory subunit PKA dominant negative transgenic mice showed deficits in long term memory, whereas short-term memory was not affected (Abel et al., 1997, Bourtchouladze et al., 1998). Studies on both amnesic patients and experimental animals showed that the hippocampus is a brain structure instrumental in the formation of memories. By using the tetracycline inducible system, Isiegas and

14 Introduction and aim of this thesis colleagues showed that genetic inhibition of PKA in the hippocampus during adulthood selectively impairs contextual fear long-term memory (Isiegas et al., 2006). In addition, several pharmacological studies also provided evidence for a role of PKA in learning and memory. For example, intra-ventricular or intra-hippocampal injection of PKA inhibitors were reported to impair long term contextual memory (Bourtchouladze et al., 1998, Wallenstein et al., 2002, Ahi et al., 2004). Altogether, these data show a fundamental role for PKA in learning and memory.

4. Compartmentalization of PKA signaling through AKAPs

It recently emerged that multifunctional binding proteins oversee the dynamic organization of the when and where signaling events take place, by clustering activator proteins with enzymes such as kinases, phosphatases and phosphodiesterases and directing them towards their downstream effectors. Particularly AKAPs play a crucial role in the dynamic organization of cellular events. To date, more than 50 AKAPs have been identified in a wide range of species, tissues and cellular compartments (Fig. 1) (Angelo and Rubin, 2000, Jackson and Berg, 2002, Sarkar et al., 1984, Wong and Scott, 2004). All members of the AKAP family share three common characteristics. First of all, all AKAPs contain an amphipathic helix of 14-18 residues, which binds to the N-terminal dimerization and docking domain of the regulatory subunits of PKA (Carr et al., 1991; Herberg et al, 2000; Newlon et al., 2001). Second, AKAPs have a unique subcellular targeting domain which directs the PKA-AKAP complex to defined subcellular structures and finally, each AKAP can interact with several other signaling proteins such as protein phosphatases or various signal termination enzymes. The AKAP family can be divided in three groups on the basis of which PKA regulatory subunits they are able to bind. Although the majority of AKAPs binds the RII subunits of PKA, RI AKAPs or dual AKAP that bind both RI and RII were characterized. Interestingly, it is believed that up to 70% of the intracellular PKA is bound to AKAPs. AKAPs have e.g. been shown to participate in macromolecular signaling complexes that include protein kinases (serine/threonine and tyrosine kinases), phosphatases,

15 Chapter 1 phosphodiesterases (PDE), adenylyl cyclases, adaptor molecules or ion channels, and also at least one member of the superfamily of G-protein-coupled receptors (GPCR) (Wong and Scott, 2004). Moreover, recent data show a role of PKA anchoring in several physiological processes including glutamate receptor trafficking (Westphal et al., 1999), synaptic function (Rosenmund et al., 1994), hormone-mediated insulin secretion (Lester et al., 1997), learning (Moita et al., 2002), cardiomyocyte contractility (Fink et al, 2001) and vasopressin-mediated translocation of aquaporin-2 into cell membranes of renal principal cells (Klussmann et al., 1999).

Glutamate receptor Calcium channel

AKAP79/150 PKA PKC CaN AKAP15/18 PKA Vesicles MAP2 Microtubules Mitochondria PKA PKA AKAP220 DAKAP-1 Centrosome PKA PKA AKAP350/450 PP1 Nucleus PKC mAKAP Gravin PP1 PKA Yotiao PKA PKA PKA AKAP-KL NMDA receptor

Fig. 1 AKAPs in different subcellular compartments. All AKAPs share the same characteristics: they all bind PKA, have a unique subcellular targeting domain and form a signalosome (adapted from Edwards and Scott, Curr Opin Cell Biol, 2000)

It is generally believed that spatio-temporal configurations of distributed cAMP/PKA activity in the brain are a major factor in changes in the strength of synaptic contacts between nerve cells, also known as synaptic plasticity. Synaptic plasticity is widely considered to be the cellular mechanism that underlies cognitive processes. It emerges from changes in neuronal transmission, both in the short-term and the long-term. Short-term synaptic plasticity depends only on covalent modifications of pre-existing proteins (mainly

16 Introduction and aim of this thesis protein phosphorylation/dephosphorylation by protein kinases and phosphatases) whereas long-term synaptic changes require new protein synthesis and are associated with the growth of new synaptic connections. Although much work has been put in revealing the identity of the molecules involved in synaptic plasticity, we are only starting to discover the mechanisms by which the multifold of intracellular signals during cognitive processes are regulated and coordinated by AKAPs. In this thesis we focus on the role of two AKAPs in the brain: neuronal A-kinase anchoring protein 79/150 (AKAP79/150) and mAKAP. Whereas coordination of cAMP signaling by AKAP79/150 is mainly related to short-term synaptic plasticity (e.g. the dynamic protein phosphorylation of the AMPA receptor (Dell'Acqua et al., 2006), we have good reasons to assume that the recently identified mAKAP is a major player in processes that affect long- term synaptic plasticity such as the induction of gene expression. Both AKAPs are likely to play a role in coordinating synaptic plasticity and learning and memory considering their location and/or their signaling constituents.

5. A-kinase anchoring protein 150

One of the most prominently investigated AKAPs in the brain is AKAP79/150. This family of proteins consists of three orthologues: bovine AKAP75, murine AKAP150 and human AKAP79. Initially AKAP75 was identified as a contaminant of PKA RII purified from cytosolic brain preparations (Bregman et al., 1991; Sarkar et al., 1984). In addition, Bregman and colleagues obtained AKAP150 by screening a rat cDNA library using radiolabeled RIIβ as functional probe (Bregman et al., 1989). Finally AKAP79 was identified as a constituent of postsynaptic densities (PSD) in human cerebral cortex (Carr et al., 1992). All three proteins are highly related and differ only in their molecular weights, which is predominantly a consequence of repeat sequences present in the larger AKAP150. This prototypic anchoring protein consists of a positively charged N terminal region which is essential for its intracellular targeting, and of a high-affinity binding site for RIIβ near the C terminus. In addition to PKA-RIIβ, binding sites for protein kinase C (PKC) and protein phosphatase 2B (PP2B, calcineurin) were mapped. As a consequence, AKAP79/150

17 Chapter 1 complexes can respond to intracellular second messengers such as cAMP, calcium and phospholipids (Klauck et al., 1996) influencing various signaling events. It was showed that AKAP150 is abundantly expressed in the rat brain and is specifically enriched in postsynaptic densities (PSD) of excitatory glutamatergic synapses (Kennedy et al., 1997, Ziff et al., 1997, Yamauchi et al., 2002, Malenka et al., 2004). Here, AKAP79/150 targets its anchored proteins via postsynaptic density (PSD)-95 family membrane-associated guanylate kinase (MAGUK) scaffold proteins and forms a multi protein complex with AMPA and NMDA receptors (AMPAR and NMDAR) (Colledge et al., 2000), synaptic adhesion molecules, and cytoskeleton proteins that may play important roles in regulating synaptic structure and receptor function in synaptic plasticity. Although a few studies have been able to ascribe the regulation of specific neuronal events to a particular AKAP, electrophysiological studies established a role for AKAP79/150 in the modulation of ion channels such as AMPAR, L-type calcium channels and various potassium channels (Tavalin et al., 2002, Hoshi et al., 2003, Oliveria et al., 2007). In summary, the AKAP79/150 family of proteins emerged as key coordinators of various signaling events at postsynaptic densities in neurons which may affect both synaptic plasticity and learning and memory.

6. Muscle AKAP

Muscle AKAP (mAKAP), also known as AKAP100 or AKAP6, was first identified by expression cloning of a truncated cDNA that encoded a 100-kDa mAKAP fragment (McCartney et al., 1995). After further characterization of the protein, it was found that mAKAP was much larger than originally thought (Kapiloff et al., 1999; McCartney et al., 1995). Due to alternative splicing, two forms of mAKAP exist: mAKAPα and mAKAPβ. Both mAKAP variants are identical except for a 244 amino acid residue N-terminal extension in mAKAPα (Michel et al., 2005). Another difference between the mAKAP proteins resides in its localization. The longer form, mAKAPα, is preferentially expressed in the brain, whereas mAKAPβ is abundant in cardiac myocytes and skeletal muscle (Michel et al., 2005). In the heart mAKAP is targeted to the nuclear envelope through the

18 Introduction and aim of this thesis binding of three spectrin repeat domains, where it forms a large macromolecular signaling complex containing several signal transduction molecules, including PKA (Kapiloff et al, 1999, Kapiloff et al, 2001), the phosphodiesterase PDE4D3 (Dodge et al., 2001), the protein phosphatases PP2A and calcineurin (Kapiloff et al, 2001, Pare et al., 2005), ryanodine receptors (RyR2) (Kapiloff et al, 2001, Marx et al., 2000, Ruehr et al., 2003), the small GTPase Rap1, the guanine exchange factor Epac1 (Dodge-Kafka et al., 2005), and the mitogen-activated protein kinases (MAPK) MEK5 and ERK5 (Dodge-Kafka et al., 2005). Both in brain and heart, mAKAP can also bind 3’-phosphoinositide-dependent kinase 1 (PDK1) and p90 ribosomal S6 kinase (RSK3) (Michel et al., 2005). While in the heart mAKAP regulates cardiac function by converging MAPK, calcium, and cAMP signaling, the role of mAKAP in the brain remains so far rather unexplored. It is very likely that mAKAP in the brain has a similar function as in the heart namely the integration of cAMP, calcium and MAPK pathways to regulate neuronal processes. Besides its localization at the nuclear membrane, there are several lines of evidence indicating that mAKAP in the brain is involved in long-term synaptic plasticity and in this way affects cognition. Long-term changes in synaptic strength are mediated through alterations in gene expression via the activation of transcription factors such as NFAT and cAMP responsive element binding protein (CREB). Like in the heart, calcineurin, which is anchored to mAKAP, mediates NFAT translocation to the nucleus and this is critically dependent on increased intracellular calcium levels in neurons (Graeff et al., 1999). Furthermore, several effectors of the mAKAP complex including intracellular calcium release and the ERK pathway are known to affect PKA-induced CREB phosphorylation and CREB-dependent transcription (Zanassi et al., 2001). Thus it can be expected that mAKAP is crucial for coordinating gene expression at the nuclear membrane and in this way plays an important role in numerous brain functions such as learning and memory.

7. Role of AKAPs in learning and memory

Although extensive research has already been performed on the role of PKA in the molecular mechanisms of learning and memory, limited data exists on the role of anchoring

19 Chapter 1 proteins or anchored PKA. In addition to PKA, AKAPs provide platforms for the assembly of several other signaling enzymes known to be involved in learning and memory (e.g. PP2B, PP1 and PKC). As a consequence, this assembly of highly dynamic signaling enzymes within ionotropic and metabotropic glutamate receptors (AMPAR, NMDAR) and different subcellular compartments including synaptic vesicles, PSD and the cytoskeleton (reviewed by Wong and Scott, 2004) makes it likely that AKAPs are involved in learning and memory. In the brain, the majority of AKAPs tether the RII subunit of PKA. In the mammalian central nervous system, RIIβ (highly expressed in the amygdala and hippocampus) is the predominant isoform and principal mediator of cAMP action (Sarkar et al., 1984). Interestingly, mice with a targeted disruption of the RIIβ gene of PKA in the amygdala showed impaired long-term memory in a conditioned taste-aversion test. The conditioned taste-aversion test is an amygdala dependent classical conditioning test, in which lab animals learn to associate a novel taste of certain food with symptoms caused by a toxic, spoiled, or poisonous substance (Koh et al., 2003). Initial evidence for a role of AKAPs and PKA anchoring in learning and memory came from a study by Moita and colleagues, who blocked PKA anchoring in the lateral amygdala of rats and subjected the animals to auditory fear conditioning (Moita et al., 2002). Auditory fear conditioning is a classical Pavlovian associative emotional learning task in which mice are presented with an auditory cue (conditioned stimulus (CS)) followed by a short mild electric shock (unconditioned stimulus (US)). In the subsequent retention, if mice learn the task, they will exhibit a clear conditioned fear response, i.e. freezing behavior (defined as the lack of movement except for respiration and heart beat) to the auditory cue (CS). In Moita’s study, rats received bilateral infusions of St-Ht31 (a peptide that blocks PKA anchoring) or vehicle solution in the amygdala, 1 hour before auditory fear conditioning. The memory for the tone was assessed 1, 4 and 24 hours after training by measuring freezing behavior. Rats showed intact freezing 1 hour after training but impaired freezing both 4 and 24 hours after training, suggesting a role of PKA anchoring in the consolidation, but not the acquisition of auditory fear memory (Moita et al., 2002). More evidence for the role of anchored PKA in learning was provided by Schwaerzel and colleagues in Drosophila. They showed that, AKAP-bound PKA is required for aversive memory in a Pavlovian olfactory learning task, in which an electric shock is used as an

20 Introduction and aim of this thesis aversive unconditioned stimulus (Schwaerzel et al., 2007). Moreover, it has been shown that AKAPs play an important role in neuronal synaptic plasticity (Bauman et al, 2004). However, many questions still remained unanswered. For example, what is exactly the contribution of AKAPs and the AKAP signalosome to short-term and long-term memory? What is their contribution in the different stages of learning and memory (e.g. acquisition, consolidation, retrieval or extinction)? Are AKAPs involved in different memory systems (e.g. declarative memory, non-declarative memory)? Which AKAP is involved in learning and memory? Progress in the field of AKAPs will lead to integrated knowledge on AKAP scaffolding in cognitive processes under physiological and pathological conditions. Ultimately this knowledge on the coordination of signaling cascades may lead to the development of novel, more fine-tuned, innovative therapeutic strategies to treat cognitive dysfunction.

8. Tools for AKAP research

There are several ways to investigate the function of AKAPs and the tethering of signaling enzymes to AKAPs. It can be done by peptides that block the binding of enzymes to the complex or via genetic modifications which imply either to knock down of a particular AKAP or to generate truncated forms of a particular AKAP which lack the ability to bind particular signaling enzymes. To determine whether PKA anchoring to AKAPs has profound implications on PKA function, several peptides which are able to compete for PKA anchoring to AKAPs, were engineered. Carr and colleagues developed the first synthetic peptide covering amino acid residues 493 to 515 of the thyroid AKAP Ht31 (Carr et al., 1992). Ht31 is able to compete for both RIIα (Carr et al., 1992) and RIIβ as well as for RIα anchoring (Herberg et al., 2000). As an inactive control for this peptide a peptide named Ht31-P was developed. Ht31-P has two isoleucine residues substituted by proline residues which disrupt the amphipathic helix structure necessary for R-subunit binding. Although the Ht31 peptide proved to be an excellent tool to ascertain the role of PKA anchoring in various cellular processes, it remained unclear which of the PKA isoforms as well which AKAP was

21 Chapter 1 responsible. To overcome this limitation, PKA isoform specific anchoring competing peptides were developed. Using computer-based and peptide array screening approaches, Alto and colleagues generated a novel PKA binding competitor named AKAP-in silico, a high-affinity RII selective binding peptide (Alto et al., 2003). Studies aimed to develop even more potent and selective peptide anchoring competitors soon followed. Gold and colleagues developed superAKAP-IS, a peptide that is 10,000-fold more selective for the RII isoform relative to the RI subunit of PKA (Gold et al., 2006). Moreover, by using a combination of bioinformatics and peptide array screening Carlson and colleagues developed a high affinity-binding peptide called RIAD (RI anchoring disruptor) which is more than 1,000-fold more selective for type I PKA over type II PKA (Carlson et a., 2003). To date, there is a broad array of PKA anchoring disrupting peptides available (Hundsrucker et al., 2006) and numerous research groups showed that these peptides are valuable tools to study the relevance of PKA anchoring in various cellular processes. However, although PKA isoform specific anchoring disrupting peptides are available, these are not specific for a particular AKAP. The interaction of PKA RII subunits with various AKAPs is shown to involve an amphipathic helix motif of 14-18 residues with a conserved structure in the AKAP where the hydrophobic face binds the RII dimer (Carr et al., 1991; Vijayaraghavan et al., 1999). However, the number and distribution of hydrophobic amino acid residues in RII-binding domains is similar in all AKAPs (Hundsruker et al., 2006). Therefore, while numerous PKA anchoring peptides were designed, developing AKAP specific PKA anchoring disruptors remains very difficult (Hundsrucker et al., 2006). To be able to place a particular AKAP in a particular physiological response, AKAP specific PKA anchoring disruptors need to be developed. Unfortunately, except PKA anchoring disruptors, inhibitory anchoring peptides for other AKAP complex members such as e.g. protein phosphatases are still not available. AKAP site-specific mutants lacking the ability to anchor PKA and/or other signaling enzyme would lead to a better understanding of spatial and temporal organization of signaling events. In the lab of J. D. Scott, deletion mutants lacking enzyme-binding sites for protein kinases A (PKA), protein kinase C (PKC), or protein phosphatase 2B (PP2B) were already used to assess the role of AKAP79/150 signalosome in modulating neuronal ion channels activity (Hoshi and Scott,

22 Introduction and aim of this thesis

2006). Overall, further investigation into the role of AKAPs and anchoring of signaling enzymes will reveal novel roles for accurate intracellular signaling in learning and memory.

9. Exchange protein directly activated by cAMP

A long time it was believed that the major effects of cAMP in mammalian cells were mediated by PKA or cyclic-nucleotide gated ion channels (CNGs). However, fairly recently, a novel family of cAMP sensor proteins named exchange protein directly activated by cAMP (Epac) or cAMP-regulated guanine exchange factor (cAMP-GEF) was characterized (de Rooij et al, 1998, Kawasaki et al., 1998). This family consists of two multi-domain isoforms, named Epac1 (cAMP-GEF-I) and Epac2 (cAMP-GEF-II) which are products of independent genes in mammals. These proteins share extensive sequence homology and both contain an cAMP binding domain (CBD) that is homologous to that of PKA R subunits. It consists of an N-terminal regulatory region and a C-terminal catalytic region. While Epac1 has one CBD, Epac2 possesses a similar additional domain with a so far unknown biological function (Bos, 2006). Another difference between the two Epac variants arises from the expression of the proteins in various tissues. It has been shown that Epac1 mRNA is ubiquitously expressed in all tissues while Epac2 mRNA is predominantly expressed in adrenal glands and brain tissue (Kawasaki et al., 1998). In addition, fluorescent microscopy studies revealed that Epac is present in the nuclear membrane, plasma membrane and mitochondria and subcellular redistribution of Epac appears to be regulated during the cellular cycle (Qiao et al. 2002). By serving as an cAMP-binding protein, Epac1 and Epac2 couple cAMP production to the activation of Rap1 and Rap2, small molecular weight GTPases of the Ras family. To date, several studies demonstrated the role of Epac in various cellular processes such as cell adhesion (Enserink et al., 2004, Kooistra et al, 2005), exocytosis (Ozaki et al., 2000), secretion (Kiermayer et al., 2005), cell differentiation (Bryn et al., 2006; Shi et al. 2006), cell proliferation (Misra and Pizzo 2005), apoptosis (Kwon et al., 2004), gene expression (Sands et al., 2006, Ulucan et al., 2007) and cardiac hypertrophy (reviewed by Roscioni et al., 2008). Despite of the growing body of literature, the role of Epac in the brain remained

23 Chapter 1 largely unknown. It is surprising that in view of the role of cAMP in brain functions (e.g.: axon growth and regeneration, addiction, synaptic, plasticity, and memory), few data are available on the role of Epac in the central nervous system. Epac was shown to enhance neurotransmitter release in glutamatergic synapses (Sakaba and Neher, 2003; Zhong and Zucker, 2005; Gekel and Neher, 2008), whereas in cultured cerebellar granule cells it can modulate neuronal excitability via Rap and p38 MAPK activation (Ster et al., 2007). Moreover, in dorsal root ganglion Epac mediates the translocation and activation of protein kinase C (PKC) leading to the establishment of inflammatory pain (Hucho et al., 2005). Thus far, evidence for a role of Epac in the process of learning and memory is limited. However, since Epac is a cAMP-responsive enzyme and cAMP signaling is established to be of critical importance in learning and memory, an involvement of PKA-independent cAMP signaling through Epac proteins can be expected. The first indications for a role of Epac in hippocampus-dependent learning and memory came from very recent studies. Gelinas and colleagues reported that Epac activation enhances the maintenance of LTP in area CA1 of mouse hippocampal slices (Gelinas et al., 2008) and co-application of a selective PKA and a selective Epac activator was shown to rescue the memory retrieval impairment observed in dopamine-beta-hydroxylase deficient mice (Ouyang et al., 2008). These initial findings indicate a likely and important role for Epac in learning and memory processes. However, the contribution of Epac1 and Epac2 in the molecular machinery of learning and memory remains to be elucidated. As such, the involvement of this additional highly coordinated cAMP effector in learning and memory represent a major research interest for future cAMP-mediated signaling studies.

10. Aim and outline of the thesis

Memory is an amazing feature of the (human) brain. It is not only defining who we are but it will also dictate who and what we are going to become. Memory has always fascinated human kind. E.g. before Christ the Greek goddess of memory was called Mnemosyne. Over time, a large number of philosophers studied the nature of memory. However, the modern

24 Introduction and aim of this thesis research on learning and memory started in 1887 with the work of Joseph Jacobs on short term memory (Jacobs, 1887). Although advances in the field of learning and memory have been made due to behavioral, electrophysiological, molecular, pharmacological and biochemical experiments, the quintessence of how the brain functions in learning and memory remains elusive. Memory and learning are so closely connected that people often confuse them with each other. However, learning and memory are not unitary processes. Learning is the process by which new information is acquired; memory is the process by which that knowledge is retained and remembered. For example: you learn a new language by studying it, but you then speak it by using your memory to retrieve the words that you have learned. The major aim of this thesis is to conduct fundamental research in the field of learning and memory. We put a specific emphasis on the role of the second messenger cyclic AMP (3'- 5'-cyclic adenosine monophosphate) in the molecular mechanisms underlying memory formation. Understanding this complex process does not only provide fundamental insights, but will also lead to the development of new therapeutic strategies for improving cognition in both neurodegenerative disease and in normal age-dependent cognitive decline.

The main downstream effector of cAMP is PKA (cyclic AMP-dependent protein kinase). PKA was found to be involved in stimulus-induced changes in synaptic strength, a process called synaptic plasticity which is believed to provide at least in part the cellular basis of learning and memory. Moreover, PKA was established to be a key enzyme in learning and memory processes. However, it remained unclear for a long time how such a broad enzyme can discriminate between different stimuli and respond accordingly. Fairly recently a key mechanism by which PKA achieves signaling specificity was reported. Specificity resides from subcellular targeting of PKA through scaffolding or adaptor proteins. Indeed, it is believed that up to 70% of the intracellular PKA is bound to a family of proteins called A- kinase anchoring proteins (AKAP). One of the AKAPs expressed in the rodent brain is AKAP150. In addition to PKA, AKAP150 was found to target protein kinase C (PKC) and protein phosphatase 2B (PP2B) to the neuronal membrane, particularly of synapses (Carr et al., 1992).

25 Chapter 1

Although it is widely acknowledged that all the complex members of AKAP150 play a crucial role in synaptic plasticity and learning and memory, little is known about the role of AKAP150 itself in these processes. Therefore, our first aim was to investigate the role of AKAP150 in learning and memory. To get a first indication of the function of AKAP150 we investigated the distribution of AKAP150 in the mouse brain (Chapter 2). Interesting, AKAP150 mRNA was shown to be upregulated in the hippocampus, 3-12 hr after the induction of LTP (long term potentiation), a long-lasting enhancement in synaptic efficacy (Genin et al., 2003). This hippocampus is a brain structure known to be critically involved in memory formation. In chapter 3 we raised the question whether AKAP150 protein levels in the mouse hippocampus, change during learning. For this, we used a learning paradigm, named contextual fear conditioning, in which mice learn to fear the context (conditioning box) in which the experiment is performed. This is done by pairing the conditioning box with an aversive mild electric foot shock (Fig. 2A). Eventually, the conditioning box alone can elicit the state of fear. Conditioned fear reflects emotional associative memory in mice and is measured as freezing, immobility-like behavior, except the movement created by breathing.

Besides determining whether the levels of AKAP150 change in learning and memory, it is at least as interesting to explore whether the anchoring of PKA to AKAP is important in learning and memory. Therefore, in chapter 4 we investigated the importance of PKA anchoring to AKAPs in the different stages of contextual fear conditioning: acquisition (encoding of memory), consolidation (storing and maintaining of memory), retrieval (accessing of memory) and extinction (“erasing” of memory). To address this question, we employed a pharmacological approach. Mice were injected intracerebroventricularly or intrahippocampally with membrane permeable PKA anchoring disrupting peptides at different time points during the memory process (Fig. 2B). To date, the AKAP family of proteins consists of more than 50 members and several of these AKAPs have been found in the brain. Initially, muscleAKAP (mAKAP) was reported to be present in the heart and skeletal muscles where it coordinates cAMP, calcium and MAPK (mitogen-activated protein kinase) signaling (Dodge-Kafka and Kapiloff, 2006). However, mRNA studies showed that a splice variant of mAKAP is also expressed in the

26 Introduction and aim of this thesis brain. Because mAKAP signaling complex members in the heart are involved in numerous brain functions, it is very likely that in the brain, mAKAP also integrates several signaling pathways to regulate neuronal processes. In chapter 5 we aimed to answer the following questions: What is the brain distribution of mAKAP? Does mAKAP expression changes during ageing? What may be the function of mAKAP in the brain?

A Acquisition Consolidation Retrieval

Context Shock Return to homecage Context

B Training -1h injection 2” Day 1-Training Day 2-Retention Acquisition 180” 30” 180”

After training injection 2” Day 1-Training Day 2-Retention Consolidation 180” 30” 180”

Retention -1h injection 2” Day 1-Training Day 2-Retention Retrieval 180” 30” 180”

Fig. 2 Contextual fear conditioning experimental setup. A. Different stages of fear memory can be distinguished: acquisition, consolidation and retrieval. B. Acquisition, consolidation and retrieval can be affected by injecting pharmacological compounds in the mouse brain at different time points during fear conditioning (adapted from Abel and Lattal, Curr Opin Neurobiol, 2001).

Although PKA was long thought to be the main effector of cAMP, fairly recently another cAMP effector was identified. cAMP can also activate Rap-guanine nucleotide exchange factor protein directly activated by cAMP (Epac). Because cAMP signaling is crucial in learning and memory, an involvement of PKA-independent cAMP signaling through Epac proteins can be expected. In chapter 6 of this thesis, we explored the role of Epac in learning and memory by pharmacological activation or gene silencing of Epac in the mouse brain in several learning paradigms. Accordingly, a cAMP analogue which specifically activates Epac but not PKA was delivered into the mouse brain and its role in the acquisition, consolidation, retrieval and extinction of contextual fear conditioning was

27 Chapter 1 determined. The data found in contextual fear conditioning were confirmed in passive avoidance, another fear motivated learning task. To exclude effects of Epac activation on anxiety we also investigated the effect of Epac activation on axiety behavior in the elevated plus maze. From the distribution of the two Epac isoforms, Epac1 and Epac2, we deduced that Epac2 is abundant in the mouse brain whereas Epac1 is hardly present. Therefore, we studied the role of reduced Epac2 expression in immediate and delayed fear memory.

Chapter 7 summarizes and discusses all findings of this thesis and provides an overall conclusion and future perspectives.

11. References

Abel, T., Nguyen, P.V., Barad, M., Deuel, T.A., Kandel, E.R., & Bourtchouladze, R. (1997). Genetic demonstration of a role for PKA in the late phase of LTP and in hippocampus-based long-term memory. Cell 88, 615-626.

Ahi, J., Radulovic, J., & Spiess, J. (2004). The role of hippocampal signaling cascades in consolidation of fear memory. Behavioral Brain Research 149, 17-31.

Alto, N.M., Soderling, S.H., Hoshi, N., Langeberg, L.K., Fayos, R., Jennings, P.A., & Scott, J.D. (2003). Bioinformatic design of A-kinase anchoring protein-in silico: a potent and selective peptide antagonist of type II protein kinase A anchoring. Proceedings of the National Academy of Sciences United States Of America 8, 4445-450.

Angelo, R.G., & Rubin, C.S. (2000). Characterization of structural features that mediate the tethering of Caenorhabditis elegans protein kinase A to a novel A kinase anchor protein. Insights into the anchoring of PKAI isoforms. Journal of Biological Chemistry 275, 4351-62.

Bauman, A.L., Goehring, A.S. & Scott, J.D. (2004). Orchestration of synaptic plasticity through AKAP signaling complexes. Neuropharmacology 46, 299-310.

Bernabeu, R., Bevilaqua, L., Ardenghi, P., Bromberg, E., Schmitz, P., Bianchin, M., Izquierdo, I., & Medina, J.H. (1997). Involvement of hippocampal cAMP/cAMP-dependent protein kinase signaling

28 Introduction and aim of this thesis pathways in a late memory consolidation phase of aversively motivated learning in rats. Proceedings of the National Academy of Sciences United States Of America 94, 7041-7046.

Bos J.L. (2006). Epac proteins: multi-purpose cAMP targets. Trends in Biochemical Sciences 31, 680- 686

Bourtchouladze, R., Abel, T., Berman, N., Gordon, R., Lapidus, K., & Kandel, E.R. (1998). Different training procedures recruit either one or two critical periods for contextual memory consolidation, each of which requires protein synthesis and PKA. Learning and Memory 5, 365-374.

Bourtchouladze, R., Patterson, S.L., Kelly, M.P., Kreibich, A., Kandel, E.R., & Abel, T. (2006). Chronically increased Gsalpha signaling disrupts associative and spatial learning. Learning and Memory 13, 745-752.

Bregman, D.B., Bhattacharyya, N., & Rubin, C.S, (1989). High affinity binding protein for the regulatory subunit of cAMP-dependent protein kinase II-β. Journal of Biological Chemistry 264, 4648-4656.

Bregman, D.B., Hirsch, A.H., & Rubin, C.S. (1991). Molecular characterization of bovine brain P75, a high affinity binding protein for the regulatory subunit of cAMP-dependent protein kinase II beta. Journal of Biological Chemistry 266, 7207-7213.

Bryn, T., Mahic, M., Enserink, J.M., Schwede, F., Aandahl, E.M., & Tasken, K. (2006). The cyclic AMP-Epac1-Rap1 pathway is dissociated from regulation of effector functions in monocytes but acquires immunoregulatory function in mature macrophages. The Journal of Immunology 176, 7361- 7370.

Byrne, J.H. & Kandel, E.R. (1996). Presynaptic facilitation revisited: state and time dependence. The Journal of Neuroscience 16, 425-435.

Cadd, G., & McKnight, G.S. (1989). Distinct patterns of cAMP-dependent protein kinase gene expression in mouse brain. Neuron 3, 71-79.

Carlson, C.R., Ruppelt, A., & Taskén. K. (2003). A kinase anchoring protein (AKAP) interaction and dimerization of the RIalpha and RIbeta regulatory subunits of protein kinase a in vivo by the yeast two hybrid system. Journal of Molecular Biology 3, 609-618.

29 Chapter 1

Carr, D.W., Stofko-Hahn, R.E., Fraser, I.D., Bishop, S.M., Acott, T.S., Brennan, R.G., & Scott, J.D. (1991). Interaction of the regulatory subunit (RII) of cAMP-dependent protein kinase with RII- anchoring proteins occurs through an amphipathic helix binding motif. Journal of Biological Chemistry 266, 14188-14192.

Carr, D.W., Hausken, Z.E., Fraser, I.D.C., Stofko-Hahn, R.E., & Scott, J.D. (1992). Association of the Type I1 CAMP-dependent Protein Kinase with a Human Thyroid RII-anchoring Protein. Cloning and characterization of the RII-binding domain. Journal of Biological Chemistry 267, 13376-13382.

Carr, D.W., Stofko-Hahn, R.E., Fraser, I.D.C., Cone, R.D., & Scott, J.D. (1992). Localization of the cAMP-dependent Protein Kinase to the Postsynaptic Densities by A-Kinase Anchoring Proteins. Journal of Biological Chemistry 24, 16816-16823.

Colledge, M., Dean, R.A., Scott, G.K., Langeberg, L.K., Huganir, R.L., & Scott, J.D. (2000). Targeting of PKA to glutamate receptors through a MAGUK-AKAP complex. Neuron 27, 107-119.

de Rooij, J., Zwartkruis, F.J., Verheijen, M.H., Cool, R.H., Nijman, S.M., Wittinghofer, A., & Bos, J.L. (1998). Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 396, 474-477.

Dell'Acqua, M.L., Smith, K.E., Gorski, J.A., Horne, E.A., Gibson, E.S., & Gomez, L.L. (2006). Regulation of neuronal PKA signaling through AKAP targeting dynamics. European Journal of Cell Biology 85, 627-33.

Dodge, K.L., Khouangsathiene, S., Kapiloff, M.S., Mouton, R., Hill, E.V., Houslay, M.D., Langeberg, L.K., and Scott, J.D. (2001). mAKAP assembles a protein kinase A/PDE4 phosphodiesterase cAMP signaling module. EMBO Journal 20, 1921-1930.

Dodge-Kafka, K.L., Soughayer, J., Pare, G.C., Carlisle Michel, J.J., Langeberg, L.K., Kapiloff, M.S., & Scott, J.D. (2005). The protein kinase A anchoring protein mAKAP coordinates two integrated cAMP effector pathways. Nature 437, 574-578.

Dodge-Kafka, K.L., & Kapiloff, M.S. (2006). The mAKAP signaling complex: Integration of cAMP, calcium and MAP kinase signaling pathways. European Journal of Cell Biology 7, 593-602.

Døskeland, S.O., Maronde, E., Gjertsen, B.T.( 1993). The genetic subtypes of cAMP-dependent protein kinase - functionally different or redundant? Biochimica et Biophysica Acta 3, 249-258.

30 Introduction and aim of this thesis

Drain, P., Folkers, E., & Quinn, W.G. (1991). cAMP-dependent protein kinase and the disruption of learning in transgenic flies. Neuron 1, 71-82.

Enserink, J.M., Price, L.S., Methi, T., Mahic, M., Sonnenberg, A., Bos, J.L., & Taskén, K. (2004). The cAMP-Epac-Rap1 pathway regulates cell spreading and cell adhesion to laminin-5 through the alpha3beta1 integrin but not the alpha6beta4 integrin. Journal of Biological Chemistry 279, 44889- 44896.

Fink, M.A., Zakhary, D.R., Mackey, J.A., Desnoyer, R.W., Apperson-Hansen, C., Damron, D.S., & Bond, M. (2001). AKAP-mediated targeting of protein kinase A regulates contractility in cardiac myocytes. Circulation Research 88, 291-297.

Francis, S. H., & Corbin, J. D. (1999). Cyclic nucleotide-dependent protein kinases: intracellular receptors for cAMP and cGMP action. Critical Reviews in Clinical Laboratory Sciences 36, 275-328.

Gekel, I., & Neher, E. (2008). Application of an Epac activator enhances neurotransmitter release at excitatory central synapses. Journal of Neuroscience 28, 7991-8002.

Gelinas, J.N., Banko, J.L., Peters, M.M., Klann, E., Weeber, E.J., & Nguyen, P.V. (2008). Activation of exchange protein activated by cyclic-AMP enhances long-lasting synaptic potentiation in the hippocampus. Learning and Memory 15, 403-411.

Genin, A., French P., Doyere, V., Davis, S., Errington, M.L., Maroun, M., Stean, T., Truchet, B., Webber, M., Wills, T., Richter-Levin, G., Sanger, G., Hunt, S.P., Mallet, J., Laroche, S., Bliss, T.V., & O'Connor, V., (2003). LTP but not seizure is associated with up-regulation of AKAP-150. European Journal of Neuroscience 17, 331-40.

Gold, M.G., Lygren, B., Dokurno, P., Hoshi, N., McConnachie, G., Tasken, K., Carlson, C.R., Scott, J.D., & Barford, D. (2006). Molecular basis of AKAP specificity for PKA regulatory subunits. Molecular Cell 24, 383-395.

Graeff, I.A., Mermelstein, P.G., Stankunas, K., Neilson, J.R., Deisseroth, K., Tsien, R.W., & Crabtree G.R. (1999). L-type calcium channels and GSK-3 regulate the activity of NF-ATc4 in hippocampal neurons. Nature 401, 703-708.

Griffiths, G., Hollinshead, R., Hemmings, B. A., & Nigg, E. A. (1990) .Ultrastructural localization of the regulatory (RII) subunit of cyclic AMP-dependent protein kinase to subcellular compartments active in endocytosis and recycling of membrane receptors. Journal of Cell Science 96, 691-703.

31 Chapter 1

Hammer, M. & Menzel R. (1995). Learning and memory in the honeybee. The Journal of Neuroscience 15, 1617-1630.

Herberg, F.W., Maleszka, A., Eide, T., Vossebein, L., & Tasken K. (2000). Analysis of A-kinase anchoring protein (AKAP) interaction with protein kinase A (PKA) regulatory subunits: PKA isoform specificity in AKAP binding. Journal of Molecular Biology 298, 329-339.

Hoshi, N., & Scott, J.D. (2006). Contextual utilization of enzymes in discrete AKAP79/ 150 signalling complexes. European Journal of Cell Biology 85, 621-622.

Hoshi, N., Zhang, J.S., Omaki, M., Takeuchi, T., Yokoyama, S., Wanaverbecq, N., Langeberg, L.K., Yoneda, Y., Scott, J.D., Brown, D.A., & Higashida, H. (2003). AKAP150 signaling complex promotes suppression of the M-current by muscarinic agonists. Nature Neuroscience 6, 564-571.

Hucho, T.B., Dina, O.A., & Levine, J.D. (2005). Epac mediates a cAMP-to-PKC signaling in inflammatory pain: an isolectin B4(+) neuron-specific mechanism. Journal of Neuroscience 25, 6119-6126.

Hundsrucker, C., Krause, G., Beyermann, M., Prinz, A., Zimmermann, B., Diekmann, O., Lorenz, D., Stefan, E., Nedvetsky, P., Dathe, M., Christian, F., McSorley, T., Krause, E., McConnachie, G., Herberg, F.W., Scott, J.D., Rosenthal, W., & Klussmann, E. (2006). High-affinity AKAP7delta- protein kinase A interaction yields novel protein kinase A-anchoring disruptor peptides. Biochemical Journal 2, 297-306.

Isiegas, C., Park, A., Kandel, E.R., Abel, T., & Lattal, K.M. (2006). Transgenic inhibition of neuronal protein kinase A activity facilitates fear extinction. Journal of Neuroscience 26, 12700-12707.

Isiegas, C., McDonough, C., Huang, T., Havekes, R., Fabian, S., Wu, L.J., Xu, H., Zhao, M.G., Kim, J.I., Lee, Y.S., Lee, H.R., Ko, H.G., Lee, N., Choi, S.L., Lee, J.S., Son, H., Zhuo, M., Kaang, B.K., & Abel, T. (2008). A novel conditional genetic system reveals that increasing neuronal cAMP enhances memory and retrieval. Journal of Neuroscience 24, 6220-6230.

Jackson, S.M., & Berg, C.A., (2002). An A-kinase anchoring protein is required for protein kinase A regulatory subunit localization and morphology of actin structures during oogenesis in Drosophila. Development 129, 4423-4433.

Jacobs, J. (1887). Experiments on prehension. Mind 12, 75-79.

32 Introduction and aim of this thesis

Kapiloff, M.S., Schillace, R.V., Westphal, A.M., & Scott, J.D. (1999). mAKAP: an A-kinase anchoring protein targeted to the nuclear membrane of differentiated myocytes. Journal of Cell Science 112, 2725-2736.

Kapiloff, M.S., Jackson, N., & Airhart, N. (2001). mAKAP and the ryanodine receptor are part of a multi-component signaling complex on the cardiomyocyte nuclear envelope. Journal of Cell Science 17, 3167-3176

Kawasaki, H., Springett, G.M., Mochizuki, N., Toki, S., Nakaya, M., Matsuda, M., Housman, D.E., & Graybiel, A.M. (1998). A family of cAMP-binding proteins that directly activate Rap1. Science 282, 2275-2279.

Kennedy, M.B., (1997). The postsynaptic density at glutamatergic synapses. Trends in Neurosciences 20, 264-268.

Keryer, G., Skålhegg, B.S., Landmark, B.F., Hansson, V., Jahnsen, T., & Taskén, K. (1999). Differential localization of protein kinase A type II isozymes in the Golgi-centrosomal area. Experimental Cell Research 1, 131-146.

Kiermayer, S., Biondi, R.M., Imig, J., Plotz, G., Haupenthal, J., Zeuzem, S., & Piiper, A. ( 2005). Epac activation converts cAMP from a proliferative into a differentiation signal in PC12 cells. Molecular Biology of the Cell 16, 5639-5648.

Klauck, T.M., Faux, M.C., Labudda, K., Langeberg, L.K., Jaken, S., & Scott, J.D. (1996). Coordination of three signaling enzymes by AKAP79, a mammalian scaffold protein. Science 271, 1589-1592.

Klussmann, E., Maric, K., Wiesner, B., Beyermann, M., Rosenthal, W. (1999). Protein kinase A anchoring proteins are required for vasopressin-mediated translocation of aquaporin-2 into cell membranes of renal principal cells. Journal of Biological Chemistry 8, 4934-4938.

Koh, M.T., Clarke, S.N., Spray, K.J., Thiele, T.E., & Bernstein, I.L. (2003). Conditioned taste aversion memory and c-Fos induction are disrupted in RIIbeta-protein kinase A mutant mice. Behavioural Brain Research 1, 57-63.

Kooistra, M.R., Corada, M., Dejana, E., & Bos, J.L. (2005). Epac1 regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Letters 579, 4966-4972.

33 Chapter 1

Kwon, G., Pappan, K.L., Marshall, C.A., Schaffer, J.E., & McDaniel, M.L. (2004). cAMP Dose- dependently prevents palmitate-induced apoptosis by both protein kinase A- and cAMP-guanine nucleotide exchange factor-dependent pathways in beta-cells. Journal of Biological Chemistry 279, 8938-8945.

Lester, L.B., Langeberg, L.K., & Scott, J.D. (1997). Anchoring of protein kinase A facilitates hormone-mediated insulin secretion. Proceedings of the National Academy of Sciences United States Of America 26, 14942-14947.

Li, M., McCann, J.D., Liedtke, C.M., Nairn, A.C., Greengard, P. & Welsh, M.J. (1988). Cyclic AMP- dependent protein kinases open chloride channels in normal but not in cystic fibrosis airway epithelium. Nature 331, 358-360.

Malenka, R.C., & Bear, M.F. (2004). LTP and LD: an embarrassment of riches. Neuron 44, 5-21.

Marx, S.O., Reiken, S., Hisamatsu, Y., Jayaraman, T., Burkhoff, D., Rosemblit, N., & Marks, A. R. (2000). PKA phosphorylation dissociates FKBP12.6 from the calcium release channel (ryanodine receptor): defective regulation in failing hearts. Cell 101, 365-376.

McCartney, S., Little, B.M., Langeberg, L.K., & Scott, J.D. (1995). Cloning and characterization of A-kinase anchor protein 100 (AKAP100). A protein that targets A-kinase to the sarcoplasmic reticulum. Journal of Biological Chemistry 270, 9327-9333.

McKnight, G.S., Clegg, C.H., Uhler, M.D., Chrivia, J.C., Cadd, G.G., Correll, L.A., & Otten, A.D. (1988). Analysis of the cAMP-dependent protein kinase system using molecular genetic approaches. Recent Progress in Hormone Research 44, 307-35.

Miche,l J.J., Townley, I.K., Dodge-Kafka, K.L., Zhang, F., Kapiloff, M.S., & Scott, J.D. (2005). Spatial restriction of PDK1 activation cascades by anchoring to mAKAPalpha. Molecular Cell 20, 661-672.

Misra, U.K., & Pizzo, S.V. (2005). Coordinate regulation of forskolininduced cellular proliferation in macrophages by protein kinase A/cAMP-response element-binding protein (CREB) and Epac1-Rap1 signaling: effects of silencing CREB gene expression on Akt activation. Journal of Biological Chemistry 280, 38276-38289.

Moita, M.A., Lamprecht, R., Nader, K., & LeDoux, J.E., (2002). A-kinase anchoring proteins in amygdala are involved in auditory fear memory. Nature Neuroscience 5, 837-838.

34 Introduction and aim of this thesis

Newlon, M.G., Roy, M., Morikis, D., Carr, D.W., Westphal, R., Scott, J.D., & Jennings, P.A. (2001). A novel mechanism of PKA anchoring revealed by solution structures of anchoring complexes. EMBO Journal 20, 1651-1662.

Oliveria, S.F., Dell’Acqua, M.L., Sather, W.A. (2007). AKAP79/150 anchoring of calcineurin controls neuronal L-type Ca2+ channel activity and nuclear signaling. Neuron 55, 261-275.

Ouyang, M., Zhang, L., Zhu, J.J., Schwede, F., & Thomas, S.A. (2008). Epac signaling is required for hippocampus-dependent memory retrieval. Proceedings of the National Academy of Sciences United States of America 105, 11993-11997.

Ozaki, N., Shibasaki, T., Kashima, Y., Miki, T., Takahashi, K., Ueno, H., Sunaga, Y., Yano, H., Matsuura, Y., Iwanaga, T., Takai, Y., & Seino S. (2000). cAMP-GEFII is a direct target of cAMP in regulated exocytosis. Nature Cell Biology 2, 805-811.

Pare, G.C., Bauman, A.L., McHenry, M., Michel, J.J., Dodge-Kafka, K.L., & Kapiloff, M.S. (2005). The mAKAP complex participates in the induction of cardiac myocyte hypertrophy by adrenergic receptor signaling. Journal of Cell Science 118, 5637-5646.

Pineda, V.V., Athos, J.I., Wang, H., Celver, J., Ippolito, D., Boulay, G., Birnbaumer, L., & Storm, D.R. (2004). Removal of G(ialpha1) constraints on adenylyl cyclase in the hippocampus enhances LTP and impairs memory formation. Neuron 41, 153-163.

Qiao, J., Mei, F.C., Popov, V.L., Vergara, L.A., & Cheng, X. (2002). Cell cycledependent subcellular localization of exchange factor directly activated by cAMP. Journal of Biological Chemistry 277, 26581-26586

Romano, A., Locatelli, F., Delorenzi, A., Pedreira, M.E., & Maldonado, H. (1996). Effects of activation and inhibition of cAMP-dependent protein kinase on long-term habituation in the crab Chasmagnathus. Brain Research 735, 131-140.

Roscioni, S.S., Elzinga, C.R., & Schmidt, M. (2008). Epac: effectors and biological functions. Naunyn-Schmiedeberg's Archives of Pharmacology 377, 345-357.

Rosenmund, C., Carr, D.W., Bergeson, S.E., Nilaver, G., Scott, J.D., & Westbrook, G.L., (1994). Anchoring of protein kinase A is required for modulation of AMPA/kainate receptors on hippocampal neurons. Nature 368, 853-856.

35 Chapter 1

Ruehr, M.L., Russell, M.A., Ferguson, D.G., Bhat, M., Ma, J., Damron, D.S., Scott, J.D., & Bond, M. (2003). Targeting of protein kinase A by muscle A kinase-anchoring protein (mAKAP) regulates phosphorylation and function of the skeletal muscle ryanodine receptor. Journal of Biological Chemistry 278, 24831-24836.

Sakaba, T., & Neher, E. (2003). Direct modulation of synaptic vesicle priming by GABA(B) receptor activation at a glutamatergic synapse. Nature 424, 775-778.

Sands, W.A., Woolson, H.D., Milne, G.R., Rutherford, C., & Palmer, T.M. (2006). Exchange protein activated by cyclic AMP (Epac)-mediated induction of suppressor of cytokine signaling 3 (SOCS-3) in vascular endothelial cells. Molecular and Cellular Biology 26, 6333-6346

Sarkar, D., Erlichman, J., & Rubin, C.S., (1984). Identification of a calmodulin binding protein that Co-purifies with the regulatory subunit of brain protein kinase II. Journal of Biological Chemistry 259, 9840-9846.

Schacher, S., Castellucci, V.F., Kandel, E.R. (1988). cAMP evokes long-term facilitation in Aplysia sensory neurons that requires new protein synthesis. Science 240, 1667-1669.

Schwaerzel, M., Jaeckel, A., & Mueller, U. (2007). Signaling at A-kinase anchoring proteins organizes anesthesia-sensitive memory in Drosophila. Journal of Neuroscience 27, 1229-1233.

Scott, J.D., (1991). Cyclic nucleotide-dependent protein kinases. Pharmacology & Therapeutics 50, 123-145.

Shi, G.X., Rehmann, H., & Andres D.A. (2006). A novel cyclic AMPdependent Epac-Rit signaling pathway contributes to PACAP38-mediated neuronal differentiation. Molecular and Cellular Biology 26, 9136-9147.

Ster, J., De Bock, F., Guérineau, N.C., Janossy, A., Barrère-Lemaire, S., Bos, J.L., Bockaert, J., & Fagni, L. (2007). Exchange protein activated by cAMP (Epac) mediates cAMP activation of p38 MAPK and modulation of Ca2+-dependent K+ channels in cerebellar neurons. Proceedings of the National Academy of Sciences United States of America 104, 2519-2524.

Sutherland, E.W., & Rall, T.W. (1958). Fractionation and characterization of a cyclic adenine ribonucleotide formed by tissue particles. Journal of Biological Chemistry 232, 1077-1091

Sutherland, E.W. (1972). Studies on the mechanism of hormone action. Science 171, 401-408

36 Introduction and aim of this thesis

Taskén, K., Andersson, K.B., Skålhegg, B.S., Taskén, K.A., Hansson, V., Jahnsen, T., & Blomhoff, H.K. (1993). Reciprocal regulation of mRNA and protein for subunits of cAMP-dependent protein kinase (RI alpha and C alpha) by cAMP in a neoplastic B cell line (Reh). Journal of Biological Chemistry 268, 23483-23489.

Tavalin, S.J., Colledge, M., Hell, J.W., Langeberg, L.K., Huganir, R.L., & Scott, J.D., (2002). Regulation of GluR1 by the A-kinase anchoring protein 79 (AKAP79) signaling complex shares properties with long-term depression. The Journal of Neuroscience 22, 3044-3051.

Taylor, S.S., Buechler, J.A., & Yonemoto, W. (1990). cAMP-dependent protein kinase: Framework for a diverse family of regulatory enzymes. Annual Review of Biochemistry 59, 971-1005.

Taylor, J.R., Birnbaum, S., Ubriani, R., & Arnsten, A.F. (1999). Activation of cAMPdependent protein kinaseAin prefrontal cortex impairs working memory performance. Journal of Neuroscience 19, 1-5.

Ulucan, C., Wang, X., Baljinnyam, E., Bai, Y.Z., Okumura, S., Sato, M., Minamisawa, S., Hirotani, S., & Ishikawa, Y. (2007). Developmental changes in gene expression of Epac and its upregulation in myocardial hypertrophy. American Journal of Physiology 293, 1162-1672.

Vijayaraghavan, S., Liberty, G.A., Mohan, J., Winfrey, V.P., Olson, G.E., & Carr, D.W. (1999). Isolation and molecular characterization of AKAP110, a novel, sperm-specific protein kinase A- anchoring protein. Molecular Endocrinology 13, 705-717.

Waddell, S., Armstrong, J.D., Kitamoto, T., Kaiser, K. and Quinn, W.G. (2000). The amnesiac gene product is expressed in two neurons in the Drosophila brain that are critical for memory. Cell 103, 805-813

Wallenstein, G.V., Vago, D.R., & Walberer, A.M. (2002). Time-dependent involvement of PKA/PKC in contextual memory consolidation. Behavioral Brain Research 133, 159-164.

Wang, H., & Storm, D.R. (2003). Calmodulin-regulated adenylyl cyclases: cross-talk and plasticity in the central nervous system. Molecular Pharmacology 63, 463-468.

Westphal, R.S., Tavalin, S.J., Lin, J.W., Alto, N.M., Fraser, I.D., Langeberg, L.K., Sheng, M., & Scott, J.D. (1999). Regulation of NMDA receptors by an associated phosphatase-kinase signaling complex. Science 285, 93-96.

37 Chapter 1

Wong, W., & Scott, J.D. (2004). AKAP signalling complexes: focal points in space and time. Nature Reviews Molecular Cell Biology 5, 959-970.

Zanassi P., Paolillo M., Feliciello A., Avvedimento E.V., Gallo V., & Schinelli S. (2001). cAMP- dependent protein kinase induces cAMP-response element-binding protein phosphorylation via an intracellular calcium release/ERK-dependent pathway in striatal neurons. Journal of Biological Chemistry 276, 11487-11495.

Zhao, W.Q., Polya, G.M., Wang, B.H., Gibbs, M.E., Sedman, G.L., & Ng, K.T. (1995).Inhibitors of cAMP-dependent protein kinase impair long-term memory formation in day-old chicks. Neurobiology of Learning and Memory 64,106-118.

Zhong, N., & Zucker, R.S. (2005). cAMP acts on exchange protein activated by cAMP/cAMP- regulated guanine nucleotide exchange protein to regulate transmitter release at the crayfish neuromuscular junction. Journal of Neuroscience 25, 208-214.

Ziff, E.B., (1997). Enlightening the postsynaptic density. Neuron 19, 1163-1174.

38 Chapter 2

A-kinase anchoring protein 150 in the mouse brain is concentrated in areas involved in learning and memory

Anghelus Ostroveanu, Eddy A. Van der Zee, Amalia M. Dolga, Paul G. M. Luiten, Ulrich L.M. Eisel, Ingrid M. Nijholt

Department of Molecular Neurobiology, Graduate School of Behavioral and Cognitive Neurosciences, University of Groningen, The Netherlands

Brain Research, 2007, 1145(11):97-107 Chapter 2

Abstract

A-kinase anchoring proteins (AKAPs) form large macromolecular signaling complexes that specifically target cAMP-dependent protein kinase (PKA) to unique subcellular compartments and thus, provide high specificity to PKA signaling. For example, the AKAP79/150 family tethers PKA, PKC and PP2B to neuronal membranes and postsynaptic densities and plays an important role in synaptic function. Several studies suggested that AKAP79/150 anchored PKA contributes to mechanisms associated with synaptic plasticity and memory processes, but the precise role of AKAPs in these processes is still unknown. In this study we established the mouse brain distribution of AKAP150 using two well- characterized AKAP150 antibodies. Using Western blotting and immunohistochemistry we showed that AKAP150 is widely distributed throughout the mouse brain. The highest AKAP150 expression levels were observed in striatum, cerebral cortex and several other forebrain regions (e.g. olfactory tubercle), relatively high expression was found in hippocampus and olfactory bulb and low/no expression in cerebellum, hypothalamus, thalamus and brain stem. Although there were some minor differences in mouse AKAP150 brain distribution compared to the distribution in rat brain, our data suggested that rodents have a characteristic AKAP150 brain distribution pattern. In general we observed that AKAP150 is strongly expressed in mouse brain regions involved in learning and memory. These data support its suggested role in synaptic plasticity and memory processes.

Keywords: AKAP150; localization; immunohistochemistry; cAMP-dependent protein kinase

40 AKAP150 in the mouse brain

Introduction cAMP-dependent protein kinase (PKA) is involved in several intracellular signaling cascades and it regulates multiple cellular functions (Scott, 1991; Skalhegg & Tasken, 2000). A potential mechanism to explain how such a multifunctional and broad substrate kinase mediates precise signaling events, is colocalization of its substrate to specific subcellular compartments. Compartmentalization arises in part from the association of the enzyme with so-called A-kinase anchoring proteins (AKAPs) (Glantz et al., 1993; Lohmann, et al., 1984). AKAPs represent a group of more than 70 identified functionally related proteins (Wong & Scott, 2004). Although they share little primary structure similarities, they all have the ability to bind PKA, and therefore to regulate specific cAMP signaling pathways by sequestering PKA to a specific subcellular location. This compartmentalization of individual AKAP-PKA complexes occurs through unique targeting domains that are present on each anchoring protein. To date, AKAPs have been identified in a wide range of species, tissues and cellular compartments (Angelo & Rubin, 2000; Jackson & Berg, 2002, Sarkar et al., 1984; Wong & Scott, 2004). In the mammalian brain, several AKAPs have been characterized. One of these AKAPs is AKAP79/150. This family of proteins consists of three orthologues: bovine AKAP75, murine AKAP150 and human AKAP79. Initially AKAP75 was identified as a contaminant of PKA regulatory subunit II (RII) purified cytosolic brain preparations (Bregman et al., 1991; Sarkar et al., 1984). In addition, Bregman and colleagues retrieved AKAP150 by screening a rat cDNA library using radiolabeled RIIβ as functional probe (Bregman et al., 1989). Finally AKAP79 was identified as a constituent of postsynaptic densities (PSD) in human cerebral cortex (Carr et al., 1992). Interestingly, AKAPs function as multi-assembly scaffold molecules interacting with other signaling enzymes. AKAP79/150 has the ability to bind protein phosphatase 2B/calcineurin (PP2B/CaN) (Dell’Acqua et al., 2002) and protein kinase C (PKC) (Faux et al., 1999) besides PKA. By tethering both kinases and phosphatases AKAP79/150 provides a unique platform for integrating opposite signaling events to the same subcellular site. It has been suggested that in excitatory synapses at the PSD, AKAP79/150 targets its anchored proteins and forms a multi protein complex with AMPA and NMDA receptors

41 Chapter 2

(AMPAR and NMDAR), synaptic adhesion molecules, and cytoskeleton proteins. These proteins play an important role in synaptic function (Colledge et al., 2000; Kennedy, 1997; Malenka & Bear, 2004; Yamauchi, 2002; Ziff, 1997). The first evidence that anchoring of PKA is crucial for the regulation of synaptic function was reported by Rosenmund et al. (1994). In their study, blocking the PKA anchoring to AKAPs prevented the PKA-mediated regulation of AMPA/kainate currents in cultured hippocampal neurons. Moreover, recent findings strongly indicate that anchored PKA is crucial for maintaining AMPA currents during glutamate stimulation (Hoshi et al., 2005). Interestingly, disruption of AKAP-PKA anchoring leads to CaN-dependent, long-term depression (LTD)-like down-regulation of AMPAR currents, implicating an important role for AKAP79/150 in AMPAR regulation (Tavalin et al., 2002). In general, the AKAP79/150 scaffold molecule has emerged as an important element in regulating AMPAR phosphorylation in long-term potentiation (LTP) and LTD at the PSD (Dell’Acqua et al., 2006; Genin et al., 2003; Snyder et al., 2005). Since strengthening or weakening of synaptic transmission is widely considered to be the cellular mechanism that underlies learning and memory, a role of AKAP79/150 in learning and memory can be expected. To date, only a few immunohistochemical localization studies illustrate the distribution of AKAP79/150 in different brain compartments in various species. In human brain high levels of AKAP79/150 were reported at the PSD of the forebrain (Carr et al., 1992). A more detailed analysis of AKAP150 protein distribution in the rat brain showed that AKAP150 is widely distributed throughout the brain and is expressed in many classes of neurons that constitute the rat CNS (Glantz et al., 1992). A more recent study focused on the distribution of AKAP150 at rat CA1 pyramidal cell asymmetric and symmetric PSD and its colocalization with several markers of excitatory and inhibitory receptors. In this study, the interaction of AKAP150 with components of the excitatory PSD was confirmed, whereas AKAP150 immunoreactivity (IR) was not associated with inhibitory synapses (Lilly et al., 2005). The distribution of AKAP150 protein in the mouse brain has not been established yet. To elucidate the specific role of AKAP150 in learning and memory processes it may be important to use genetically modified mice in future research. Therefore, besides characterizing AKAP150 expression throughout the whole mouse brain, we specifically

42 AKAP150 in the mouse brain focused on AKAP150 expression in areas known to be involved in learning and memory processes. We established the distribution of AKAP150 protein in the mouse brain using immunohistochemistry and Western blot techniques.

Material and Methods

Animals and housing conditions The present study is based on brain tissue from 9-week-old male C57Bl/6J inbred mice (n=8) obtained from Harlan, Horst, The Netherlands. Animals were group-housed in standard macrolon cages placed in an environmentally controlled room for temperature (22 ± 1 °C) and humidity (45 ± 10%) under a 12:12h light/dark cycle with lights on at 7.00 a.m. Animals had free access to water and standard food pellets. Collection of brain tissue was performed after at least one week of accommodation under the above conditions. All procedures concerning animal care and treatment were in accordance with the regulation of the ethical committee for the use of experimental animals of the University of Groningen, The Netherlands (License: DEC 4278A).

Western blotting

43 Chapter 2

Protein samples (20 µg per sample) were separated on a 10 % SDS polyacrylamide gel and transferred to PVDF membranes (Millipore, USA). The blots were blocked for 1 h in blocking buffer (0.2 % I-Block (Tropix), 0.1 % Tween 20) and then incubated overnight at 4 °C either with goat anti-AKAP150 N-19 (1:500, sc-6446 Santa Cruz, CA, USA) or goat anti-AKAP150 C-20 (1:2,500, sc-6445 Santa Cruz, CA, USA). Mouse anti-actin antibody (1:40,000; MP Biomedicals, Irvine, CA, USA) served as control for protein loading. The blots were incubated with alkaline phosphatase-conjugated secondary antibodies [AP conjugated donkey anti-goat IgG (1:10,000)] (sc-2022 Santa Cruz, CA, USA) and AP- conjugated goat anti-mouse (1:10,000) (AC32ML, Tropix). Western blots were developed using the chemiluminescence method (Nitroblock and CDP-Star, Tropix). The immunoblots were digitized and quantified using a Leica DFC 320 image analysis system (Leica, Cambridge, UK).

Immunohistochemistry Animals were anesthetized by intraperitoneal injection with sodium pentobarbital 6% solution and sacrificed by transcardial perfusion with saline solution containing heparin, followed by 4 % paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB). After perfusion, the brains were kept at 4 °C in 0.01 M phosphate buffered saline (PBS) containing 0.1% sodium azide for 72 h. All brains were dehydrated for 48 h in a 30 % buffered sucrose solution. After dehydration, 20 µm coronal or sagittal sections were cut on a cryostat microtome. Sections were stored at 4 °C in 0.01 M PBS containing 0.1 % sodium azide. The avidin/biotin immunoperoxidase staining method was used to visualize mouse AKAP150 IR. Coronal and sagittal sections covering the whole mouse brain were rinsed in

0.01 M PBS. Sections were preincubated with 0.3 % H2O2 to reduce endogenous peroxidase activity and afterwards rinsed in PBS. Non-specific binding sites were blocked by preincubating the sections with 5 % normal rabbit serum in 0.01 M PBS for 30 min. Subsequently sections were incubated either with goat anti-AKAP150 N-19 (1:500) or goat anti-AKAP150 C-20 (1:500) in 0.01 M PBS containing 5 % normal rabbit serum and 0.3% Triton X-100 for 2 h at room temperature (RT) and left overnight at 4 °C. Afterwards the sections were rinsed (3 x 15 min in PBS) and incubated at RT for 2 h with biotin- conjugated rabbit anti-goat antibody (1:400) (Jackson Inc.) in 0.01 M PBS containing 1 %

44 AKAP150 in the mouse brain normal rabbit serum and 3 % Triton X-100. Another rinsing step with 0.01 M PBS at 4 °C was followed by incubation with avidin complex containing biotinylated horseradish peroxidase (1:400) (Vectastain ABC Kit, Burlingame, CA, USA) for 2 h at RT. Finally, staining was visualized with 0.03 % diaminobenzidine (DAB) chromogen substrate and

0.001 % H2O2. The incubation reactions were stopped by rinsing the sections with 0.01 M PBS.

Immunostained sections were analyzed with an Olympus BH2 microscope (Olympus, Japan). Two independent investigators established semi-quantification of mouse AKAP150 IR. For each brain region, material from 6 animals was examined to establish AKAP150 immunostaining. The scoring system used for establishing the relative AKAP IR was classified as follows: absent (-), low (+-), moderate (+), high (++) and very high (+++). In addition, for a subjective quantification of AKAP150 protein distribution in the mouse brain, relative densitometry measures for some major brain regions and nuclei have been determined (Table 2) (Quantimet 550 IW, Cambridge, UK). Photographs were taken with a DM1000/DFC280 Leica image analysis system (Leica, Cambridge, UK).

Antibody specificity Both AKAP150 antibodies are affinity purified goat polyclonal antibodies raised against a peptide mapping at the carboxyl terminus part (C-20) or N-terminus (N-19) of AKAP150 of rat origin. They showed, in parallel staining experiments, high specific AKAP150 IR in the mouse brain and displayed a similar pattern of staining. Because polyclonal antibodies were used, different affinities and avidities of the antibodies could influence the staining intensity. Therefore Table 1 reflects only relative amounts of AKAP150 IR, rather than a quantitative comparison between the two AKAP150 antibodies. The specificity of both antibodies in the mouse brain was assessed by parallel staining performed without primary antibodies and with secondary antibodies alone and showed no staining signals (data not shown). In addition, incubation of the conjugate with AKAP150 antibody/blocking peptide (1:10; Santa Cruz, CA, USA) was carried out at 4 °C overnight, followed by 25 min centrifugation (12,000 rpm), before application to the tissue sections. Following preabsorption, no AKAP150 staining was observed (Fig. 4B). Furthermore, preincubation

45 Chapter 2 steps were performed to increase the specificity of AKAP150 antibodies in the mouse brain.

Results

Western blot analysis of AKAP150 expression levels in various brain compartments Western blot analysis with two AKAP150 antibodies (a N-terminal or C-terminal antibody) in protein extracts from different mouse brain regions revealed the highest expression level of AKAP150 protein in cortex and striatum (Fig. 1). High expression levels were also detected in the hippocampus and olfactory bulb, while the cerebellum and the hypothalamus revealed low levels of AKAP150 expression. The lowest expression level of AKAP150 was found in the brain stem (Fig. 1).

A Fig. 1 Western blot analysis of AKAP150 expression levels in ) 140

(% mouse brain. (A) Bar graph representing AKAP150 expression 120

level level in different compartments of mouse brain. Values represent 100 mean percentages of integrated optical density (I.O.D). ± S.E.M. 80 Cortex was set to 100% (B) Representative Western blot with

60 expression 40 protein extracts from different brain regions. Actin served as 20 control for protein load.

AKAP150 0 B AKAP150 The results of the Western blot expression levels Actin corresponded with the distribution pattern of this

bulb stem protein in the immunohistochemical analysis of the Cortex Striatum Cerebellum Brain mouse brain (Fig. 2; Tables 1 and 2). HypothalamusOlfactoryHippocampus

General overview of AKAP150 IR in mouse brain AKAP150 IR was widely distributed throughout the brain. Highest IR was found in striatum and olfactory tubercle, but in cortex, hippocampus and amygdala AKAP150 expression was also very abundant (Fig. 2).

46 AKAP150 in the mouse brain

In several brain regions, AKAP150 IR was limited to specific cell layers (e.g. the Purkinje cell layer of the cerebellum) or nuclei (e.g. reticular thalamic nucleus, ethmoid nucleus) (Table 1). Some brain regions did not show any IR for AKAP150 protein (e.g. numerous nuclei in midbrain and hindbrain) (Fig. 2; Table 1).

A AP 1,98mm B AP 1,18mm C AP -0,70mm

CPu CPu LGP

LH Acb OT

D AP -1,46mm E AP -2,92mm

H SC

SNR BLA AH

Fig. 2. Localization of AKAP150 protein in coronal sections of the brain. Distance from bregma (AP) is indicated directly on the photographs (Franklin & Paxinos, 1997): (A) OT - olfactory tubercle, (B) CPu - caudate putamen, Acb - accumbens nucleus, (C) CPu - caudate nucleus, LGP - lateral globus pallidus, LH - lateral hypothalamic area, (D) H - hippocampus, VL - ventrolateral thalamic nucleus, BLA - basolateral amygdala, AH - anterior hypothalamic area, (E) SC - superior colliculus, SNR - substantia nigra reticular part. Scale bar = 1 mm.

Detailed description of AKAP150 expression in mouse brain

Olfactory system AKAP150 IR varied from moderate to relatively high in the olfactory bulb and its various layers. Superficial layers (glomerular layer and external plexiform layer) as well as deep layers (internal plexiform layer and granule layer) of the olfactory bulb showed moderate expression of AKAP150 (Table 1). Only the mitral cell layer of the olfactory bulb was highly immunoreactive for AKAP150 protein (Table 1). The olfactory tubercle showed a different staining pattern than the rest of the olfactory system. Here we observed a very

47 Chapter 2 high AKAP150 IR pattern. The excessive staining made it difficult to distinguish subcellular compartments within the olfactory tubercle (Figs. 2A and B).

Cerebral cortex In mouse cerebral cortex, AKAP150 protein was found to be expressed throughout all cortical layers (Figs. 2 and 3A). Very high AKAP150 IR levels were observed in cortical layers I (molecular layer), II (external granular layer) and IV (internal granular layer). Cortical layers II and IV were highly immunoreactive for AKAP150 protein in perikarya of granule cells (Figs. 3B and C).

A I B C B I-III

II-IV D C

E V

E F F VI

Fig. 3 Distribution of AKAP150 protein in the cerebral cortex. (A) Overview of AKAP150 protein distribution in cerebral cortex somatosensory 1, trunk region. Boxed areas represent the localization of panel B-F. (B) Dense staining is observed in cortical layers I-III. (C) Perikarya staining in cortical layer IV. (D) AKAP150 staining in barrel cortex. (E) Both fibers and cell bodies are immunopositive for AKAP 150 in cortical layer V. Note the weaker staining and less perikarya IR for AKAP150 in this layer as compared with cortical layer IV. (F) AKAP150 IR in cortical layer VI.

In layer IV of the somatosensory region of the mouse cerebral cortex, cortical barrels displayed a dense and characteristic AKAP150 IR pattern (Fig. 3D). Cortical layer III (external pyramidal layer) showed pronounced AKAP150 IR (Figs. 3A and B) whereas

48 AKAP150 in the mouse brain cortical layer V (internal pyramidal layer), which represents a principal output system of the neocortex, revealed only moderate immunostaining for AKAP150 (Fig. 3E). The deepest cortical layer, layer VI (polymorphic layer), was strongly AKAP150 IR. (Fig. 3F). The highest IR of AKAP150 protein in mouse cerebral cortex was found in cerebral cortex layer I (molecular layer) and in the stellate cells (interneurons) of both external and internal granular layers (cortex layers II and IV). In addition, the entorhinal cortex, the major source of afferents to the hippocampal formation, was intensely stained (Table 1).

Hippocampal formation Although there were some differences within this laminar structure, overall the hippocampal formation displayed high IR for AKAP150 protein (Fig. 4A; Tables 1 and 2).

A SO C Py SR SLM CA1 SLu D CA3 E GCL B hil ML

CD E ML

GCL Py Py

hil

Fig. 4 Immunostaining pattern for AKAP150 in the hippocampal formation. (A) AKAP150 IR in stratum oriens (SO), stratum radiatum (SR), stratum lacunosum moleculare (SLM), stratum pyramidale (Py) of CA1-3 and stratum lucidum (SLu) of CA3 and in molecular layer (ML), dentate granular cell layer (GCL) and hilus (hil) of the dentate gyrus. (B) AKAP150 IR after preabsorption of AKAP150 antiserum with the antigenic peptide. (C) Higher magnification of boxed area in panel A, showing immunopositive perikarya of pyramidal neurons (Py) in CA1, (D) Higher magnification of boxed area in panel A, showing stained pyramidal cell bodies (Py) in CA3. (E) Higher magnification of boxed area in panel A, showing immunopositive granule cell perikarya in the dentate gyrus (GCL).

49 Chapter 2

Especially in the CA1-3 region of the hippocampus, AKAP150 showed a very high expression in basal dendrites (stratum oriens) and apical dendrites (stratum radiatum) of pyramidal neurons (Fig. 4A). A somewhat lower IR was observed in both stratum lacunosum moleculare and stratum lucidum (Fig. 4A). A characteristic staining pattern was observed in the pyramidal cell body layer. In CA1 the pyramidal cell bodies were moderately stained (Figs. 4A and C) whereas in CA3 pyramidal neurons perikarya were rather densely stained (Figs. 4A and D). The subiculum also showed high AKAP150 IR (Table 1). In the dentate gyrus, AKAP150 IR was characterized by very high staining in the granule cell dendrites of the medial and lateral blade of the molecular layer (Fig. 4A). However, the dentate granule cells showed only moderate staining, whereas the hilus was slightly more immunoreactive (Figs. 4A and E). Interestingly, scattered non-principal cells in the dentate gyrus were also strongly stained for AKAP150 protein (Fig. 4E).

Amygdala In the amygdala, AKAP150 IR patterns revealed a high level of this protein in the various subcompartments. The distribution of AKAP 150 was clear and strong throughout all amygdaloid nuclei with only slight differences in the density of staining (Table 1). Uniform but strong staining was detected in cortical amygdaloid nuclei, medial amygdaloid nucleus, basomedial amygdaloid nucleus, amygdalohippocampal area and amygdalopiriform transition area (Table 1, Fig. 5A).

A B Fig. 5 Distribution of AKAP150 in the amygdala. (A) AKAP150 IR in the lateral amygdaloid nucleus, dorsolateral part LaDL (LaDL), lateral amygdaloid nucleus, CeMPV ventrolateral part (LaVL) and the anterior B CeL part of the basolateral amygdaloid nucleus (BLA). (B) Higher magnification LaVL CeC of boxed area in panel A, showing BLA AKAP150 IR in the central amygdala. The capsular part (CeC), lateral division (CeL) and the medial posteroventral part of the central amygdala (CeMPV) show staining in both cell bodies and fibers (see magnification in panel).

50 AKAP150 in the mouse brain

The most prominent staining in the amygdala for AKAP150 protein was detected in the central amygdaloid nucleus, lateral amygdaloid nucleus and basolateral amygdaloid nucleus (anterior, posterior and ventral) (Fig. 5A). The basolateral complex revealed a strong IR whereas the central amygdala was characterized by a dense staining in which both neuronal perikarya and fibers were detected (Figs. 5A and B; Table 1).

Septal nuclei and striatopallidal system In these brain regions, well-defined AKAP150 IR was observed. In the septum, the lateral nucleus was strongly stained, whereas the medial nucleus showed only faint AKAP150 staining (Fig. 2B; Tables 1 and 2). The caudate putamen was the most strongly stained structure in the entire mouse brain. The very dense AKAP150 IR pattern made it difficult to distinguish the subcellular distribution of AKAP150, although some perikarya were observed. Claustrum and ventral pallidum exhibited moderate to high AKAP150IR, while the globus pallidus showed almost no staining (Fig. 2; Table 1).

Epithalamus and thalamus In the epithalamus AKAP150 protein showed a moderate expression. Both the medial and lateral habenula displayed a diffuse IR with faint perikarya staining (Table 1). At the level of the thalamus distinct patterns of expression were detectable although the overall distribution of AKAP150 protein in this brain region was rather low (Fig. 2; Table 1). Detailed analysis of the stained thalamus revealed moderate staining in reticular thalamic nucleus, subgeniculate nucleus and ethmoid nucleus, where scattered fibers (varicose fibers) were specifically stained. Low levels of AKAP150, described as faint staining, were found in geniculate nucleus, anterior nucleus, lateral nucleus, ventral nucleus and zona incerta (Fig. 2). The remainder of the thalamus showed no AKAP150 IR (Table 1).

Hypothalamus In general, the expression level of AKAP150 in the hypothalamus was moderate to low. Accordingly, the supraoptic nucleus, paraventricular nucleus and suprachiasmatic nucleus

51 Chapter 2 showed a very faint and diffuse pattern of staining (Fig. 2; Table 1). Among the regions with the highest IR, the median eminence, arcuate nucleus and mammillary nuclei displayed a homogeneous staining (Table 1).

Midbrain AKAP150 protein showed a moderate to low expression in the midbrain region (Table 1). Most of the staining was observed in the superior colliculus, substantia nigra pars reticulata and pars compacta and periaqueductal gray (Fig. 2E). The IR in ventral tegmental area showed only moderate levels of AKAP150 (Fig. 2E). Low levels of AKAP150 IR were detectable in inferior colliculus and interpeducular nuclei (Table 1).

Cerebellum In the cerebellum, AKAP150 IR showed a moderate to low expression. Most of the staining was seen in the molecular layer of the cerebellar cortex in what appeared to be dendrites of the Purkinje neurons (Fig. 6B). The Purkinje perikarya were not well stained although punctuate staining of AKAP150 IR was detectable in the Purkinje layer throughout the whole cerebellar cortex (Fig. 6A). In the granule cell layer, the staining was considered faint, corresponding to a low expression of AKAP150 protein (Fig. 6A).

A B Fig. 6 AKAP150 expression in the cerebellum. (A) AKAP150 IR in the cerebellar molecular layer (Mol). (B) Higher magnification of boxed area in panel A, showing staining of Purkinje Gr Mol neurons (arrow, also in panel A) staining in both cell body and dendrites. B

Pons and medulla oblongata In the brainstem, the overall expression level of AKAP150 was low. Only the inferior olivary complex and nucleus of the solitary tract showed moderate immunostaining (Table 1). The remainder of this brain structure showed low or no AKAP150 IR.

52 AKAP150 in the mouse brain

Table 1 - Expression of AKAP150 protein in various Septum lateral nucleus ++ ++ compartments of the mouse central nervous system. Septum medial nucleus +- +- Brain region C- N- Diagonal band of Broca + + term term Striatopallidal system Telencephalon Caudate putamen +++ +++ Olfactory system Globus pallidus +- +- Glomerular layer + + Ventral pallidum +/++ ++ External plexiform layer + + Claustrum +/++ +/++ Mitral cell layer ++ ++ Islands of Calleja +++ +++ Internal plexiform layer + + Granule layer + + Diencephalon Anterior olfactory nucleus + + Thalamus Olfactory tubercle ++ ++/+++ Anterior nuclei - +- Cerebral cortex Lateral nuclei - +- Layer I +++ +++ Ventral nuclei +- +- Layer II ++ ++ Mediodorsal nucleus - - Layer III ++ ++ Central nuclei - - Layer IV +++ ++/+++ Paracentral nucleus - - Layer V + + Parafascicular thalamic - - Layer VI ++ ++ nuclei Piriform cortex ++ ++ Paraventricular thalamic - - Hippocampal formation nuclei Dentate gyrus Geniculate nuclei +- +- Granule cell layer + + Subgeniculate nucleus + + Molecular layer +++ +++ Reticular thalamic nucleus + ++ Hilus +++ ++ Posterior nuclear group - - CA1 region Paratenial nucleus - - Stratum radiatum +++ +++ Rhomboid nucleus - - Stratum oriens +++ +++ Ethmoid nucleus +/++ + Stratum lacunosum ++ ++ Reuniens +- - moleculare Subthalamic nucleus + -/+- Stratum pyramidale + + Zona incerta +- + CA3 region Epithalamus Stratum oriens +++ +++ Habenula medial + + Stratum radiatum +++ +++ Habenula lateral + + Stratum pyramidale +++ +++ Hypothalamus Stratum lucidum + ++ Median eminence +/++ + Subiculum ++ ++ Arcuate nucleus + +-/+ Entorhinal cortex ++ ++ Supraoptic nucleus +- +- Amygdala Paraventricular nucleus +- +- Central amygdaloid nucleus +++ ++ Periventricular nucleus +-/+ +- Medial amygdaloid nucleus ++ ++ Suprachiasmatic nucleus +- +-/- Lateral amygdaloid nucleus ++ ++/+++ Mammillary nuclei ++ + Basolateral amygdaloid +++ ++ Fornix - - nucleus Basomedial amygdaloid ++ ++ Mesencephalon nucleus Midbrain Cortical amygdaloid nuclei ++ ++ Superior colliculus + + Amygdalohippocampal ++ ++ Inferior colliculus +- +- area Interpeduncular nuclei +-/+ + Amygdalopiriform ++ ++ Substantia nigra transition area Pars reticulata + + Intercalated nuclei +++ +++ Pars compacta + + Septal and basal magnocellular nuclei Periaqueductal gray + + Bed nucleus stria terminalis ++ ++ Nucleus accumbens core ++ ++ Rhombencephalon Nucleus accumbens shell ++ ++ Cerebellum Substantia innominata + + Molecular layer + +

53 Chapter 2

Table 1 - Continuation Brain region C- N- term term Rhombencephalon Cerebellum Purkinje cell layer + + Granule cell layer +- +- Deep cerebellar nuclei +- +- Pons and medulla oblongata Locus coeruleus + + Reticular nuclei +- +- Ambiguus nucleus +- - Nucleus of the solitary tract +- +- Inferior olive + + Scoring was classified as absent (-), low (+-), moderate (+), high (++), very high (+++).

Table 2 - Mean optical density of AKAP150 protein Stratum oriens 0.39 0.018 in various compartments of the mouse central Stratum pyramidale 0.32 0.024 nervous system. Stratum lucidum 0.10 0.015 Brain region O.D. ±SEM Amygdala Telencephalon Central amygdala 0.33 0.011 Olfactory system Basolateral amygdala 0.29 0.004 Olfactory tubercle 0,52 0.006 Lateral amygdala 0.33 0.015 Cerebral cortex Septal and basal magnocellular nuclei Layer I 0.52 0.020 Septum lateral nucleus 0.32 0.012 Layer II 0.46 0.026 Septum medial nucleus 0.14 0.009 Layer III 0.41 0.027 Striatopallidal system Layer IV 0.41 0.022 Caudate putamen 0.46 0.015 Layer V 0.23 0.018 Layer VI 0.30 0.014 Diencephalon Barrel cortex 0.52 0.023 Thalamus Hippocampal formation Lateral nuclei 0.16 0.003 Dentate gyrus Hypothalamus Granule cell layer 0.16 0.019 Dorsomedial hypothalamus 0.09 0.011 Molecular layer 0.42 0.046 CA1 region Rhombencephalon Stratum radiatum 0.37 0.054 Cerebellum Stratum oriens 0.37 0.003 Molecular layer 0.17 0.002 Stratum lacunosum 0.21 0.035 Granule cell layer 0.13 0.005 moleculare Values represent mean of optical density and ± Stratum pyramidale 0.18 0.020 S.E.M. of each brain compartment (n=5). O.D.= CA3 region optical density.

Discussion

General overview of AKAP150 distribution in the mouse brain Our results showed that AKAP150 is widely distributed throughout the mouse brain. However, clear differences between brain compartments were observed. Both Western

54 AKAP150 in the mouse brain blotting and immunohistochemistry showed the highest expression levels of AKAP150 in striatum and cerebral cortex. In addition, relatively high AKAP150 expression was found in hippocampus and olfactory bulb and low/no expression in cerebellum, hypothalamus, thalamus and brain stem. Although the overall expression of AKAP150 in thalamus, hypothalamus, midbrain and hindbrain was limited, a few nuclei in these regions showed a moderate to high AKAP150 IR (e.g. reticular thalamic nucleus, ethmoid nucleus, mammillary nuclei). Although it remains difficult to speculate on the specific function of a brain region based solely on AKAP150 expression, a higher expression in these specific nuclei may suggest a more prominent role of AKAP150 in these nuclei. E.g. the thalamic reticular nucleus is a sheet of GABAergic neurons. It was recently shown that AKAP150 facilitates the phosphorylation of GABA(A) receptors by PKA which may have profound local effects on neuronal excitation (Brandon et al., 2003). It is interesting to note that in cholinergic areas such as the nucleus basalis, the medial septum and the diagonal band of Broca, which are known to be involved in the modulation of learning and memory, rather low levels of AKAP150 expression were observed. AKAP150 expression was absent in the large brain fiber systems (e.g. fornix and corpus callosum). With a few exceptions, our mouse AKAP150 data showed a distribution pattern similar to that previously described for the rat brain (Glantz et al., 1992). Like in the mouse brain, in the rat brain AKAP150 is abundantly expressed in olfactory bulb, cerebral cortex, hippocampus and cerebellar Purkinje neurons whereas in thalamus, hypothalamus, midbrain and hindbrain, AKAP150 is quantitatively characterized as modest or extremely low (Glantz et al., 1992 and Lilly et al., 2005).

AKAP150 IR in the cortex, hippocampus and amygdala In the mouse brain AKAP150 showed high levels of IR in the cortex. Dense staining was observed throughout all cortical layers, except for moderate staining in immunoreactive layer V. The same pattern of expression was observed in the rat neocortex (Glantz et al., 1992). Interestingly, in the mouse barrel cortex (cortical layer IV), AKAP150 showed a strong and characteristic staining which was not reported for the rat brain. The barrel cortex and its afferent pathway from the facial vibrissae in rodents is often used as a model for studying neuronal plasticity (Fox et al., 1996). Although the function of AKAP150 in the

55 Chapter 2 barrel cortex is not known, we can speculate on its involvement in mechanisms of neuronal plasticity in this brain region. AKAP150 was expressed in all mouse hippocampal subfields but mainly in apical and basal dendrites of pyramidal neurons and dendrites of granule cells in the dentate gyrus. In general these findings correspond to the rat hippocampal AKAP150 distribution as reported previously (Glantz et al., 1992; Lilly et al., 2005). However, we also observed several differences between mouse and rat AKAP150 expression in this region. Overall, CA1 and dentate gyrus showed stronger IR than CA3 in the rat hippocampus (Lilly et al., 2005), whereas the mouse hippocampus has a more homogeneous staining. On a more cellular level the pyramidal neurons of the mouse hippocampus showed a somewhat different staining in comparison to the rat AKAP150 IR. Mouse pyramidal neurons in the CA3 showed very high AKAP150 IR in their cell bodies, whereas cell bodies in the CA1 exhibited a much weaker staining. This specific staining for pyramidal neurons was not reported previously for the rat brain (Glantz et. al., 1992; Lilly et al., 2005). Interestingly, detailed observation of the mouse hippocampus indicated that interneurons might also contain AKAP150. For example, AKAP150-positive cells that resemble interneurons (based on shape, size and localization) were seen in the CA1-3 hippocampal area and in the dentate gyrus, which were not previously reported in the rat hippocampus (Glantz et al., 1992; Lilly et al., 2005). However, in the hippocampus from post-mortem human brains of healthy individuals and lobectomy samples from patients with intractable epilepsy, AKAP79 positive cells resembling interneurons were also reported in the CA1 area and in the hilus (Sik et al., 2000). Double staining experiments for AKAP79 and calretin and parvalbumin as interneuron markers demonstrated numerous double stained dendrites and somata at the electron microscopic level of the human hippocampus (Sik et al., 2000). Together these findings and our data suggest that AKAP79/150 may have a specific function in these interneurons. Nevertheless, the nature of the non-principal cells, which showed high expression levels of AKAP150 in the mouse brain, was not identified. Since we did not use any cellular markers for interneurons, the staining could also reflect e.g. excitatory mossy neurons. Substantial levels of AKAP150 were demonstrated both in mouse and rat amygdala (Glantz et al., 1992). The highest expression levels were found in central and basolateral amygdala

56 AKAP150 in the mouse brain where dense staining marked both perikarya and fibers. This pattern of expression was also reported previously for AKAP79 in human foetal amygdala where it was enriched in dendrites and various neuronal cell types (Ulfig & Setzer, 1999).

AKAP150 expression parallels the expression of PKA-RIIβ and AMPA receptor subunits The AKAP79/150 family has a high affinity for the RIIβ-PKA subunit (Bregman et. al., 1989). RIIβ is the predominant isoform and principal mediator of cAMP action in mammalian central nervous system (Sarkar et. al., 1984). Ventra and colleagues showed that in the rat neocortex and corpus striatum RIIβ mRNA levels paralleled the presence of AKAP150 protein. Conversely, in brain areas showing low RIIβ levels such as cerebellum, hypothalamus and cerebellum, the anchoring protein was absent (Ventra et al., 1996). In the adult mouse brain, high RIIβ mRNA levels were shown in the neocortex, caudate-putamen, hippocampus and reticular thalamic nuclei and a reduced level in the thalamus, midbrain and hindbrain (Cadd & McKnight, 1989). This RIIβ mRNA pattern in the mouse brain corresponds to our AKAP150 expression pattern confirming a parallel presence of both proteins in specific mouse brain regions. AKAP79/150 has been shown to regulate hippocampal AMPA receptor phosphorylation and function (Colledge et al., 2000; Tavalin et al., 2002). Interestingly, in the murine hippocampus high expression levels of both GluR1 and GluR2 AMPA receptor subunits were reported in CA1-3 stratum oriens, radiatum and lacunosum moleculare and in the stratum moleculare of the dentate gyrus (Yoneyama et al., 2004). This also parallels the AKAP150 expression pattern we observed in this brain structure.

Concluding remarks Overall, our results showed that AKAP150 protein is strongly expressed in numerous mouse brain compartments, and particularly in those areas that were reported to be involved in learning and memory processes. This supports the general notion that AKAP150 might be involved in learning and memory processes. AKAP79/150 bound PKA was found to play an important role in the regulation of AMPA receptor surface expression and synaptic plasticity (Rosenmund et al., 1994). Changes in synaptic plasticity are suggested to be the

57 Chapter 2 possible mechanism underlying learning and memory processes, therefore a role of AKAP79/150 in learning and memory can be expected. Initial evidence for a role of AKAP150 in learning and memory came from a study by Moita and colleagues, who reported that blocking PKA binding on AKAPs in the rat lateral amygdala leads to an impairment of memory consolidation of auditory fear conditioning (Moita et al., 2002). In addition, we could recently show that AKAP150 is upregulated in the mouse hippocampus after exposing mice to a novel context and after associative learning (Nijholt et al., 2007). In summary, we presented a detailed description of AKAP150 in discrete brain regions of the mouse. Although we observed minor differences between AKAP150 staining in the mouse and rat, our data suggested a characteristic pattern of AKAP150 distribution in these two rodent species.

Acknowledgments

We thank Jan Keyser for valuable microscopy technical assistance. Part of this work was supported by The Netherlands Organization for Scientific Research (NWO- Vernieuwingsimpuls E.A.V.d.Z (Grant 016.021.017)).

References

Brandon, N.J., Jovanovic, J.N., Colledge, M., Kittler, J.T., Brandon, J.M., Scott, J.D., & Moss, S.J. (2003). A-kinase anchoring protein 79/150 facilitates the phosphorylation of GABA(A) receptors by cAMP-dependent protein kinase via selective interaction with receptor beta subunits. Molecular and Cellular Neuroscience 22, 87-97.

Bregman, D.B., Bhattacharyya, N., & Rubin, C.S, (1989). High affinity binding protein for the regulatory subunit of cAMP-dependent protein kinase II-B. Journal of Biological Chemistry 264, 4648-4656.

58 AKAP150 in the mouse brain

Cadd, G., & McKnight, G.S. (1989). Distinct patterns of cAMP-dependent protein kinase gene expression in mouse brain. Neuron. 3, 71-9.

Colledge, M., Dean, R.A., Scott, G.K., Langeberg, L.K., Huganir, R.L., & Scott, J.D. (2000). Targeting of PKA to glutamate receptors through a MAGUK-AKAP complex. Neuron 27, 107-119.

Dell'Acqua, M.L., Dodge, K.L., Tavalin, S.J., & Scott, J.D. (2002). Mapping the protein phosphatase- 2B anchoring site on AKAP79. Binding and inhibition of phosphatase activity are mediated by residues 315-360. Journal of Biological Chemistry 277, 48796-802.

Dell'Acqua, M.L., Smith, K.E., Gorski, J.A., Horne, E.A., Gibson, E.S., & Gomez, L.L. (2006). Regulation of neuronal PKA signaling through AKAP targeting dynamics. European journal of cell biology 85, 627-33.

Faux, M.C., Rollins, E.N., Edwards, A.S., Langeberg, L.K., Newton, A.C., & Scott, J.D. (1999). Mechanism of A-kinase-anchoring protein 79 (AKAP79) and protein kinase C interaction . Biochemical Journal 343, 443-52.

Fox, K., Glazewski, S., Chen, C.M., Silva, A., & Li, X., (1996). Mechanisms underlying experience- dependent potentiation and depression of vibrissae responses in barrel cortex. Journal of Physiology- Paris 90, 263-269.

Franklin, K.B.J., & Paxinos, G., (1997). The mouse brain in stereotaxic coordinates. Academic Press, San Diego.

59 Chapter 2

Glantz, S.B., Amat, J.A., & Rubin, C.S., (1992). cAMP signaling in neurons: patterns of neuronal expression and intracellular localization for a novel protein, AKAP150, that anchors the regulatory subunit of cAMP-dependent protein kinase II beta. Molecular Biology of the Cell 3, 1215-1228.

Glantz, S.B., Li, Y., & Rubin, C.S., (1993). Characterization of distinct tethering and intracellular targeting domains in AKAP75, a protein that links cAMP-dependent protein kinase II beta to the cytoskeleton. Journal of Biological Chemistry 268, 12796-804.

Hoshi, N., Langeberg, L., & Scott, J., (2005). Distinct enzyme combinations in AKAP signaling complexes permit functional diversity. Nature Cell Biology 7, 1066–1073.

Kennedy, M.B., (1997). The postsynaptic density at glutamatergic synapses. Trends in Neurosciences 20, 264-268.

Lilly, S.M., Alvarez, F.J., & Tietz, E.I., (2005). Synaptic and subcellular localization of A-kinase anchoring protein 150 in rat hippocampal CA1 pyramidal cells: co-localization with excitatory synaptic markers. Neuroscience 1, 155-163.

Lohmann, S. M., De Camilli, P., Einig, I., & Walter, U., (1984). High-affinity binding of the regulatory subunit (RII) of cAMP dependent protein kinase to microtubule-associated and other cellular proteins. Proceedings of the National Academy of Sciences 81, 6723-6727.

Malenka, R.C., & Bear, M.F., (2004). LTP and LD: an embarrassment of riches. Neuron 44, 5-21.

Moita, M.A., Lamprecht, R., Nader, K., & LeDoux, J.E., (2002). A-kinase anchoring proteins in amygdala are involved in auditory fear memory. Nature Neuroscience 5, 837-838.

Nijholt, I.M., Ostroveanu, A., De Bruyn, M., Luiten, P.G.M., Eisel, U.L.M., & Van der Zee, E.A., (2007). Both exposure to a novel context and associative learning induce an upregulation of AKAP150 protein in mouse hippocampus. Neurobiology of Learning and Memory, 87, 693-696.

60 AKAP150 in the mouse brain

Scott, J.D., (1991). Cyclic nucleotide-dependent protein kinases. Pharmacology & Therapeutics 50, 123-145.

Sik, A., Gulacsi, A., Lai, Y., Doyle, W.K., Pacia, S., Mody, I., & Freund, T.F., (2000). Localization of the A kinase anchoring protein AKAP79 in the human hippocampus. European Journal of Neuroscience 12, 1155-64.

Skalhegg, B.S., & Tasken, K., (2000). Specificity in the cAMP/PKA signaling pathway. Differential expression, regulation, and subcellular localization of subunits of PKA. Frontiers in Bioscience 5, 678-93.

Snyder, E.M., Colledge M., Crozier R.A., Chen, W.S., Scott, J.D., & Bear, M.F., (2005). Role for A kinase-anchoring proteins (AKAPs) in glutamate receptor trafficking and long term synaptic depression. Journal of Biological Chemistry 280, 16962-16968.

Ulfig, N., & Setzer, M., (1999). Expression of a kinase anchoring protein 79 in the human fetal amygdala. Microscopy Research and Technique 46, 48-52.

Ventra, C., Porcellini, A., Feliciello, A., Gallo, A., Paolillo, M., Mele, E., Avvedimento, V.E., & Schettini, G., (1996). The differential response of PKA to cyclic AMP in discrete brain areas correlates with the abundance of regulatory subunit II. Journal of Neurochemistry 66, 1752-1761.

Wong, W., & Scott, J.D., (2004). AKAP signaling complexes: focal points in space and time. Nature Reviews Molecular Cell Biology 12, 959-970.

Yamauchi, T., (2002). Molecular constituents and phosphorylation-dependent regulation of the postsynaptic density. Mass Spectrometry Reviews 21, 266-286.

Yoneyama, M., Kitayama, T., Taniura, H. & Yoneda, Y., (2004). Immunohistochemical detection by immersion fixation with Carnoy solution of particular non-N-methyl-D-aspartate receptor subunits in

61 Chapter 2 murine hippocampus. Neurochemistry International 44, 413-422.

Ziff, E.B., (1997). Enlightening the postsynaptic density. Neuron 19, 1163-1174.

62 Chapter 3

Both exposure to a novel context and associative learning induce an upregulation of AKAP150 protein in mouse hippocampus

Ingrid M. Nijholt, Anghelus Ostroveanu, Ulrich L.M. Eisel, Eddy A. Van der Zee, Paul G.M. Luiten

Department of Molecular Neurobiology, University of Groningen, P.O. Box 14, 9750 AA Haren, The Netherlands

Neurobiology of Learning and Memory, 2007, 87(4):693-6 Chapter 3

Abstract

A-kinase anchoring protein 150 (AKAP150) is a multi-enzyme signaling complex that coordinates the action of PKA, PKC, and PP2B at neuronal membranes and synapses. We measured levels of AKAP150 protein in the hippocampus 6 h after training mice in a contextual fear conditioning paradigm. In contextual fear conditioning mice learn to associate a context with a footshock presentation. Mice were divided in four experimental groups with different training protocols: naive, no footshock exposure, immediate footshock exposure, and footshock 3 min after exposure to the context. We found that AKAP150 protein levels were increased upon exposing mice to the novel context independent of the training protocol. However, when the animals were habituated to the experimental context, only mice that learned to associate the context with the footshock showed an upregulation of AKAP150. We suggest that upregulated levels of AKAP150 contribute to processing the exposure to a novel context and associative learning.

Key words: A-kinase anchoring protein, Contextual fear conditioning, Memory

64 AKAP150 in learning

Brief Communication

The phosphorylation of intracellular proteins is a general mechanism used to control diverse cellular processes that occur in response to extracellular signals. Since many protein kinases and phosphatases are widely distributed throughout the cell and often exhibit a broad substrate specificity, additional mechanisms are used to contribute to the organization and specificity of signal transduction pathways by favoring the accessibility to certain substrates. The subcellular localization of cAMP-dependent protein kinase (PKA) is tightly controlled by a family of A-kinase anchoring proteins (AKAPs) (Rubin, 1994). AKAPs have been shown to interact with a number of signaling proteins, allowing for the localization and segregation of multi-enzyme signaling complexes. The capacity of AKAPs to coordinate multi-enzyme signaling complexes is very well exemplified by the neuronal AKAP79/150 family of anchoring proteins. This family consists of three structurally similar orthologs: bovine AKAP75, murine AKAP150, and human AKAP79 (Carr et al., 1992). AKAP79/150 targets PKA, protein kinase C (PKC), and protein phosphatase 2B (PP2B/calcineurin) to the same intracellular locus. At the postsynaptic membrane of glutamatergic synapses, this AKAP79/150 is recruited to NMDA and AMPA glutamate receptors by postsynaptic density (PSD)-95 family membrane-associated guanylate kinase (MAGUK) scaffold proteins (Colledge et al., 2000). It was found that this multi-enzyme signaling complex plays an important role in coordinating changes in synaptic structure and receptor signaling functions underlying synaptic plasticity (Dell'Acqua et al., 2006). AKAP150 mRNA was shown to be upregulated in the hippocampus 3-12 hr after the induction of LTP, a long-lasting enhancement in synaptic efficacy (Genin et al., 2003). In addition, pharmacological inhibition of PKA anchoring to AKAPs impaired late-phase LTP in hippocampal slices (Huang et al., 2006). Since synaptic plasticity is widely considered to be the cellular mechanism that underlies information storage in the brain, an important role of AKAP79/150 in learning and memory processes may be expected.

To date the only evidence for a role of AKAP79/AKAP150 in learning and memory processes came from a study by Moita, Lamprecht, Nader and Ledoux (2002). They

65 Chapter 3 reported that inhibition of PKA anchoring to AKAP150 in the rat lateral amygdala impairs memory consolidation of auditory fear conditioning (Moita et al., 2002). To investigate if learning is associated with changes in the expression of AKAP150 we assessed in the present study the expression of AKAP150 in the mouse hippocampus after a single training session in a contextual fear conditioning paradigm. Contextual fear conditioning is a hippocampus-dependent form of associative emotional learning. All experiments were performed with 9-12 weeks old male C57BL/6J mice (Harlan, Horst, The Netherlands). The procedures concerning animal care and treatment were in accordance with the regulations of the ethical committee for the use of experimental animals of the University of Groningen Upon arrival mice were individually housed in standard macrolon cages and maintained on a 12 h light/dark cycle (lights on at 7.30 a.m.) with food (hopefarm® standard rodent pellets) and water ad libitum. A layer of sawdust served as bedding. The animals were allowed to adapt to the housing conditions for 1-2 weeks before the experiments started.

Contextual fear conditioning was performed in a plexiglas cage (44 x 22 x 44 cm) with constant illumination. The training (conditioning) consisted of a single trial. The mouse was exposed to the conditioning context for 180 s followed by an electric footshock (ES; 0.7 mA, 2 s, constant current) delivered through a stainless steel grid floor. The mouse was removed from the fear conditioning box 30 s after ES termination to avoid an aversive association with the handling procedure (P-ES group, Fig 1A). The specificity of induction of fear was controlled by groups consisting of mice exposed to the context only (No-ES group) or to an immediate ES (ES-P group) (Fig. 1A). This latter group was introduced in order to discriminate between the fear response representing a conditioned response to the context, from an unconditioned response to the ES and possible sensitization induced by the ES. Memory tests were performed 24 h after fear conditioning. Contextual fear memory was tested in the fear conditioning box for 180 s without ES presentation. Freezing, defined as the lack of movement except for respiration and heart beat, was assessed by a time- sampling procedure every 10 s throughout memory tests. In addition, mean activity of the animal during the training and retention test was measured with the Ethovision system (Noldus, The Netherlands).

66 AKAP150 in learning

Fig. 1 Exposure to a novel context results in an increase in AKAP150 expression. (A) Schematic diagram of the behavioral protocols used for the three experimental groups. (B) Freezing behavior measured in the memory test performed 24 hr after training. Statistically significant differences: * p< 0.05 versus no-ES group. (C) Representative Western blot of four independent experiments. The bar graph summarizes the Western blot data and shows the levels of AKAP150 protein as a percentage of naive animals (set to 100%). Statistically significant differences: * p < 0.05 versus naive.

A separate set of animals was used to measure the expression levels of AKAP150 protein 6 h after the training session. Both hippocampi of the mice were collected 6 h after training. At this timepoint AKAP150 mRNA levels showed the largest increase after LTP induction (Genin et al., 2003) and this time point corresponds to the memory consolidation phase (Igaz et al., 2002). AKAP150 protein levels were measured using Western blotting. Hippocampi were homogenized at 4ºC with a plastic homogenizer, in a homogenization buffer containing 50 mM HEPES (pH 7,4), 150 mM NaCl, 0,2% NP-40, 4 mM EGTA, 10 mM EDTA, 15 mM sodium pyrophosphate, 100 mM β-glycerophosphate, 50 mM sodium fluoride, 5 mM sodium orthovanadate, 1 mM dithiothreitol, 1 mM PMSF, and Complete Mini Protease Inhibitor Cocktail (Roche). The insoluble material was removed by centrifugation at 20,000g for 10 min at 40C, and the resulting supernatant was assayed for

67 Chapter 3 protein concentration. 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, USA). The blot was probed using an AKAP150 antibody (1: 2.500; sc-6445 Santa Cruz, CA, USA) and an anti-actin antibody (1:40.000; MP biomedicals, Irvine, CA, USA). Western blots were developed using the chemiluminescence method. Immunoreactive bands were digitized and quantified using a Quantimet 500 image analysis system (Leica, Cambridge, UK). Statistical comparisons were made by analysis of variance (ANOVA). Data were expressed as mean ± SEM. Significance was determined at the level of p < 0.05.

There was no significant difference between the three experimental groups in activity during training (data not shown). During the retention test, mice of the no-ES group did not exhibit any fear-related behavior, as indicated by no freezing after re-exposure to the context (Fig. 1B). Mice that received an immediate footshock during training (ES-P) showed low freezing scores in the retention test. However, mice of the P-ES group had significantly higher freezing scores, than mice of the no-ES or ES-P groups upon re- exposure to the context (Fig. 1B). This finding was in full agreement with previous studies (Fanselow, 1980) and indicated that under entirely non-associative conditions, the shock itself did not produce a strong freezing response upon subsequent re-exposure of the mice to the chamber used in the training phase. The finding that mice exposed to context paired with shock exhibited significantly more freezing than mice exposed to immediate shock or to context only, demonstrated that the freezing behavior was induced by associative learning and did not represent an unconditional or non-associative response to the footshock employed. A separate set of mice was subjected to the training (No-ES, P-ES, ES-P, n=7-8 per group) trials described above and together with a group of naïve mice sacrificed 6 h after training. Their hippocampi were used for Western blotting to assess AKAP150 expression levels. Mice subjected to the three different training protocols all showed a strong increase of AKAP150 protein compared to naive animals (Fig. 1C). Thus, it appeared that upregulation of AKAP150 resulted from exposure of the animals to the novel environment. To prevent any interference of novel stimuli on AKAP150 expression, in the next set of experiments, the training was preceded by a habituation procedure to completely familiarize the mice

68 AKAP150 in learning with the stimuli of the experimental set-up. Habituation consisted of a 5 min exposure to the conditioning box for three trials per day on two consecutive days (Fig. 3A). Although animals showed a decreased in activity (Fig. 2) during habituation, animals never showed any freezing behavior during these sessions.

Fig. 2 Habituation causes a decrease in activity. The graph shows the time course of animal activity during the six habituation sessions.

Habituation did not significantly affect the freezing behavior in the retention test. Animals that did not receive a footshock during the training (H No-ES) or were immediately exposed to the footshock (H ES-P) showed no freezing and moderate freezing, respectively, whereas strong fear-related behavior was observed in the associative learning group (H P- ES)(Fig. 3B). When training was preceded by habituation, only associative learning was paralleled by an increase in AKAP150 expression in the hippocampus 6 h after training (Fig. 3C). By exposing the animals repeatedly to the context, the novelty effect wears off as indicated by AKAP150 proteins levels in the H No-ES group that were comparable to naive animals. Habituated animals that received a footshock immediately after exposure to the box showed no difference in AKAP150 expression compared to naive animals. Thus it can be concluded that shock exposure by itself does not lead to an upregulation of AKAP150. In summary, we conclude that both exposure to a novel context and associative learning upregulated AKAP150 expression in mouse hippocampus.

Recent research has shown that exposure to a novel event triggers a cascade of neural events that is also relevant to learning and memory. The hippocampus is an essential component of the network that detects and responds to novel stimuli (Knight, 1996). The

69 Chapter 3 upregulation of AKAP150 may be involved in this processing of novelty detection or may be closely related to arousal and anxiety induced by the experimental conditions.

Fig. 3 Associative learning is associated with an upregulation of AKAP150. (A) Experimental paradigm for the fear conditioning tests with the training and memory test sequences preceded by a six trial habituation procedure. (B) Freezing behavior measured in the memory test performed 24 h after training. Statistically significant differences: * p< 0.05 versus no-ES group. (C) Results are representative of five Western blot experiments. Statistically significant differences: * p < 0.05 versus naive.

In addition, we cannot rule out the possibility that the increased AKAP150 is, in fact, related to the formation of some type of associative new learning triggered by the novel environment. Habituation to the conditioning box resulted in a decrement of spatial exploration during successive exposures. Behavioral habituation to a novel environment is one of the most elementary forms of non-associative learning. In our experiments non- associative learning was not paralleled by an increase in AKAP150 expression in the hippocampus. Since AKAP150 is an important coordinator of cAMP pathways (Dell’acqua et al., 2006), it is interesting to note that non-associative learning does not involve cAMP signaling pathways. It is also not impaired by the inhibition of protein synthesis in the

70 AKAP150 in learning hippocampus (Vianna et al., 2000). In contrast, habituated animals showed an associative learning-specific increase in AKAP150 levels in the hippocampus 6 h after fear conditioning. Long-term formation of a hippocampal-dependent form of contextual fear conditioning was shown to depend on two consolidation periods during which hippocampal gene expression is critical: around the time of training and 3-6 h after training (Igaz et al., 2002). Whereas the first period of protein synthesis involves enhanced expression of immediately early genes and transcription factors, the second period concerns enhanced expression of structural proteins. This late phase of memory consolidation of fear motivated associative learning in rats is critically dependent on cAMP signaling pathways. It was found that PKA activity is enhanced in the hippocampal CA1 region (Bernabeu et al., 1997).In addition, extracellular regulated kinase/mitogen-activated protein (ERK/MAP) kinase is necessary for the consolidation of associative memories in the mammalian nervous system (Atkins et al., 1998). Coactivation of PKA and MAPK signaling leads to the concurrent activation of CREB-dependent gene expression required for hippocampal long-term memory formation (Impey et al., 1998). Our findings suggest a role for AKAP150 in the second consolidation period of long-term memory. Since AKAP150 levels were only measured 6 h after training we can not exclude that AKAP150 may also be important during other stages of the memory process. In general, the upregulation of AKAP150 may be the result of de novo protein synthesis or decreased protein degradation. Although we can only speculate on the possible role of the increased expression of AKAP150, it might very well be that elevated AKAP150 levels results in a more efficient propagation of signals carried by locally generated cyclic AMP (Colledge et al., 2000; Feliciello et al., 1997), which in turn may contribute to processing the exposure to a novel context and the consolidation of associative memory.

Acknowlegdements

Part of this work was supported by The Netherlands Organization for Scientific Research (NWO- Vernieuwingsimpuls E.A.V.d.Z (Grant 016.021.017)).

71 Chapter 3

References

Atkins, C. M., Selcher, J. C., Petraitis, J. J., Trzaskos, J. M., & Sweatt, J. D. (1998). The MAPK cascade is required for mammalian associative learning. Nature Neuroscience, 1, 602-609.

Bernabeu, R., Bevilaqua, L., Ardenghi, P., Bromberg, E., Schmitz, P., Bianchin, M., Izquierdo, I., & Medina, J.H. (1997). Involvement of hippocampal cAMP/cAMP-dependent protein kinase signaling pathways in a late memory consolidation phase of aversively motivated learning in rats. Proceedings of the National Academy of Sciences United States Of America, 94, 7041-7046.

Carr, D. W., Stofko-Hahn, R. E., Fraser, I. D., Cone, R. D., & Scott, J. D. (1992). Localization of the cAMP-dependent protein kinase to the postsynaptic densities by A-kinase anchoring proteins. Characterization of AKAP 79. Journal of Biological Chemistry, 267, 16816-16823.

Colledge, M., Dean R. A., Scott, G. K., Langeberg, L. K., Huganir, R. L., & Scott, J. D. (2000). Targeting of PKA to glutamate receptors through a MAGUK-AKAP complex. Neuron, 27, 107-119.

Dell'Acqua, M. L., Smith, K. E., Gorski, J. A., Horne, E. A., Gibson, E. S., & Gomez, L. (2006). Regulation of neuronal PKA signaling through AKAP targeting dynamics. European Journal of Cell Biology, 85, 627-633.

Fanselow, M.S. (1980). Conditional and unconditional components of post-shock freezing. Pavlovian Journal Biological Science, 15, 177–182.

Feliciello A., Li, Y., Avvedimento, E.V., Gottesman, M. E., &Rubin, C. S. (1997). A-kinase anchor protein 75 increases the rate and magnitude of cAMP signaling to the nucleus. Current Biology, 7, 1011-1014.

Genin, A., French, P., Doyere, V., Davis, S., Errington, M. L., Maroun, M., Stean, T., Truchet, B., Webber, M., Wills, T., Richter-Levin, G., Sanger, G., Hunt, S. P., Mallet, J., Laroche, S., Bliss, T.V., & O'Connor, V. (2003). LTP but not seizure is associated with up-regulation of AKAP-150. European Journal of Neuroscience, 17, 331-340.

Huang, T., McDonough, C. B., & Abel T. (2006). Compartmentalized PKA signaling events are required for synaptic tagging and capture during hippocampal late-phase long-term potentiation. European Journal of Cell Biology, 85, 635-642.

72 AKAP150 in learning

Igaz, L.M., Vianna, M.R., Medina, J.H., & Izquierdo, I. (2002). Two time periods of hippocampal mRNA synthesis are required for memory consolidation of fear-motivated learning. Journal of Neuroscience, 22, 6781-6789.

Impey, S., Smith, D. M., Obrietan, K., Donahue, R., Wade, C., & Storm, D. R. (1998).Stimulation of cAMP response element (CRE)-mediated transcription during contextual learning. Nature Neurosience, 1, 595-601.

Knight, R. (1996). Contribution of human hippocampal region to novelty detection. Nature, 383, 256 -259.

Moita, M. A., Lamprecht, R., Nader, K.,& LeDoux, J. E. (2002). A-kinase anchoring proteins in amygdala are involved in auditory fear memory. Nature Neuroscience, 5, 837-838.

Rubin, C. S. (1994). A-kinase anchor proteins and the intracellular targeting of signals carried by cyclic AMP. Biochimica et Biophysica Acta, 1224, 467-79.

Vianna, M.R., Alonso, M., Viola, H., Quevedo, J., de Paris, F., Furman, M., de Stein, M.L., Medina, J.H., & Izquierdo, I. (2000). Role of hippocampal signaling pathways in long-term memory formation of a nonassociative learning task in the rat. Learning and Memory, 7, 333-340.

Chapter 4

Inhibition of PKA anchoring to A-kinase anchoring proteins impairs consolidation and facilitates extinction of contextual fear memories

Anghelus Ostroveanu*,1, Ingrid M. Nijholt*,1, Wouter A. Scheper1, Botond Penke2, Paul G.M. Luiten1, Eddy A. Van der Zee1, Ulrich L.M. Eisel1

1Department of Molecular Neurobiology, University of Groningen, P.O.Box 14, 9750 AA Haren, The Netherlands, 2Department of Medical Chemistry, University of Szeged, 8 Dóm tér, Szeged, H-6720, Hungary * both authors contributed equally to this work

Neurobiology of Learning and Memory, 2008, 90(1):223-9 Chapter 4

Abstract:

Both genetic and pharmacological studies demonstrated that contextual fear conditioning is critically regulated by cyclic AMP-dependent protein kinase (PKA). Since PKA is a broad range protein kinase, a mechanism for confining its activity is required. It has been shown that intracellular spatial compartmentalization of PKA signaling is mediated by A-kinase anchoring proteins (AKAPs). Here, we investigated the role of PKA anchoring to AKAPs in different stages of the memory process (acquisition, consolidation, retrieval and extinction) using contextual fear conditioning, a hippocampus-dependent learning task. Mice were injected intracerebroventricularly or intrahippocampally with the membrane permeable PKA anchoring disrupting peptides St-Ht31 or St-superAKAP-IS at different time points during the memory process. Blocking PKA anchoring to AKAPs resulted in an impairment of fear memory consolidation. Moreover, disrupted PKA anchoring promoted contextual fear extinction in the mouse hippocampus. We conclude that the temporal and spatial compartmentalization of hippocampal PKA signaling pathways, as achieved by anchoring of PKA to AKAPs, is specifically instrumental in long-term contextual fear memory consolidation and extinction, but not in acquisition and retrieval.

Key words: A-kinase anchoring protein, Ht31, learning, memory, fear conditioning, mouse, hippocampus, extinction, superAKAP-IS

76 PKA anchoring in fear conditioning

Introduction

Contextual fear conditioning is a form of associative learning in which animals learn to fear a new environment because of its temporal association with an aversive unconditioned stimulus (US), usually an electrical footshock. The neuroanatomical systems and neurochemical basis underlying conditioned fear have been extensively investigated. It affects multimodal sensory information processing of continuously present (tonic) stimuli and it depends on a time-limited function of the hippocampus (see for review e.g. Sanders et al., 2003). Studies investigating the intracellular signal transduction pathways involved, have shown a crucial role for cAMP-dependent protein kinase (PKA) in contextual fear conditioning. Abel and colleagues generated transgenic mice which express R(AB), an inhibitory form of the regulatory subunit of PKA, only in forebrain regions such as the hippocampus. In these mice hippocampal PKA activity is reduced which is paralleled by behavioral deficits in long-term but not short-term memory for contextual fear conditioning (Abel et al., 1997). The time course of amnesia in these transgenic mice is similar to the time course observed in mice treated with inhibitors of PKA (Bourtchouladze et al., 1998). Other studies using pharmacological approaches also reported that PKA inhibitors impair contextual fear conditioning (Ahi et al., 2004; Schafe et al., 1999; Wallenstein et al., 2002). Although much is known about the mechanisms involved in the storage of contextual fear memories, the processes underlying the extinction of fear memories are far less understood. Recently, a role for PKA in fear extinction was proposed. Transgenic mice which express R(AB), show facilitated extinction of both recent and remote contextual fear memories (Isiegas et al., 2006) whereas increased PKA activity was found to impair extinction (McNally et al., 2005; Wang et al., 2004). In general these studies suggest that the PKA signal transduction pathway is important in the consolidation and extinction of contextual fear memories. However, PKA is a multifunctional enzyme with a broad substrate specificity and thus coordinated control of PKA signaling is required. This is partly achieved by association of the enzyme with so called A-kinase anchoring proteins (AKAPs) (Rubin, 1994). AKAPs are a group of more than 50 identified functionally related proteins. Although they share

77 Chapter 4 little primary structure similarities, they all have the ability to bind the regulatory subunits of PKA, and therefore to coordinate specific cAMP signaling pathways by sequestering PKA to a particular subcellular location (Beene & Scott, 2007; Wong & Scott, 2004). Up to 75% of the total cellular PKA is believed to be associated with some member of the AKAP family. Compartmentalization of individual AKAP-PKA complexes occurs through specialized targeting domains that are present on each anchoring protein. Interestingly, several AKAPs bind more than one signaling enzyme simultaneously. These multivalent AKAPs serve as scaffolds for the assembly of signaling complexes consisting of several kinases and phosphatases. Compartmentalization of both kinases and phosphatases to the same location may provide a coordinated activity of two enzymes with opposite catalytic activities. Previous studies mainly focused on the effect of changes in PKA activity on learning and memory processes. However, recent findings suggest that positioning of PKA at its proper subcellular location by AKAPs is crucial for its efficient catalytic activation and accurate substrate selection and may thus be important in learning and memory processes. Hitherto knowledge on the importance of PKA anchoring to AKAPs in learning and memory processes is limited. In an initial study Moita and colleagues showed that local inhibition of PKA anchoring in the rat lateral amygdala impaired memory consolidation of auditory fear conditioning (Moita et al., 2002). More recent studies in Drosophila reported an important role for AKAPs in olfactory memory processing (Lu et al., 2007; Schwaerzel et al., 2007). Furthermore, data from genetically modified mice that conditionally express Ht31, an inhibitor of PKA anchoring to AKAPs, showed that an anchored pool of PKA is important in theta-burst LTP and hippocampus-dependent spatial memory storage (Nie et al., 2007). In aplysia sensory neurons Ht31 was found to prevent both short- and long-term facilitation (Liu et al., 2004).

In the present study we investigated the importance of PKA anchoring in the distinct stages of the memory process during contextual fear conditioning.

78 PKA anchoring in fear conditioning

Materials and methods

Animals All experiments were performed with 9-12 weeks old male C57BL/6J mice (Harlan, Horst, the Netherlands). Individually housed mice were maintained on a 12 h light/dark cycle (lights on at 7.00 a.m.) with food (Hopefarm® standard rodent pellets) and water ad libitum. A layer of sawdust served as bedding. The animals were allowed to adapt to the housing conditions for 1-2 weeks before the experiments started. The procedures concerning animal care and treatment were in accordance with the regulations of the Ethical Committee for the use of experimental animals of the University of Groningen (DEC 4174C).

Fear conditioning Fear conditioning was performed in a plexiglas cage (44 x 22 x 44 cm) with constant illumination (12 V, 10 W halogen lamp, 100-500 lux). The training (conditioning) consisted of a single trial. Before each individual mouse entered the box, the box was cleaned with 70% ethanol. The mouse was exposed to the conditioning context for 180 s followed by a footshock (0.7 mA, 2 s, constant current) delivered through a stainless steel grid floor. The mouse was removed from the fear conditioning box 30 s after shock termination to avoid an aversive association with the handling procedure. Memory tests were performed 1 or 24 h after fear conditioning. Contextual memory was tested in the fear conditioning box for 180 s without footshock presentation. 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 by a time-sampling procedure every 10 s throughout memory tests. In addition, mean activity of the animal during the training and retention test was measured with the Ethovision system (Noldus, The Netherlands). In some experiments, animals were exposed to an alternative context 24 h after the training session. This alternative context consisted of a white plastic chamber (39×29×19 cm) which was exposed to 500-1000 lux, did not have a rod floor and was washed with 1 % acetic acid, before each individual mouse entered the chamber. To assess fear extinction mice underwent a daily re-exposure to the

79 Chapter 4 conditioning chamber for 3 min after the retention test. During these extinction trials freezing behavior and mean activity was measured.

Animal surgery Double guide cannulae (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 as previously described (Nijholt et al., 2004) into both lateral brain ventricles (i.c.v.) with anteroposterior (AP) coordinates zeroed at Bregma AP 0 mm, lateral 1 mm, depth 3 mm or directed toward both dorsal hippocampi (i.h.), AP –1.5 mm, lateral 1 mm, depth 2 mm (Franklin & Paxinos, 1997). Each double guide cannula with inserted dummy cannula and dust cap was fixed to the skull with dental cement (3M AG, Germany). Administration of 1 mg/ml finadyne (0.005 ml/g i.p.) before the surgery served as pain killer. The animals were allowed to recover for 6-7 days before the behavioral experiments started.

Brain injections Bilateral injections were performed during a short isoflurane anesthesia using a Hamilton microsyringe fitted to a syringe pump unit (TSE systems, Bad Homburg, Germany) at a constant rate of 0.5 µl/min (final volume: 1 µl per side) for the i.c.v. injections and 0.34 µl/min (final volume: 0.3 µl per side) for the i.h. injections. PKA anchoring to AKAPs was inhibited by intracerebroventricular (i.c.v.) or intrahippocampal (i.h.) injection of the peptide Ht31 (InCELLect® AKAP St-Ht31 inhibitor peptide (Promega, Madison, WI)) or superAKAP-IS. These peptides inhibit the interaction between the regulatory subunits of PKA and AKAP (Gold et al., 2006; Vijayaraghavan et al. 1997). SuperAKAP-IS was synthesized by solid phase peptide synthesis using BOC-chemistry and purified after cleavage from the matrix by preparative HPLC. Purity was controlled by analytical HPLC and mass spectrometry. The stearated form of Ht31 and superAKAP-IS was used to enhance the cellular uptake of the peptide through the membrane. St-Ht31 was injected in a final concentration of 10 mM (i.c.v. 20 nmol/mouse and i.h. 6 nmol/mouse) and St-superAKAP-IS in a final concentration of 5-500 µM (i.h. 0.003-0.3 nmol/mouse per injection). Unfortunately, it was not possible to prepare concentrations of St-superAKAP-IS higher than 500 µM. 50 mM Tris-HCl (pH 7.5) served

80 PKA anchoring in fear conditioning as vehicle. To test the specificity of the observed effects another set of animals was injected with either InCELLect® St-Ht31P, a proline-substituted derivative which does not inhibit PKA anchoring (control peptide; final concentration 10 mM in 50 mM Tris-HCl pH 7.5; i.c.v. 20 nmol/mouse and i.h. 6 nmol/mouse) or vehicle alone (50mM Tris-HCl pH 7.5). Untreated animals without cannula served as controls for possible cannulation and injection effects. The number of animals per group varied from 6 to 18.

Histology Immediately after the behavioral test mice were injected during 1.2 % avertin anesthesia (0.02 ml/g, i.p.) with methylene blue solution i.c.v., or i.h. Brains were removed and serially sectioned at 50 µm, collecting the sections on glass slides. Sections were stained on glass for 5 minutes in 0.1% nuclear fast red solution. To identify the location of the injection, sections were analyzed using light microscopy (Fig. 1).

Fig. 1 Representative coronal brain sections of bilateral (A) intracerebroventricular (i.c.v.) and (B) dorsal hippocampal (i.h.) injections with methylene blue injections after counterstaining with nuclear fast red.

Only data from animals in which the exact site of injection was confirmed after the behavioral experiments were evaluated. The methylene blue injections in the dorsal hippocampus did not show a diffusion of the solution to other brain or hippocampal areas.

Immunoprecipitation One hour after intrahippocampal injection of PKA-anchoring disruptor peptide or vehicle solution, the dorsal hippocampus was excised and mechanically homogenized in 10 volumes of homogenization buffer [50 mM Hepes (pH 7.4), 150 mM NaCl, 0.2 % NP-40, 4 mM EGTA, 10 mM EDTA, 15 mM sodium pyrophosphate, 100 mM β-glycerophosphate, 50 mM sodium fluoride, 5 mM sodium orthovanadate, 1 mM dithiothreitol, 1 mM PMSF,

81 Chapter 4 and Complete Mini Protease Inhibitor Cocktail (Roche)]. The homogenate was centrifuged at 20,000g for 10 min at 4 °C, and the resulting supernatant was used for AKAP150 immunoprecipitation. Per sample 100 µl of Dynabeads protein A (Dynal Biotech) was washed twice with Na- phosphate buffer (0.1 M, pH 8.1). Ten micrograms of goat anti-AKAP150 C-20 antibody (1:2500, sc-6445 Santa Cruz, CA, USA) was incubated with the beads for 10 min. Afterwards, the beads were washed three times with Na-phosphate buffer (0.1 M, pH 8.1) and twice with triethanolamine (0.2 M). IgGs were crosslinked with dimethyl pimelimidate (20 mM in 0.2 M trietholamine) for 30 min. The beads were washed for 15 min with Tris (50 mM, pH 7.5) and three times with phosphate buffered saline. Unbound IgG was removed by washing twice for 30 min with Na-citrate (0.1 M, pH 2-3). The dorsal hippocampus homogenate was incubated for 1 h with the beads. Bound proteins were eluted by denaturation at 95 °C for 5 minutes. The immunoprecipitated sample was stored at -80 ºC until use. All the steps of the immunoprecipitation procedure were performed at room temperature.

Western blotting AKAP150 immunoprecipitates were separated on a 10% SDS-polyacrylamide gel and transferred to PVDF membranes (Millipore, USA). The blots were blocked for 1 h in blocking buffer (0.2 % I-Block (Tropix), 0.1% Tween 20) and then incubated overnight at 4 °C with goat anti-AKAP150 C-20 (1:2500, sc-6445, Santa Cruz) and mouse anti-PKA RIIβ (1:2.000, 610625, BD Biosciences). The blots were incubated with horse radish peroxidase- conjugated secondary antibodies [HRP conjugated donkey anti-goat IgG (1:4.000)] (sc- 2020 Santa Cruz, CA, USA) and HRP-conjugated donkey anti-mouse (1:4.000) (sc-2005 Santa Cruz, CA, USA). Western blots were developed using the chemiluminescence method (Pierce ECL, 32106). The immunoblots were digitized and quantified using a Leica DFC 320 image analysis system (Leica, Cambridge, UK).

Statistical analysis Statistical comparisons were made by analysis of variance (ANOVA). For each significant F ratio, Fisher's protected least significant difference (PLSD) test was used to analyze the

82 PKA anchoring in fear conditioning statistical significance of appropriate multiple comparisons. Data were expressed as mean ± s.e.m. Significance was determined at the level of p < 0.05.

Results

Consolidation of contextual fear memory is impaired by i.c.v St-Ht31 injection To investigate the effect of inhibition of PKA anchoring to AKAPs on the acquisition and consolidation of fear memory, animals were injected i.c.v. with St-Ht31, control peptide or vehicle 1 h before training. Injection of none of these substances resulted in changes in mean activity during training or shock reactivity when compared to untreated animals without cannula (data not shown). However, injection of St-Ht31 caused a significant reduction in freezing behavior during the retention test 24 h after training in comparison to control peptide, vehicle injected and untreated animals (one-way ANOVA: F(3, 31) = 5.471, p = 0.004, Fig. 2A). Similarly, injection of St-Ht31 immediately after training significantly attenuated conditioned fear (one-way ANOVA: F(3,30) = 3.932, p = 0.018, Fig. 2B). The learning deficit observed when St-Ht31 was injected immediately after training was similar to the effect of St-Ht31 injected 1 h before training (43.8 ± 8.1%, n=9 versus 40.0 ± 7.3%, n=7 respectively). To be able to distinguish between acquisition and consolidation, we performed a retention test 1 h after training with mice that were injected 1 h before training. Overall, the contextual fear response was somewhat lower 1 h after training than 24 h after training (Fig. 2A versus Fig. 2C). This result is in full agreement with previous studies of Rudy and Morledge who investigated the time course of the expression of context- dependent fear (Rudy & Morledge, 1994). Interestingly, the performance of St-Ht31 injected animals did not differ from the control groups when the retention test was performed 1 h after training (one-way ANOVA: F(3,20) = 0.257, p = 0.855, Fig. 2C). The finding that mice which received St-Ht31 1 h before training, showed unimpaired freezing 1h after training but attenuated freezing 24 h after training, suggests that PKA anchoring onto AKAPs plays a specific role in the consolidation of contextual fear memories but not in acquisition.

83 Chapter 4

Injection Injection A B Training 2” 24hr retention Training 2” 24hr retention

180” 30” 180” 180” 30” 180”

80 80

60 60 * *

(%)

(%) 40 40

Freezing Freezing 20 20

0 0

ehicle ehicle St-Ht31 V St-Ht31 V Untreated St-Ht31-P Untreated St-Ht31-P

Injection D Injection C Training 2” 1hr retention Training 2” 24hr retention

180” 30” 180” 180” 30” 180”

80 80

60 60

(%)

(%) 40 40

20 20

Freezing

0 0

ehicle ehicle V St-Ht31 V St-Ht31 Untreated St-Ht31-P Untreated St-Ht31-P

Fig. 2 Intracerebroventricular injection of St-Ht31 impairs the consolidation of contextual fear memory. Mice were injected either one hour before training (A and C), immediately after training (B) or one hour before the retention test (D) with St-Ht31, control peptide or vehicle. Untreated mice served as controls. The training consisted of a 180 s exposure to the fear conditioning box followed by a footshock ( , 0.7mA, 2 sec). 30 s after the footshock mice were returned to their home cage. Freezing behavior was measured in the memory test 1 h (C) or 24 h (A, B and D) after training. Error bars indicate standard error of the mean. Statistically significant differences: * p< 0.05 versus all control groups (vehicle, control peptide and untreated).

The importance of PKA anchoring in the retrieval of memories was studied by injecting mice with St-Ht31 1 h before the retention test 24 h after training. There was no significant difference in freezing behavior between all groups (one-way ANOVA: F(3,25) = 0.071, p = 0.975, Fig. 2D).

84 PKA anchoring in fear conditioning

Intrahippocampal injection of PKA anchoring disrupting peptides impairs consolidation of contextual fear memory We tested the subregion-specific contribution of the hippocampus by i.h. injection of St- Ht31, different concentrations of St-superAKAP-IS, control peptide or vehicle. When injected immediately after training, both St-Ht31 and St-superAKAP-IS caused an impairment of contextual fear memory when compared to the control groups (one-way ANOVA: F(7,69) = 4.219, p = 0.001, Fig. 3A). The effect of St-superAKAP-IS on freezing behavior appeared to be dose-dependent (Fig. 3A).

A Injection B Injection Training 2” 24hr retention Training 2” 24hr retention

180” 30” 180” 180” 30” 180”

80 80

60 * * * 60

(%) (%) 40 40

Freezing 20 Freezing 20

0 0

ehicle ehicle St-Ht31 V V Untreated St-Ht31-P Untreated

St-superAKAP-ISSt-superAKAP-ISSt-superAKAP-ISSt-superAKAP-IS St-superAKAP-IS μM μM μM μM 5 100 300 500

Fig. 3 Hippocampal PKA anchoring plays an important role in the consolidation of contextual fear memory. Mice were injected intrahippocampally with St-Ht31, St-superAKAP-IS, control peptide or vehicle immediately after training. Untreated mice served as controls. Freezing was measured in the memory test in the same context (A) or in an alternative context (B) 24 h after training. Error bars indicate standard error of the mean. Statistically significant differences: *p < 0.05 versus all control groups.

In addition, consistent with other studies (Radulovic et al., 1998), mice showed contextual generalization of fear in an alternative context 24 h after the training session. However, freezing in this alternative context was much lower than in the conditioning context and was not affected by 500 µM St-superAKAP-IS injection (one-way ANOVA: F(2,15) = 1.154, p = 0.342, Fig. 3B), indicating that the non-associative component of the freezing response is not dependent on PKA anchoring.

85 Chapter 4

Overall we can conclude that PKA anchoring to AKAPs located in the hippocampus is instrumental in associative memory consolidation. However, we cannot completely rule out the additional involvement of extrahippocampal PKA signaling pathways. In all experiments, the injection procedure itself had no effect on conditioned fear as indicated by the finding that there was never a significant difference between vehicle- injected and non-injected animals (Fig. 2 and 3).

Intrahippocampal injection of St-superAKAP-IS promotes fear extinction Next we assessed the role of PKA anchoring in the extinction of contextual fear memory. Mice underwent a single training trial and retention test and after the retention test mice were daily re-exposed to the conditioning chamber for 3 min. St-superAKAP-IS (500 µM) or vehicle was injected i.h. immediately after each extinction trial. Inhibition of PKA anchoring by St-superAKAP-IS significantly facilitated fear extinction (Extinction 5, one- way ANOVA: F(1,10) = 7.836, p = 0.019; Extinction 6, one-way ANOVA: F(1,10) = 8.188, p = 0.017; Extinction 7, one-way ANOVA: F(1,10) = 10.152, p = 0.010, Fig. 4).

Fig. 4 Intrahippocampal injection of St-superAKAP-IS 80 * * facilitates the extinction of contextual fear memory. Mice * 60 were injected intrahippocampally with St-superAKAP-IS

(%) and vehicle immediately after each extinction. Freezing was 40 measured in the memory test performed 24 h after training and on 8 consecutive days, starting 24 after the memory Freezing 20 St-superAKAP-IS Vehicle test. Error bars indicate standard error of the mean.

1 2 3 4 5 6 7 8 Statistically significant differences: *p < 0.05 versus all Extinction control groups. Retention

Intrahippocampal St-superAKAP-IS injection reduced PKA anchoring to AKAP150 Using immunoprecipitation we specifically assessed the amount of PKA anchored to AKAP150 in the dorsal hippocampus one hour after intrahippocampal injection of vehicle or 500 µM St-superAKAP-IS. The AKAP150 complex was immunoprecipitated with an antibody directed against AKAP150. Subsequent analysis of the amount of PKA bound to

86 PKA anchoring in fear conditioning

AKAP150 showed that St-superAKAP-IS reduced the amount of PKA anchored to AKAP150 in the dorsal hippocampus (one-way ANOVA: F(1, 7) = 12.115, p = 0.01, Fig. 5)

A 120 (%) 100 Fig. 5 Intrahippocampal injection of St-superAKAP-IS impairs * 80 PKA anchoring to AKAP150. Dorsal hippocampus was excised 60

/AKAP150 1 h after St-superAKAP-IS or vehicle injection. AKAP150 was β 40 immunoprecipitated from the dorsal hippocampus. (A) Bar 20 graph showing the ratio of PKA-RIIβ complexed to AKAP 150. PKA-RII 0 The ratio in the vehicle-injected group was set at 100% for each B AKAP150 150kD experiment. Results shown represent three separate experiments. Error bars indicate standard error of the mean. 57kD PKA-RIIβ Statistically significant differences: *p < 0.05 versus the vehicle group. (B) Representative Western blot for AKAP150 and

ehicle PKA-RIIβ. V

St-superAKAP-IS

Discussion

In summary, we conclude that hippocampal PKA anchoring to AKAPs is important for the consolidation and extinction of contextual fear memories whereas acquisition and retrieval are not affected. These findings are consistent with earlier studies using genetic and pharmacological approaches to inhibit PKA activity. The genetic reduction of hippocampal PKA activity in mice that express PKA-R(AB) selectively impairs hippocampus-dependent long-term memory for contextual fear conditioning (Abel et al., 1997). To exclude the developmental effects as a result of transgene expression Abel and colleagues confirmed their data via injection of a PKA inhibitor (Bourtchouladze et al., 1998). Both i.c.v. and i.h. injections of PKA or PKA/PKC inhibitors before or after training did not effect memory after 1 h but

87 Chapter 4 significantly impaired memory after 24 h (Bourtchouladze et al., 1998; Schafe et al., 1999; Wallenstein et al., 2002). Overall, these data suggest an important role for PKA signaling in the long-term consolidation of contextual fear memories. Besides PKA, extracellular regulated kinase/mitogen-activated protein (ERK/MAP) kinase is necessary for the consolidation of associative memories in the mammalian nervous system (Atkins et al., 1998). It is suggested that coactivation of PKA and MAPK signaling leads to the concurrent activation of CREB-dependent gene expression required for hippocampal long-term memory formation (Impey et al., 1998). From our data it can be concluded that not only PKA activity is necessary for proper consolidation of memories but also the spatial and temporal compartmentalization of PKA achieved via anchoring to AKAPs.

Mammalian PKA includes four regulatory (RIα, RIβ, RIIα, RIIβ) and three catalytic (Cα, Cβ, Cγ) subunits, each encoded by a separate gene. PKA consists of an inactive heterotetramer of two catalytic subunits bound to two regulatory subunits (Taylor et al., 1990). PKA is associated to AKAPs with its regulatory subunits via an amphipathic helix binding motif (Herberg et al., 2000). In studies by Fink and colleagues inhibition of PKA anchoring by Ht31 resulted in redistribution of the regulatory subunits and decreased compartmentalization of PKA (Fink et al., 2001). Thus, disrupted spatial compartmentalization of PKA attenuates the specificity of the cAMP/PKA signaling pathway. This will affect downstream proteins such as the phosphorylation of CREB and may finally lead to impaired long-term memory consolidation. Our finding that only long- term memory consolidation is affected and not acquisition or retrieval indicates that there is a critical time window in which PKA anchoring is essential in contextual fear memories.

The specific ways in which inhibition of PKA anchoring accelerates extinction remains to be determined. However our findings are in line with the facilitated extinction of contextual fear memories observed in mice with a transgenic inhibition of PKA (Isiegas et al., 2006) and the impaired extinction in mice with increased PKA activity (McNally et al., 2005; Wang et al., 2004). It has been hypothesized recently that both memory formation as well as extinction are actively controlled by a tightly regulated balance between PKA and protein phosphatase 2B (PP2B) in which the one opposes the activity of the other (Mansuy,

88 PKA anchoring in fear conditioning

2003). In line with these findings it was reported that a reduction of PP2B signalling in forebrain neurons improves memory consolidation whereas it deteriorates fear extinction (Havekes et al., in press; Ikegami & Inokuchi, 2000; Lin et al., 2003). Our data showed that St-superAKAP-IS injection into the CA1 area of the dorsal hippocampus specifically reduced the amount of PKA bound to AKAP150 in this area. AKAP79/150 targets PKA to postsynaptic densities in neurons (Dell’Acqua et al., 2006) and is also able to bind PP2B (Dell’Acqua et al., 2002). In vitro studies using the peptide Ht31 showed that displacement of PKA from AKAP75/79/150 shifts the balance to PP2B activity (Snyder et al., 2005). Thus, AKAP79/150 might be an important coordinator of PKA and PP2B activity in memory consolidation and extinction. Recently, we and others provided additional evidence for an important role of AKAP79/150 in learning and memory. Electrophysiological measurements from hippocampal slices of mice with a stop codon inserted into the AKAP150 gene to truncate the last 36 residues, which constitute the PKA binding site, showed the importance of AKAP150-anchored PKA in LTP (Lu et al., 2007). We observed that AKAP150 is highly abundant in the mouse brain especially in those areas that are known to be involved in learning and memory (Ostroveanu et al., 2007). Moreover, the levels of hippocampal AKAP150 were elevated after exposure of animals to a novel context and during the consolidation phase of contextual fear conditioning, indicating that upregulated levels of AKAP150 contribute to processing the exposure to a novel context and the consolidation of associative learning (Nijholt et al., 2007). Although we cannot exclude the involvement of additional AKAPs, it thus seems likely that at least AKAP79/150 is important in the spatial compartmentalization of PKA signal transduction pathways that are active in the consolidation of contextual fear memories.

Both superAKAP-IS and Ht31 inhibit the anchoring of PKA to several AKAP species. However, whereas Ht31 has the potential to disrupt RII but also some RI mediated localization (Herberg et al., 2000), superAKAP-IS is a peptide that is 10,000-fold more selective for the RII isoform relative to RI (Gold et al., 2006). Our results show that RII anchoring is important in the consolidation and extinction of contextual fear memories. In future experiments the impact of the RI isoform-selective anchoring on learning and memory processes could be assessed using the RI anchoring disruptor (RIAD) (Carlson et

89 Chapter 4 al., 2006). To study in greater detail which specific AKAP is involved, it would be necessary to develop inhibitors that disrupt the interaction of PKA with one particular AKAP or to disrupt the interaction of PKA by introducing site-specific mutations in the PKA binding domain of a specific AKAP. Overall, our data suggest that the temporal and spatial specificity of the hippocampal PKA signaling pathway, mediated by AKAPs, is critical to consolidate long-term contextual fear memory whereas PKA anchoring to AKAPs may put a constraint on extinction.

Acknowlegdements

Part of this work was supported by The Netherlands Organization for Scientific Research (NWO-Vernieuwingsimpuls E.A.V.d.Z (Grant 016.021.017)) and grants to U.E. from the Dutch brain foundation (Hersenstichting Nederland), the International Alzheimer Foundation and by the European Union’s FP6 funding, NeuroproMiSe, LSHM-CT-2005- 018637. We thank Janne Papma for valuable technical assistance. This work reflects only the author’s views. The European Community is not liable for any use that may be made of the information herein.

References

Ahi, J., Radulovic, J., & Spiess, J. (2004). The role of hippocampal signaling cascades in consolidation of fear memory. Behavioral Brain Research, 149, 17-31.

Atkins, C. M., Selcher, J. C., Petraitis, J. J., Trzaskos, J. M., & Sweatt, J. D. (1998). The MAPK cascade is required for mammalian associative learning. Nature Neuroscience, 1, 602–609.

90 PKA anchoring in fear conditioning

Beene, D.L., & Scott, J.D. (2007). A-kinase anchoring proteins take shape. Current Opinion in Cell Biology, 19, 192-198.

Carlson, C.R., Lygren, B., Berge, T., Hoshi, N., Wong, W., Taskén, K., & Scott, J.D. (2006). Delineation of type I protein kinase A-selective signaling events using an RI anchoring disruptor. Journal of Biological Chemistry, 281, 21535-21545.

Franklin, K.B.J., & Paxinos, G. (1997). The mouse brain in stereotaxic coordinates. San Diego: Academic Press.

Havekes, R., Nijholt, I.M., Visser, A.K., Eisel, U.L., & Van der Zee, E.A. (2008). Transgenic inhibition of neuronal calcineurin activity in the forebrain facilitates fear conditioning, but inhibits the extinction of contextual fear memories. Neurobiology of Learning and Memory, 4, 595-598

91 Chapter 4

Ikegami, S., & Inokuchi, K. (2000). Antisense DNA against calcineurin facilitates memory in contextual fear conditioning by lowering the threshold for hippocampal long-term potentiation induction. Neuroscience, 98, 637-646.

Impey, S., Smith, D. M., Obrietan, K., Donahue, R., Wade, C., & Storm, D.R. (1998). Stimulation of cAMP response element (CRE)-mediated transcription during contextual learning. Nature Neuroscience, 1, 595–601.

Isiegas, C., Park, A., Kandel, E.R., Abel, T., & Lattal, K.M. (2006). Transgenic inhibition of neuronal protein kinase A activity facilitates fear extinction. Journal of Neuroscience, 26, 12700-12707.

Lin, C. H., Yeh, S. H., Leu, T. H., Chang, W. C., Wang, S. T., & Gean, P.W. (2003). Identification of calcineurin as a key signal in the extinction of fear memory. Journal of Neuroscience, 23, 1574-1579.

Liu, J., Hu, J.Y., Schacher, S., & Schwartz, J.H. (2004). The two regulatory subunits of aplysia cAMP-dependent protein kinase mediate distinct functions in producing synaptic plasticity. Journal of Neuroscience, 24, 2465-2474.

Lu, Y., Allen, M., Halt, A.R., Weisenhaus, M., Dallapiazza, R.F., Hall, D.D., Usachev, Y.M., McKnight, G.S., & Hell, J.W. (2007) Age-dependent requirement of AKAP150-anchored PKA and GluR2-lacking AMPA receptors in LTP. EMBO Journal, 26, 4879-4890.

Lu, Y., Lu, Y.S., Shuai, Y., Feng, C., Tully, T., Xie, Z., Zhong, Y., & Zhou, H.M. (2007). The AKAP Yu is required for olfactory long-term memory formation in Drosophila. Proceedings of the National Academy of Sciences United States ofAmerica, 104, 13792-13797.

Mansuy, I.M. (2003). Calcineurin in memory and bidirectional plasticity. Biochemical and Biophysical Research Communications, 311, 1195-1208.

McNally, G.P., Lee, B.W., Chiem, J.Y., & Choi, E.A. (2005). The midbrain periaqueductal gray and fear extinction: Opioid receptor subtype and roles of cyclic AMP, protein kinase A, and mitogen- activated protein kinase. Behavioral Neuroscience, 119, 1023-1033.

Moita, M.A., Lamprecht, R., Nader, K., & LeDoux, J.E. (2002). A-kinase anchoring proteins in amygdala are involved in auditory fear memory. Nature Neuroscience, 5, 837-838.

Nie, T., McDonough, C.B., Huang, T., Nguyen, P.V., & Abel, T. (2007). Genetic disruption of protein kinase A anchoring reveals a role for compartmentalized kinase signaling in theta-burst long-

92 PKA anchoring in fear conditioning term potentiation and spatial memory. Journal of Neuroscience, 27, 10278-10288.

Nijholt, I., Farchi N., Kye, M., Sklan, E.H., Shoham, S., Verbeure, B., Owen, D., Hochner, B., Spiess, J., Soreq, H., & Blank, T. (2004). Stress-induced alternative splicing of acetylcholinesterase results in enhanced fear memory and long-term potentiation. Molecular Psychiatry, 9, 174-183.

Nijholt, I.M., Ostroveanu, A., de Bruyn, M., Luiten, P.G., Eisel, U.L., & Van der Zee E.A. (2007). Both exposure to a novel context and associative learning induce an upregulation of AKAP150 protein in mouse hippocampus. Neurobiology of Learning and Memory, 87, 693-696.

Ostroveanu, A., Van der Zee, E.A., Dolga, A.M., Luiten, P.G., Eisel, U.L. & Nijholt, I.M. (2007). A- kinase anchoring protein 150 in the mouse brain is concentrated in areas involved in learning and memory. Brain Research, 1145, 97-107.

Radulovic, J., Kammermeier, J., & Spiess, J.(1998) Generalization of fear responses in C57BL/6N mice subjected to one-trial foreground contextual fear conditioning. Behavioral Brain Research, 95, 179-189.

Rubin, C.S. (1994). A kinase anchor proteins and the intracellular targeting of signals carried by cyclic AMP. Biochimica et Biophysica Acta, 1224, 467-479.

Rudy, J.W. & Morledge, P. (1994). Ontogeny of contextual fear conditioning in rats: implications for consolidation, infantile amnesia, and hippocampal system function. Behavioral Neuroscience, 108, 227-234.

Sanders, M.J., Wiltgen, B.J., & Fanselow, M.S. (2003). The place of the hippocampus in fear conditioning. European Journal of Pharmacology, 463, 217-223.

Schafe, G.E., Nadel, N.V., Sullivan, G.M., Harris, A., & LeDoux, J.E. (1999). Memory consolidation for contextual and auditory fear conditioning is dependent on protein synthesis, PKA, and MAP kinase. Learning and Memory, 6, 97-110.

Schwaerzel, M., Jaeckel, A., & Mueller, U. (2007). Signaling at A-kinase anchoring proteins organizes anesthesia-sensitive memory in Drosophila. Journal of Neuroscience, 27, 1229-1233.

Snyder, E.M., Colledge, M., Crozier, R.A., Chen, W.S., Scott, J.D., & Bear M.F. (2005). Role for A kinase-anchoring proteins (AKAPS) in glutamate receptor trafficking and long term synaptic depression. Journal of Biological Chemistry, 280, 16962-16968.

93 Chapter 4

Taylor, S.S., Buechler, J.A., & Yonemoto, W. (1990). cAMP-dependent protein kinase: Framework for a diverse family of regulatory enzymes. Annual Review of Biochemistry, 59, 971-1005.

Vijayaraghavan, S., Goueli, S.A., Davey, M.P., & Carr, D.W. (1997). Protein kinase A-anchoring inhibitor peptides arrest mammalian sperm motility. Journal of Biological Chemistry, 272, 4747- 4752.

Wallenstein, G.V., Vago, D.R., & Walberer, A.M. (2002). Time-dependent involvement of PKA/PKC in contextual memory consolidation. Behavioral Brain Research, 133, 159-164.

Wang, H., Ferguson, G.D., Pineda, V.V., Cundiff, P.E., & Storm, D.R. (2004). Overexpression of type-1 adenylyl cyclase in mouse forebrain enhances recognition memory and LTP. Nature Neuroscience, 7, 635– 642.

Wong, W., & Scott, J.D. (2004). AKAP signalling complexes: focal points in space and time. Nature Reviews Molecular Cell Biology, 5, 959-970.

94 Chapter 5

Detailed analysis of mAKAP expression in the brain of young and old mice

Anghelus Ostroveanu, Eddy A. van der Zee, Ulrich L. M. Eisel, Ingrid M. Nijholt

Department of Molecular Neurobiology, Graduate School of Behavioral and Cognitive Neurosciences, University of Groningen, The Netherlands

In preparation Chapter 5

Abstract

A-kinase anchoring proteins (AKAPs) form large macromolecular signaling complexes that specifically target cAMP-dependent protein kinase (PKA) to unique subcellular compartments and thus provide high specificity to PKA signaling. One important member of this family is mAKAP. Although mAKAP was reported to be present in the brain, so far knowledge on its distribution and function was lacking. In the present study we established a detailed distribution of mAKAP the mouse brain. In general, mAKAP could be found throughout the whole mouse brain. mAKAP expression was particularly abundant in the hippocampus, hypothalamus and olfactory bulb, whereas the cortex, cerebellum and striatum revealed a more moderate mAKAP immunoreactivity. The lowest expression level was detected in the brain stem. At subcellular levels, mAKAP was abundantly expressed both in perikarya (e.g: pyramidal neurons, Purkinje neurons), dendrites (e.g.: apical and basal dendrites of hippocampus pyramidal neurons) and to some extent in fibers (e.g: amygdala, cortex). Interestingly, the highest mAKAP expression was detected presynaptically in brain regions such as central amygdaloid nucleus, supraoptic nucleus, paraventricular hypothalamus nucleus, and suprachiasmatic nucleus. In aged mice, we found a dramatic decrease in mAKAP expression in all brain regions. Our finding that mAKAP is abundantly expressed, particularly presynaptically, suggest an important role for mAKAP in coordinating signal transduction pathways involved in proper brain functioning. Moreover, the strongly decreased expression in senescent mice suggests that mAKAP may contribute to changes in brain functioning in aged mice.

Key words: mAKAP; immunohistochemistry; brain; mouse, cAMP-dependent protein kinase, localization.

96 mAKAP in the mouse brain

Introduction cAMP-dependent protein kinase (PKA) mediates cellular responses to a wide range of hormones, neurotransmitters, and other signaling substances. Specificity of PKA signaling is in part achieved by binding to A-kinase anchoring protein (AKAP) molecules via interaction with the regulatory subunits (RI and RII) of PKA. Besides binding to PKA, AKAPs are capable of forming multi-protein complexes to coordinate the action of several signalling molecules all at a single location. To date, over 50 AKAPs have been described from diverse species ranging from C. elegans and rodents to humans. A prototypic example of an AKAP which carries multiple signaling enzymes is muscle specific AKAP (mAKAP, AKAP6 or AKAP100). mAKAP was for the first time identified in 1995 as AKAP100 using an interaction cloning strategy with an RIIα protein probe on a human hippocampal cDNA expression library (McCartney et al., 1995). Later AKAP100 was renamed mAKAP because of its presence in striated muscle (Kapiloff et al., 1999). A targeting domain is responsible for linkage of the complex to the nuclear envelope and sacroplasmatic reticulum (Kapiloff et al., 1999). The mAKAP gene structure was shown to be conserved in humans and rodents and two alternatively spliced forms of mAKAP were characterized: mAKAPα and mAKAPβ. mAKAPα and mAKAPβ are identical except for a 244 amino acid residue N-terminal extension in mAKAPα (Michel et al., 2005). The longer form, mAKAPα, is preferentially expressed in the brain, whereas mAKAPβ is abundant in cardiac myocytes and skeletal muscle (Michel et al., 2005). Hitherto, the function of mAKAP in signaling events in cardiac myocytes has become increasingly clear under physiological as well as under pathological conditions. In cardiac myocytes, PKA anchored to mAKAP participates in calcium signaling via phosphorylation of the calcium-activated calcium release channel RyR (Kapiloff et al., 2001; Ruehr et al., 2003). Increased intracellular levels of calcium can activate the calcium-dependent protein phosphatase 2B (PP2B) which was also found to bind mAKAP (Pare et al., 2005). This phosphatase dephosphorylates the transcription factor NFAT, allowing for movement of NFAT into the nucleus and the transcription of genes. Interestingly, mAKAP was also found to bind a phosphodiesterase (PDE4D3). In cardiac myocytes, by tethering both PKA

97 Chapter 5 and PDE4D3, mAKAP assembles a negative feedback loop that functions to regulate both PKA and PDE activity (Dodge-Kafka et al., 2005). Upon stimulation of the cell increased levels of cAMP lead to PKA activation, while activated PKA phosphorylates PDE4D3. This phosphorylation of PDE4D3 in turn increases the metabolic activity of PDE4D3 resulting in decreased cAMP levels. These effects were found to be counterbalanced by the MAP kinase extracellular-signal-regulated kinase 5 (ERK5) which phosphorylates PDE4D3 on a different serine site which results in suppressed PDE activity and increased PKA activity (Dodge-Kafka et al., 2005). Overall, by binding PKA, PP2B, RyR, PDE and a MAP kinase, mAKAP serves to integrate cAMP, calcium, and MAPK signals for the coordination of intracellular processes in the heart. The importance of proper coordination of signaling pathways is perfectly exemplified by the function of mAKAP in the heart. Defective mAKAP signaling has e.g. been linked to cardiac hypertrophy (Pare et al., 2005; Bauman et al., 2007).

The role of mAKAP in the brain remains so far rather unexplored. It was recently reported that the unique amino terminus of the most prominent mAKAP form in the brain, mAKAPα, can spatially restrict the activity of 3-phosphoinositide-dependent kinase-1 (PDK1). Moreover, the simultaneous recruitment of PDK1 and ERK onto mAKAPα was reported to facilitate activation and release of the downstream target p90RSK (Michel et al., 2005). AKAP100 mRNA was detected in the human cerebellum, basal ganglia, hippocampus, frontal and motor cortex (McCartney et al., 1995). These data, together with the importance of the different mAKAP signaling complex members in numerous brain functions, makes its very likely that mAKAP in the brain has a similar function as in the heart i.e. the integration of several signaling pathways to regulate neuronal processes.

The present study adds to the fundamental knowledge of mAKAP in the brain by mapping its distribution in detail in young and aged mice.

98 mAKAP in the mouse brain

Materials and methods

Animals and housing conditions 3 and 26 months old male C57Bl/6J inbred mice (Harlan, Horst, The Netherlands, n=10) were group housed in standard macrolon cages and placed in an environmentally controlled room under a 12:12 light/dark cycle. Animals had free access to water and standard food pellets. All procedures concerning animal care and treatment were in accordance with the regulation of the ethical committee for the use of experimental animals of the University of Groningen, The Netherlands (DEC 4278A).

Western blotting

CO2/O2 anesthetized animals were quickly decapitated and brain tissue was removed on ice. The following brain regions were excised, immediately frozen in liquid nitrogen and stored at -80 °C before further processing: hippocampus, striatum, olfactory bulb, cortex, cerebellum, hypothalamus and brain stem. Brain, colon, and heart tissue was mechanically homogenized in 10 volumes of homogenization buffer (50 mM HEPES (pH 7.4), 150 mM NaCl, 0.2 % NP-40, 4 mM EGTA, 10 mM EDTA, 15 mM sodium pyrophosphate, 100 mM β-glycerophosphate, 50 mM sodium fluoride, 5 mM sodium orthovanadate, 1 mM dithiothreitol, 1 mM PMSF, and Complete Mini Protease Inhibitor Cocktail (Roche)). The homogenate was centrifuged at 20,000 x g for 10 min at 4 °C, and the resulting supernatant was assayed for protein concentration using the Bradford method. Protein samples (20 µg per sample) were separated on a 6 % SDS polyacrylamide gel and transferred to PVDF membranes (Millipore, USA). The blots were blocked for 1 h in blocking buffer (0.2 % I-Block (Tropix), 0.1 % Tween20) and then incubated overnight at 4 °C either with rabbit anti-mAKAP (1:8.000; Abcam, ab24639). Mouse anti-actin antibody (1:40,000; MP Biomedicals, Irvine, CA, USA) served as control for protein loading. The blots were incubated with horse radish peroxidase conjugated secondary antibodies (HRP conjugated anti-mouse (1:5,000)) (sc-2005 Santa Cruz, CA, USA) and HRP conjugated anti-rabbit (1:5,000) (NA934V, Amersham). Western blots were developed using the

99 Chapter 5 chemiluminescence method (Pierce). The immunoblots were digitized and quantified using a Leica DFC 320 image analysis system (Leica, Cambridge, UK).

Immunohistochemistry Animals were anesthetized by intraperitoneal injection with sodium pentobarbital 6% solution and sacrificed by transcardial perfusion with saline solution containing heparin, followed by 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB). After perfusion, the brains were kept at 4 °C in 0.01 M phosphate buffered saline (PBS) containing 0.1% sodium azide for 72 h. All brains were dehydrated for 48 h in a 30% buffered sucrose solution. After dehydration, 20 µm coronal or sagittal sections were cut on a cryostat microtome. Sections were stored at 4 °C in 0.01 M PBS containing 0.1% sodium azide. The avidin/biotin immunoperoxidase staining method was used to visualize mouse mAKAP

IR. Coronal and sagittal sections were preincubated with 0.3% H2O2 to reduce endogenous peroxidase. Non-specific binding sites were blocked by preincubating the sections with 5% normal rabbit serum in 0.01 M PBS for 30 min. Subsequently sections were incubated with rabbit polyclonal anti mAKAP, 1:1000 dilution (Abcam, ab24639) in 0.01 M PBS containing 5% normal goat serum and 0.3% Triton X-100 for 2 h at room temperature (RT) and left overnight at 4 °C. Afterwards the sections were rinsed and incubated at RT for 2 h with biotin-conjugated goat anti-rabbit antibody (1:400) (Jackson Inc., 115-065-145) in 0.01 M PBS containing 1% normal goat serum and 3% Triton X-100. This was followed by incubation with avidin complex containing biotinylated horseradish peroxidase (1:400) (Vectastain ABC Kit, Burlingame, CA, USA) for 2 h at RT. Finally, staining was visualized with 0.03% diaminobenzidine (DAB) chromogen substrate and 0.001% H2O2. Immunostained sections were analyzed with an Olympus BH2 microscope (Olympus, Japan). Two independent investigators established semi-quantification of mouse mAKAP IR. For each brain region, material from 6 animals was examined to establish mAKAP immunostaining. The scoring system used for establishing the relative AKAP IR was classified as follows: absent (-), low (-/+), moderate (+), high (++) and very high (+++). In case of very high staining in which presynaptic elements were detected, the immunoreactivity was classified as ++++.

100 mAKAP in the mouse brain

Photographs were taken with a DM1000/DFC280 Leica image analysis system (Leica, Cambridge, UK).

Antibody specificity Since there are no blocking peptides available for any mAKAP antibodies, we ascertained the specificity of both antibodies using several different approaches. Initially, parallel staining in free-floating brain sections was performed without primary antibodies. In these sections we could not observe any staining (Fig. 1A).

A C CA1

CA2

DG CA3

b a a

Fig. 1. mAKAP antibody specificity. (A) Control staining without mAKAP primary antibody in the hippocampus. CA1, CA2, CA3, cornu ammonis 1, 2, 3. DG, dentate gyrus, (B) mAKAP IR in the colon. a: lamina mucosa, b: lamina submucosae, c: lamina muscularis. (C) Positive mAKAP staining in mouse heart tissue. Enlarged: transversal section of muscular cells in the right ventricular wall of the mouse heart. Scale bar: A = 620 µm, B = 350 µm, C = 250 µm.

Furthermore, preincubation steps were performed to increase the specificity of mAKAP antibodies in the mouse brain. It was previously reported the mAKAP is not expressed in the colon (Kapiloff et al., 1999). Therefore, we tested the antibody specificity using mouse colon as a negative control and mouse heart and brain tissue as positive control. Colon, heart and brain tissue from perfused mice were dehydrated for 48 h in a 30% buffered sucrose solution. After dehydration, 20 µm coronal or sagittal sections were directly collected on precoated microscope glasses. mAKAP staining was performed on glass as described for immunostaining of free-floating sections. Controls without primary antibodies were also included. mAKAP immunoreactivity could not be detected in the colon (Fig. 1B)

101 Chapter 5 while brain and heart tissue showed high mAKAP staining (Fig. 1C). In addition, no staining was detected in the control sections without primary or secondary antibody. Also in Western blot experiments mAKAP was not found in colon tissue. Overall we conclude that the rabbit polyclonal anty mAKAP antibody is specific for mAKAP in mouse tissue.

Results

Western blot analysis of mAKAP expression levels in various brain compartments We detected mAKAP in homogenates of heart and cortex (brain) tissue but not in the colon of mice. Western blot analysis showed that our antibody can specifically detect both mAKAPα in the brain and mAKAPβ in the heart (Fig. 2A).

A 250 kDa mAKAP Fig. 2. Western blot analysis of mAKAP mAKAP expression levels in mouse brain. (A) mAKAP

Cortex Colon Heart antibody detects both mAKAPα and mAKAPβ. Tissue source is indicated below each lane. Colon B 100

) was used as negative control (B) Bar graph (% 80 representing mAKAP expression levels in different

60 compartments of the mouse brain. Values represent

expression mean percentages of integrated optical density 40 (I.O.D) ± S.E.M. Hippocampus was set to 100% 20 mAKAP (C) Representative Western blot with protein extracts from different brain regions. Actin served C mAKAP as control for protein load. Blots are representative Actin of three independent experiments.

bulb stem Cortex Striatum CerebellumBrain Hippocampus OlfactoryHypothalamus

Western blot analysis on protein extracts from different mouse brain regions revealed the highest expression level of mAKAP in hippocampus, hypothalamus, and olfactory bulb (Fig. 2B, C). Relatively high expression levels were also detected in the cortex, striatum, and cerebellum, while the brain stem showed low levels of mAKAP (Fig. 2B, C).

102 mAKAP in the mouse brain

General overview of mAKAP IR in mouse brain Immunohistochemical staining on coronal brain slices was performed to investigate the distribution of mAKAP in more detail. Antibody specificity was ascertained as described in materials and methods (section “Antibody specificity”, Fig. 1). Overall, mAKAP was found to be widely distributed throughout the mouse brain. The highest mAKAP expression was detected presynaptically. For example, areas such as periventricular nucleus, supraoptic nucleus, suprachiasmatic nucleus and central amygdala were highly immunoreactive for mAKAP (Table 1, Fig. 3). High mAKAP immunoreactivity (IR) was also observed in various other regions like olfactory bulb, cortex, cerebellum, and hippocampus (Fig. 3). At the cellular level mAKAP was mostly expressed in cell bodies (e.g. the Purkinje cell layer of the cerebellum, pyramidal neurons in the cortex and hippocampus), but it was also found in fibers (e.g. stratum radiatum of the hippocampus, mossy fibers of the dentate gyrus). Some brain regions did not show any IR, such as the corpus callosum and the fornix (Fig. 3). Immunopositive endothelial cells or glia could not be detected.

Detailed description of mAKAP expression in mouse brain

Olfactory system mAKAP is strongly expressed in the olfactory system. The accessory olfactory bulb (one of the strongest mAKAP immunoreactive regions in the mouse brain) was very densely stained and revealed mAKAP staining in fibers. The external plexiform layer and mitral cell layer also showed very high levels of IR of mAKAP. In these areas clear perikarya staining could be detected. In contrast, the glomerular layer, the granule layer, and internal plexiorm layer showed only moderate levels of mAKAP IR. Very high mAKAP IR was also found in the olfactory tubercle, in tenia tecta layers 1-3 and anterior olfactory nucleus, where perikarya staining was pronounced. (Fig. 3A, B, Table 1).

Cerebral cortex mAKAP protein was found to be expressed throughout all layers of the cortex (Fig. 4A) The highest mAKAP IR in the mouse cortex was observed in cortical layer II (external granular layer), cortical layer III (external pyramidal layer) and cortical layer V (internal

103 Chapter 5 pyramidal layer). Cortical layer II was highly immunoreactive for mAKAP protein in the perikarya of its granule cells (Fig. 4A). In layers III and V, which represent the principal output system of the neocortex, pyramidal neurons were strongly stained for mAKAP (Fig. 4A). Layer IV (internal granular layer) and the deepest cortical layer, layer VI (polymorphic layer), were immunoreactive for mAKAP to a somewhat lower extent (Fig. 4A). The lowest cortical IR was found in cerebral cortex layer I (molecular layer). In addition, the entorhinal cortex, the major source of afferents to the hippocampal formation, was intensely stained (Table 1). At subcellular level, besides perikarya, dendrites were immunoreactive for mAKAP while occasionally immunopositive fibers with varicosities were detected (data not shown).

A B C D E 3.92mm 3.20mm 1.98mm 0.62mm -0.70mm

AT AOB Cpu Cpu

OB Pir BLA AO PaV

F G H -1.34mm -2.80mm -5.80mm P H Gr SC Mol

CA SNR LH

Fig. 3. Overview of mAKAP expression in coronal sections of the brain. Distance from bregma is indicated directly on the photographs (Franklin & Paxinos, 1997). (A) OB, olfactory bulb, (B) AOB, accessory olfactory bulb, (C) Pir, piriform cortex, AO, anterior olfactory nucleus, (D) Cpu, caudate putamen (striatum), (E) AT, anterior thalamus, BLA, basolateral amygdala, PaV, paraventricular hypothalamic nucleus ventral, Cpu, caudate putamen, (F) LH, lateral hypothalamic area, H, hippocampus, CA, central amygdala, (G) SC, superior colliculus, SNR, substantia nigra pars reticulata, (H) Mol, cerebellar molecular layer, Gr, cerebellar granular layer, P, Purkinje cell layer. Scale bar = 1300 µm

104 mAKAP in the mouse brain

Hippocampus Overall within this laminar brain structure, mAKAP displayed high IR (Fig. 4B; Table 1). Especially in the CA1-3 region of the hippocampus, mAKAP showed a very high expression particularly in the pyramidal layer where very high IR was detected in the perikarya of the pyramidal neurons (Fig. 4B, D, E).

B A I SO Py SR II-IV SLM

GCL

V hil ML

CDE

Fig. 4. Distribution of mAKAP protein in the cerebral cortex and hippocampus. (A) Overview of mAKAP protein expression in the cerebral cortex somatosensory 1, trunk region: I-VI cortical layers: I - molecular layer, II - external granular layer, III - external pyramidal layer, IV - internal granular layer, V - internal pyramidal layer, VI polymorphic layer. Enlarged: densely stained pyramidal neurons in cortical layer V, (B) mAKAP IR in the hippocampus: SO, stratum oriens, SR, stratum radiatum, SLM, stratum lacunosum moleculare, Py, stratum pyramidale, ML, dentate gyrus molecular layer, GCL, dentate granular cell layer and hil, hilus , (C) mAKAP staining in subiculum, (D) mAKAP IR in CA1 area of the hippocampus. Non-principal cells are also stained for mAKAP (arrows), (E) mAKAP IR in the CA3 area of the hippocampus. Enlarged: CA3 pyramidal neurons densely immunoreactive for mAKAP. Scale bar: A = 175 µm, B = 400 µm, C, D and E = 100 µm.

In stratum oriens and stratum lacunosum moleculare moderate to high mAKAP IR was observed. A divergent staining pattern was observed in the stratum radiatum (the apical dendrites of pyramidal neurons): Whereas in the CA1 area the apical dendrites were moderately stained (Fig. 4D), in the CA3 area extremely dense staining could be observed

105 Chapter 5

(Fig. 4E). Moreover, the densely stained stratum lucidum revealed a mixture of mAKAP IR in mossy fibers and dendrites of CA3 pyramidal neurons. The subiculum also showed very high mAKAP IR in perikarya (Fig. 4C). Interestingly, scattered non-principal cells in the hippocampus were strongly stained for mAKAP protein (Fig. 4D). In the dentate gyrus, mAKAP IR was characterized by moderate staining in the granule cells, while the granule cell layer showed a somewhat higher mAKAP IR (Fig. 4B). The staining made it difficult to distinguish subcellular compartments within the granule cell layer. The molecular layer of the dentate gyrus showed only moderate staining, whereas the hilus was slightly more immunoreactive (Fig. 4B).

Amygdala The distribution of mAKAP was clear and strong throughout all amygdaloid nuclei (Table 1). The most prominent staining was detected in the central amygdaloid nucleus (Fig. 5), the basomedial amygdaloid nucleus (anterior and posterior) and intercalated nuclei (Table 1). In spite of the dense staining, perikarya staining could still be observed in these nuclei.

Fig. 5. mAKAP IR in the amygdala. CA, central amygdaloid nucleus, LaDL, lateral amygdaloid nucleus LaDL dorsolateral, LaVL, lateral amygdaloid nucleus ventrolateral, BLA, basolateral amygdaloid nucleus. Enlarged: cluster of perikarya in the basolateral amygdala. CA LaVL Scale bar = 200 µm. BLA

The basolateral complex revealed a strong IR in the ventral nucleus and a more moderate staining in the anterior nucleus in which both neuronal perikarya and fibers were detected (Figs. 3E, F, 5, Table 1). Moderate mAKAP IR was also observed in the medial amygdaloid nucleus, lateral amygdaloid nucleus, amygdalohippocampal area and amygdalopiriform transition area (Table 1).

106 mAKAP in the mouse brain

Septal nuclei and striatopallidal system Overall, the distribution of mAKAP in the septal nuclei and the striatopallidal system was strong with only slight differences in the density of staining (Table 1). In the septum, the lateral nucleus was strongly stained, whereas the medial nucleus showed a somewhat lower expression (Fig. 3D; Table 1). In the caudate putamen all transected fiber bundles are mAKAP immunoreactive and also perikarya staining is present. Claustrum and ventral pallidum were also highly IR for mAKAP.

Epithalamus and thalamus In the epithalamus mAKAP showed a moderate to high expression. In the medial habenula we observed high IR with visible perikarya staining (Table 1), whereas the lateral habenula was only moderately stained. Moreover, in the habenula, occasionally mAKAP positive could be detected. At the level of the thalamus the overall distribution of mAKAP revealed a very high expression (Fig. 3E, Table 1). Detailed analysis of the stained thalamus displayed strong staining in the posterior nuclear group, the parafascicular thalamic nuclei and the subgeniculate nucleus. The ventrolateral thalamic nucleus was also highly immunoreactive for mAKAP. In contrast to the other areas, stained perikarya could be observed in this region. The remainder of the thalamus (i.e.: paratenial nucleus, rhomboid nucleus, ethmoid nucleus, mediodorsal nucleus) showed only moderate levels of mAKAP (Table 1). The lowest mAKAP IR was observed in the rhomboid nucleus and reuniens.

Hypothalamus In general, the expression level of mAKAP in the hypothalamus was high. Accordingly, paraventricular and suprachiasmatic nucleus displayed the highest mAKAP staining whereas the median eminence, arcuate nucleus, mammilary nuclei and periventricular nucleus showed a reduced but still strong pattern of staining (Fig. 3F, Table 1), (Table 1).

Midbrain mAKAP protein was moderately to very strongly expressed in the midbrain region (Table 1). The highest staining was observed in the substantia nigra pars reticulata, interpeduncular

107 Chapter 5 and pointine nuclei (Fig. 3G). Moderate levels of mAKAP IR were detectable in inferior and superior colliculus. Low levels were found in the periaqueductal gray and brachium pontis (Table 1).

Cerebellum A high homogeneous staining was seen throughout the entire cerebellar cortex with no apparent expression difference between the molecular and granule cerebellar layer (Fig. 6A). The Purkinje cell bodies and their fibers were very clearly stained covering the complete cerebellar cortex (Fig 6, Table 1).

A B Fig. 6. mAKAP expression in the cerebellum. (A) Mol, molecular layer, Gr, granular layer. White arrow shows Mol purkinje neurons with staining in both cell body and dendrites, (B) facial nucleus, (C) raphne magnus nucleus, (D) spinal trigeminal tract. Scale bar: A, C and D = 15 µm,

Gr B = 30 µm.

C D

Brain stem High levels of mAKAP were also detected in the raphne nucleus, facial nucleus, lateral (dentate) cerebellar nucleus and the dorsal cochlear nucleus (Fig. 3H, Tabel 1). In these nuclei, at higher magnification, the staining was detected in both cell bodies and fibers. In the spinal trigeminal tract, moderate staining was detected in fibers. Lower mAKAP IR was observed in the inferior cerebellar peduncle (Tabel 1).

108 mAKAP in the mouse brain mAKAP expression in aged mice Finally, we compared mAKAP protein levels of young (4-6 months) and old (24-26 months) mice. Immunohistochemical analysis revealed a different expression level for mAKAP in young and old mice. Accordingly, mAKAP IR was dramatically decreased throughout the entire brain of old mice. Brain regions in which mAKAP was very highly expressed in young mice (e.g.: paraventricular hypothalamic nucleus, central amygdala, etc.) exhibit only low mAKAP expression in old mice (Fig. 7).

A B Fig. 7. mAKAP expression in several brain regions of 26 SO I months old mouse brain: (A) mAKAP expression in the

Py cortex. I-VI cortical layers: I - molecular layer, II - external granular layer, III - external pyramidal layer, IV - internal granular layer, V - internal pyramidal layer, VI II-IV SR polymorphic layer, (B) mAKAP expression in the CA1 area of the hippocampus: SO, stratum oriens, Py, stratum C pyramidale, SR, stratum radiatum, (C) Arrow indicates low mAKAP IR in paraventricular hypothalamus nucleus, (D) mAKAP IR in the amyglada: CA, central amygdaloid V nucleus, LaDL, lateral amygdaloid nucleus dorsolateral, LaVL, lateral amygdaloid nucleus ventrolateral, BLA, basolateral amygdaloid nucleus. Scale bar: A = 180 µm, B D = 100 µm, C and D = 250 µm VI LaDL CA

LaVL

BLA

To confirm these results, we compared mAKAP protein expression in the hippocampus and cortex from young and old mice using Western blot analysis. Both in the cortex and the hippocampus of old mice, the levels of mAKAP were significantly decreased when compared to young controls (cortex 61% reduction, hippocampus 47 % reduction; Fig. 8).

109 Chapter 5

A Hippocampus Fig. 8. Western blot analysis of mAKAP Cortex

100 expression levels in young and old mouse )

(% brain (A) mAKAP expression levels of 80 hippocampus and cortex in brain tissues from 3

60 and 26 months old mice. Values represent expression mean percentages of integrated optical density 40

(I.O.D). ± S.E.M. Hippocampus was set to mAKAP 20 100%, (B) Representative Western blot for mAKAP. Actin was used as control for protein 4 months old 24 months old load. B mAKAP Actin

Cortex Cortex

Hippocampus Hippocampus

Table 1 - Expression of mAKAP protein in Hilus ++ various compartments of the mouse central CA1 region nervous system. Stratum oriens ++ Stratum radiatum ++ Brain region mAKAP Stratum lacunosum moleculare ++/+++ Stratum pyramidale Telencephalon CA3 region Olfactory system Stratum oriens ++ Glomerular layer ++ Stratum radiatum +++ External plexiform layer +++ Stratum pyramidale +++ Mitral cell layer +++ Stratum lucidum +++ Internal plexiform layer +++ Subiculum +++ Granule layer ++ Entorhinal cortex ++/+++ Anterior olfactory nucleus ++ Amygdala Olfactory tubercle +++ Central amygdaloid nucleus ++++ Tenia tecta, layers 1-3 ++/+++ Medial amygdaloid nucleus +/++ Anterior commisure, anterior -/+ Lateral amygdaloid nucleus ++ Accessory olfactory bulb +++ Basolateral amygdaloid nucleus, ++ Cerebral cortex anterior Layer I +/++ Basolateral amygdaloid nucleus, ++/+++ Layer II +++ ventral Layer III +++ Basomedial amygdaloid nucleus ++/+++ Layer IV ++ Amygdalohippocampal area +++ Layer V ++/+++ Amygdalopiriform transition area ++ Layer VI ++ Intercalated nuclei +++ Piriform cortex ++ Posteromedial cortical +++ Hippocampal formation amygdaloid nucleus Dentate gyrus Septal and basal magnocellular nuclei Granule cell + Bed nucleus stria terminalis ++++ Granule cell layer ++ Nucleus accumbens core ++/+++ Molecular layer ++/+++ Nucleus accumbens shell ++/+++

110 mAKAP in the mouse brain

Tabel 1 (Continued) Habenula lateral ++ Hypothalamus Brain region mAKAP Median eminence ++++ Arcuate nucleus +++ Septal and basal magnocellular nuclei Supraoptic nucleus +++ Substantia innominata +++ Paraventricular nucleus ++++ Septum lateral nucleus +++ Periventricular nucleus +++ Septum medial nucleus ++ Suprachiasmatic nucleus ++++ Striatopallidal system Mammillary nuclei +++ Caudate putamen ++ Fornix ++ Globus pallidus ++ Lateral globus pallidus ++ Mesencephalon Ventral pallidum ++ Midbrain Claustrum ++ Superior colliculus ++ Islands of Calleja +/++ Inferior colliculus ++ Interpeduncular nuclei +++ Diencephalon Pontine nuclei +++ Thalamus Substantia nigra Ventrolateral nucleus ++/+++ pars reticulata ++/+++ Ventral posterolateral ++/+++ pars compacta ++ Ventral posteromedial ++/+++ Periaqueductal gray + Mediodorsal nucleus ++ Brachium pontis + Central nuclei ++ Paracentral nucleus ++ Rhombencephalon Parafascicular thalamic nuclei +++ Cerebellum Paraventricular thalamic nuclei ++ Molecular layer ++ Subgeniculate nucleus ++/+++ Purkinje cell layer +++ Reticular thalamic nucleus +/++ Granule cell layer ++ Posterior nuclear group +++ Raphne nucleus ++/+++ Paratenial nucleus ++ Facial nucleus +++ Rhomboid nucleus +/++ Lateral cerebellar nucleus ++/+++ Ethmoid nucleus ++ Dorsal cochlear nucleus ++ Reuniens +/++ Spinal trigeminal tract ++ Subthalamic nucleus ++/+++ Inferior cerebellar peduncle + Zona incerta ++ Scoring was classified as absent (-), low (-/+), Epithalamus moderate (+), high (++), very high (+++), very Habenula medial +++ high including presynaptic staining (++++).

Discussion

Previous studies reported that mAKAP is predominantly expressed in cardiac and skeletal muscle, but it was also observed to some extent in the brain (Kapiloff et al., 1999). Nevertheless, detailed knowledge on the localization and the function of mAKAP in the brain remained so far rather limited. In this study we investigated the distribution of mAKAP in the brain of young and old mice in detail. Our results show that overall, mAKAP is abundantly expressed in the mouse brain and its distribution is thus not

111 Chapter 5 restricted to a particular compartment. Very high mAKAP IR was detected in the cortex, cerebellum, thalamus, hypothalamus and hippocampus. These results are in good agreement with the reported mRNA expression of AKAP100/mAKAP in several human brain regions (Kapiloff et. al., 1999; McCartney et al., 1995). AKAP100 mRNA was reported to be highly expressed in the human motor and frontal cortex, hippocampus, basal ganglia and cerebellum (McCartney et al., 1995). Moreover, our results from immunohistochemistry were confirmed by Western blot analysis in several compartments of the mouse brain. So far two mAKAP variants have been identified: mAKAPα (preferentially expressed in the brain) and mAKAPβ (preferentially expressed in the heart) (Michel et al., 2005). mAKAPβ is a shorter form of the anchoring protein that lacks the first 244 amino acids. The unique amino terminus of mAKAPα can spatially restrict the activity of 3-phosphoinositide- dependent kinase-1 (PDK1) (Michel et al., 2005). Indeed, we detected on Western blot proteins with a lower motility in the mouse brain corresponding to mAKAPα whereas in the mouse heart the shorter mAKAPβ was observed (Kapiloff et al., 1999).

Michel and colleagues generated mice that have mAKAPα exchanged for mAKAPβ. mAKAPα null mice had reduced viability and a smaller body size as compared to wild-type and heterozygous littermates and showed craniofacial defects (Michel et al., 2005). These data suggest that switching the tissue-specific expression pattern of mAKAP forms has critical effects on the formation of brain signaling complexes which are maintained by this anchoring protein.

Although our results showed mAKAP IR throughout the entire mouse brain, detailed analysis revealed some striking differences between particular brain regions. For example, the highest mAKAP IR was detected in discrete regions of the limbic system such as paraventricular hypothalamic nucleus, supraoptic nucleus, suprachiasmatic nucleus, central amygdala, and bed nucleus stria terminalis. Interestingly, this presynaptic expression pattern overlaps remarkably well with the expression pattern for corticotrophin releasing factor (CRF) (Pilcher & Joseph, 1984). CRF and the related urocortin peptides mediate behavioral, cognitive, autonomic, neuroendocrine and immunologic responses to aversive stimuli by activating CRF receptor 1 or CRF receptor 2 in the central nervous system and

112 mAKAP in the mouse brain anterior pituitary. The paralleled distribution of mAKAP with the CRF system suggests that mAKAP may also be involved in a variety of functions such as emotion, behaviour, memory, and autonomic responses. For example, it has been shown that cardiac function is critically regulated by the autonomic nervous system to adapt to the physical activity and emotional stress and mAKAP was reported to play a very important role in heart physiology. However, it remains to be determined whether there is a crosstalk between CRF and mAKAP in the brain and in the heart.

Besides its localization in the limbic system at similar sites as CRF, there are several additional lines of evidence indicating that mAKAP could be involved in cognition. Long- term changes in synaptic strength are thought to be the cellular basis underlying learning and memory processes. These changes are mediated through alterations in gene expression via the activation of transcription factors such as NFAT and cAMP responsive element binding protein (CREB) (Wang et al., 2006). Like in the heart, calcineurin, which is anchored to mAKAP, mediates NFAT translocation to the nucleus and this is critically dependent on increased intracellular calcium levels in neurons (Graeff et al., 1999). Furthermore, several effectors of the mAKAP complex including intracellular calcium release and the ERK pathway are known to affect PKA-induced CREB phosphorylation and CREB-dependent transcription (Zanassi et al., 2001). Thus one could speculate that mAKAP is important in coordinating gene expression at the nuclear membrane which is essential for e.g. long-term memory storage. We observed reduced levels of mAKAP in aged mice. In this respect it is interesting to note that cAMP levels are also reduced over the course of aging leading to improper regulation of CREB activation (Sugawa & May, 1994).

Our immunohistochemical analysis revealed that mAKAP is expressed both in cell bodies (e.g.: pyramidal neurons in the cortex and hippocampus, neurons in the striatum and amygdala) and in dendrites (e.g.: apical and basal dendrites of hippocampal pyramidal neurons). It was previously reported that mAKAP is targeted to the nuclear membrane of differentiated myocytes (Kapiloff et al., 1999). Moreover, immunohistochemical analysis revealed that AKAP100 (mAKAP) was localized to the sarcoplasmic reticulum in both cardiac and skeletal cell lines (McCartney et al., 1995; Yang et al., 1998). Smooth

113 Chapter 5 endoplasmic reticulum is the best characterized Ca2+ signaling organelle and represents a heterogeneous endomembrane system that extends from the neuronal soma to most cell compartments, including axons and dendritic spines (Meldolesi, 2001). In the mouse brain mAKAP is also detected in dendrites. This could be due to the fact that mAKAP is attached to the endoplasmic reticulum in this subcellular compartment.

In the heart, mAKAP recruits PKA to the cardiac ryanodine receptor (RyR) via an interaction with a leucine zipper-like motif (Marx et al., 2001). This channel is found on both the sarcoplasmatic reticulum and the nuclear envelope in cardiac and skeletal muscle cells and mediates calcium-induced calcium release from intracellular stores (Bootman et al., 2000). In cardiomyocytes, mAKAP facilitates the PKA-induced increase in channel conductance by linking this kinase to the RyR (Kapiloff et al., 2001, Ruehr et al., 2003). Moreover, localization of PKA by mAKAP at RyR1 increases both PKA-dependent RyR phosphorylation as well as efflux of Ca2+ through the RyR (Ruehr et al., 2003). RyR are also present in dendritic spines in the brain as indicated by immunohistochemical localization studies (Martone et al., 1997; Sharp et al., 1993) and by Ca2+ imaging (Emptage et al., 1999; Raymond & Redman, 2006). Although is difficult to speculate on the role of mAKAP in the brain on the basis of its distribution, it may very well be that mAKAP is involved in Ca2+ store-mediated signaling in the brain similar to its function in the heart. For example, a target of the Ca2+ store-mediated events could be CaMKII, which is known to associate with PSD following LTP induction (Shen et al., 2000). Overall, the widespread distribution of mAKAP suggests an important role for this protein in multiple brain functions. This hypothesis is further strengthened by the fact that all the molecules so far identified as members of the mAKAP macromolecular complex such as PKA, PDE, Epac, calcineurin and MAPK play a crucial role in multiple brain processes and mAKAP may act as an important coordinator of these signaling pathways. Moreover, the age-dependent decrease in mAKAP expression may influence the proper coordination of signal transduction pathways and thereby contribute to changes in brain function during ageing.

114 mAKAP in the mouse brain

Acknowledgments

We thank Martijn Clausen and Marcia Peters for valuable technical support.

References

Bauman, A.L., Michel, J.J., Henson, E., Dodge-Kafka, K.L., & Kapiloff, M.S. (2007). The mAKAP signalosome and cardiac myocyte hypertrophy. IUBMB Life 59, 163-169.

Bootman, M.D., Thomas, D., Tovey, S.C., Berridge, M.J., & Lipp, P. (2000). Nuclear calcium signaling. Cellular and Molecular Life Sciences 57, 371-378.

Emptage, N., Bliss, T.V., & Fine, A. (1999). Single synaptic events evoke NMDA receptor-mediated release of calcium from internal stores in hippocampal dendritic spines. Neuron 22, 115-124.

Ghavami, A., Hirst, W.D., & Novak, T.J. (2006). Selective phosphodiesterase (PDE)-4 inhibitors: a novel approach to treating memory deficit? Drugs R D 7, 63-71.

Graeff, I.A., Mermelstein, P.G., Stankunas, K., Neilson, J.R., Deisseroth, K., Tsien, R.W., & Crabtree, G.R. (1999). L-type calcium channels and GSK-3 regulate the activity of NF-ATc4 in hippocampal neurons. Nature 401, 703-708.

115 Chapter 5

Martone, M.E., Alba, S.A., Edelman, V.M., Airey, J.A., & Ellisman, M.H. (1997). Distribution of inositol-1,4,5-trisphosphate and ryanodine receptors in rat neostriatum. Brain Research 756, 9-21.

Marx, S.O., Reiken, S., Hisamatsu, Y., Gaburjakova, M., Gaburjakova, J., Yang, Y.M., Rosemblit, N., & Marks, A.R. (2001). Phosphorylation-dependent regulation of ryanodine receptors: a novel role for leucine/isoleucine zippers. The Journal of Cell Biology 153, 699-708.

Meldolesi, J. (2001). Rapidly exchanging Ca2+ stores in neurons: molecular, structural and functional properties. Progress in Neurobiology 65, 309-338.

Michel, J.J., Townley, I.K., Dodge-Kafka, K.L., Zhang, F., Kapiloff, M.S., & Scott, J.D. (2005). Spatial restriction of PDK1 activation cascades by anchoring to mAKAPalpha. Molecular Cell 20, 661-672.

Pilcher, W.H., & Joseph, S.A. (1984). Co-localization of CRF-ir perikarya and ACTH-ir fibers in rat brain. Brain Research 299, 91-102.

Raymond, C.R., & Redman, S.J. (2006). Spatial segregation of neuronal calcium signals encodes different forms of LTP in rat hippocampus. Journal of Physiology 570, 97-111.

Sharp, A.H., McPherson, P.S., Dawson, T.M., Aoki, C., Campbell, K.P., & Snyder, S.H. (1993). Differential immunohistochemical localization of inositol 1,4,5-trisphosphate- and ryanodine- sensitive Ca2+ release channels in rat brain. Journal of Neuroscience 13, 3051-3063.

116 mAKAP in the mouse brain

Shen, K., Teruel, M.N., Connor, J.H., Shenolikar, S., & Meyer, T. (2000). Molecular memory by reversible translocation of calcium/calmodulin-dependent protein kinase II. Nature Neuroscience 3, 881-886.

Sugawa, M., & May, T. (1994). Signal transduction in aging. Archives of Gerontology and Geriatrics 19, 235-246.

Wang, H., Hu, Y., & Tsien, J.Z. (2006). Molecular and systems mechanisms of memory consolidation and storage. Progress in Neurobiology 79, 123-135.

Yang, J., Drazba, J.A., Ferguson, D.G., & Bond, M. (1998). A-kinase anchoring protein 100 (AKAP100) is localized in multiple subcellular compartments in the adult rat heart. The Journal of Cell Biology 142, 511-522.

Zanassi, P., Paolillo, M., Feliciello, A., Avvedimento, E.V., Gallo, V., & Schinelli, S. (2001). cAMP- dependent protein kinase induces cAMP-response element-binding protein phosphorylation via an intracellular calcium release/ERK-dependent pathway in striatal neurons. Journal of Biological Chemistry 276, 11487-11495.

117

Chapter 6

Exchange protein activated by cyclic AMP 2 (Epac2) plays a specific and time-limited role in memory retrieval

Anghelus Ostroveanu1, Eddy A. van der Zee1, Ulrich L.M. Eisel1, Martina Schmidt2, Ingrid M. Nijholt1

1 Department of Molecular Neurobiology, 2 Department of Molecular Pharmacology, Graduate School of Behavioral and Cognitive Neurosciences, University of Groningen, The Netherlands

Submitted: Hippocampus Chapter 6

Abstract

Knowledge on the molecular mechanisms involved in memory retrieval is limited due to the lack of tools to study this stage of the memory process. Here we report that exchange proteins activated by cAMP (Epac) play a surprisingly specific role in memory retrieval. Intrahippocampal injection of the Epac activator 8-pCPT-2’O-Me-cAMP was shown to improve fear memory retrieval in contextual fear conditioning whereas acquisition and consolidation were not affected. The retrieval enhancing effect of the Epac activator was even more prominent in the passive avoidance paradigm. Downregulation of Epac2 expression in the hippocampal CA1 area impaired fear memory retrieval when the memory test was performed 72 h after training, but not when tested after 17 days. Our data thus identify an important time-limited role for hippocampal Epac2 signaling in cognition and opens new avenues to investigate the molecular mechanisms underlying memory retrieval.

Key words: fear, conditioning, learning, cognition, hippocampus, mouse

120 Epac2 in memory retrieval

Introduction

To date significant advances have been made in understanding the neurophysiological basis of learning and memory. In particular, cyclic adenosine monophosphate (cAMP) signaling was shown to play a pivotal role. Originally cAMP-dependent protein kinase (PKA) was thought to be the major if not the sole effector of cAMP and its importance in memory consolidation is now widely acknowledged (Abel & Nguyen, 2008). However, fairly recently, a new effector of cAMP signaling has been identified named exchange protein directly activated by cAMP (Epac). In independent studies, two variants of the Epac protein, namely Epac1 (also called cAMP-GEF-I) and Epac2 (also called cAMP-GEF-II), were characterized (de Rooij et al., 1998; Kawasaki et al., 1998). Both Epac proteins are multi-domain proteins that function as guanine-nucleotide-exchange factors (GEFs) for Rap1 and Rap2, members of the Ras superfamily of small GTPases. Activation of Epac by cAMP leads to activation of Rap1 and Rap2, which then act as molecular switches on downstream signaling cascades. While Epac1 has one cAMP binding domain, Epac2 possesses a similar additional domain, the biological function of which is still unknown (Bos, 2006). The two Epac variants also differ in their expression patterns. Epac1 has been found to be expressed ubiquitously, whereas expression of Epac2 was found mainly in adrenal glands and brain tissue (Kawasaki et al., 1998). Since their discovery, Epac proteins have been found to control key cellular processes, including cellular calcium handling, integrin-mediated cell adhesion, gene expression, cardiac hypertrophy, inflammation, and exocytosis (Roscioni et al., 2008). However, the exact nature of any involvement that Epacs have in neuronal function, has only recently begun to be investigated. Epac was shown to enhance neurotransmitter release in glutamatergic synapses (Sakaba & Neher, 2003; Zhong & Zucker, 2005; Gekel & Neher, 2008), whereas in cerebellar granule cells it can modulate neuronal excitability (Ster et al., 2007). In dorsal root ganglion Epac mediates the translocation and activation of protein kinase C (PKC) leading to the establishment of inflammatory pain (Hucho et al., 2005) and promotes neurite outgrowth (Murray & Shewan, 2008). In spinal cord tissue Epac advances neurite regeneration (Murray & Shewan, 2008).

121 Chapter 6

Thus far, evidence for a role of Epac in the process of learning and memory is limited. However, since Epac is a cAMP-responsive enzyme and cAMP signaling is established to be of critical importance in learning and memory, an involvement of PKA-independent cAMP signaling through Epac proteins can be expected. Indeed the first indications for a role of Epac in hippocampus-dependent learning and memory came from very recent studies. Gelinas and colleagues reported that Epac activation enhances the maintenance of LTP in area CA1 of mouse hippocampal slices (Gelinas et al., 2008) and co-application of a selective PKA and a selective Epac activator was shown to rescue the memory retrieval impairment observed in dopamine-beta-hydroxylase deficient mice whereas application of the Epac activator alone had no effect (Ouyang et al., 2008). In the current study we investigated the role of Epac signaling in the different phases of the memory process; acquisition, consolidation and retrieval. Epac signaling via Epac2 was shown to play a specific and time-limited role in memory retrieval.

Materials and methods

Animals and housing conditions Male C57BL/6J mice (Harlan, Horst, the Netherlands), 9 to 12 weeks old, were individually housed in standard macrolon cages. Subjects were maintained on a 12 hour light/dark cycle (lights on at 7.30 a.m.) with food (hopefarm® standard rodent pellets) and water ad libitum. A layer of sawdust served as bedding. The procedures concerning animal care and treatment were in accordance with the regulations of the ethical committee for the use of experimental animals of the University of Groningen (DEC 4174I-K).

Cannulation Double guide cannulae (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 (Nijholt et al., 2008). The cannulae were placed into both dorsal hippocampi (intrahippocampal; i.h.), AP –1.5 mm, lateral 1 mm, depth 2 mm (Franklin & Paxinos, 1997). The animals were allowed to recover for 6-7 d before the experiments started.

122 Epac2 in memory retrieval

Bilateral injections were performed during a short anesthetic period of isoflurane inhalation using a syringe pump (TSE systems, Bad Homburg, Germany) at a constant rate of 0.33 µl/min (final volume: 0.3 µl per side). The exact site of injection was confirmed after the behavioral experiments by injection of methylene blue solution into each hemisphere and subsequent histological evaluation (Fig. 5A). Data were evaluated only from those mice that received an injection at the correct target site.

Drug treatment The Epac activator 8-pCPT-2’O-Me-cAMP (Biolog, Bremen, Germany) was injected in a final concentration of 1 mM (300 ng/brain) in artificial cerebrospinal fluid (ACSF) solution of the following composition (in mM): 130 NaCl, 3.5 KCl, 1.25 NaH2PO4, 1.5 MgSO4, 2

CaCl2, 24 NaHCO3, and 10 glucose (pH 7.4). 8-pCPT-2’O-Me-cAMP was stored as a 100 mM stock solution in H2O. A separate set of animals was injected with vehicle (ACSF pH=7.4). Untreated animals without cannula served as controls for possible cannulation and injection effects.

Fear Conditioning Fear conditioning was performed as described before (Nijholt et al., 2008) in a Plexiglas cage (44 x 22 x 44 cm) with constant illumination (12 V, 10 W halogen lamp, 100-500 lux). The training (conditioning) consisted of a single trial. The mouse was exposed to the conditioning context for 180 sec followed by a footshock (0.7 mA, 2 sec, constant current) delivered through a stainless steel grid floor. The mouse was removed from the fear conditioning box 30 sec after shock termination to avoid an aversive association with the handling procedure. Memory tests were performed 24 hr, 72 h or 14 days after fear conditioning. Contextual memory was tested in the fear conditioning box for 180 sec without footshock presentation. 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 by a time-sampling procedure every 10 s throughout memory tests. In addition, mean activity of the animal during the training and retention test was measured with the Ethovision system (Noldus, The Netherlands).

123 Chapter 6

Passive avoidance Passive avoidance experiments were performed in a plexiglas cage (44 x 22 x 44 cm) consisting of a dark compartment (22 x 22 x 20 cm) equipped with a stainless steel grid floor and a light compartment (22 x 22 x 44 cm) with a plastic floor. Both compartments were separated by a guillotine door. The light compartment was brightly illuminated by a 100 W bulb. Mice were habituated to the experimental set-up during three sessions 30, 24 and 6 hr prior to the training session. During habituation sessions, the mouse was introduced into the light compartment facing the closed guillotine door. After 60 sec the door was opened and the mouse was allowed to enter the dark compartment. Upon entering the dark compartment the door was closed and the mouse was allowed to explore the compartment for 60 sec. During the training session, the mouse was again introduced into the light compartment, and the guillotine door was opened after 60 sec. Latency (defined as the time between the opening of the door and the mouse entering the dark compartment with all four paws) was recorded for each animal. Upon entering the dark chamber the door was closed and a single footshock (0.3 mA, 2 sec, constant current) was delivered to the mouse. The mouse was removed from the apparatus 30 sec after shock termination to avoid an aversive association with the handling procedure. Memory tests were performed 24 hr after training. During the memory test the guillotine door was opened 60 sec after introducing the mouse into the light compartment and left opened for maximally 480 sec. During this time period, latency to enter the dark compartment was recorded and assessed as the behavioral parameter. If a mouse did not enter the dark compartment, latency was set to 480 sec.

Elevated plus maze Elevated plus maze experiments were performed using a plus maze (50 cm above the floor) with two opposite closed and two opposite open arms (50 cm long, 5 cm wide). The mouse was placed in the central zone of the plus maze, facing an open arm and allowed to explore the maze for 480 sec. Time spent in dark arms, open arms and center compartment were recorded for each animal with the Ethovision system (Noldus, The Netherlands). The ratio of time spent in the open arms to total time spent in the maze was calculated for each group

124 Epac2 in memory retrieval of mice and taken as a measure of anxiety-related behaviour, with a higher ratio being indicative of lower anxiety levels.

Immunohistochemistry 30 µm thick coronal sections of C57Bl/6J mice perfused with 4 % paraformaldehyde, were preincubated with 0.3 % H2O2 to reduce endogenous peroxidase. Non-specific binding sites were blocked by preincubating the sections with 5 % normal goat serum in 0.01 M PBS for 30 min. Subsequently, sections were probed with antibodies specific for Epac1 (from 1:300 to 1:1000, several batches [1C8, 4D9 and 5D3] kindly provided by J. Bos, University Utrecht, the Netherlands and Epac1 A5, sc-28360, Santa Cruz) or Epac2 (1:1000) (2B12, provided by J. Bos) in 0.01 M PBS containing 5 % normal goat serum and 0.3 % Triton X-100 for 2 h at room temperature (RT) and subsequently for 72 h at 4 °C. After several washing steps, sections were incubated with biotin SP-Conjugated AffiniPure goat anti mouse secondary antibody overnight at 4 °C (1:400) (115-065-166, Jackson Laboratories INC) followed by the ABC complex (Vector ABC kit). For visualization, DAB was used as a chromogen (Sigma fast tablet set). Sections were examined using light microscopy. The specificity of Epac antibodies was assessed by parallel staining without primary antibodies. In these sections we could not observe any staining (data not shown). Photographs were taken with a DM1000/DFC280 Leica image analysis system (Leica, Cambridge, UK).

Semi-quantitative RT-PCR To determine Epac1 and Epac2 mRNA levels in the mouse hippocampus, total RNA was extracted from a single hippocampus of a naive mouse (n=5). Total RNA was isolated according to the manufacturer’s protocol (NucleoSpin RNA II kit, Macherey-Nagel, 740955.250). RT-PCR was performed using Superscript III One-Step RT-PCR with Platinum Taq DNA Polymerase (Invitrogen, 12574) as described before (Nijholt et al., 2004). 120 ng of RNA was used for each RT-PCR reaction. The reverse transcriptase reaction was performed at 55 °C for 30 min. PCR cycling was at 94 °C for 15 sec, annealing at 55 °C for 30 sec, extension at 72 °C for 30 sec and a final extension at 72 °C

125 Chapter 6 for 10 min. 10 µl of each sample was removed every 3 cycles from 25 to 37 cycles in each reaction to amplify Epac1 and Epac2. To test the efficiency of the siRNA probes, hippocampi were collected 24 hr after the last siRNA injection and the injection site excised. Total RNA isolation and the reverse transcriptase reaction were performed as described above. Hypoxanthine-guanine phosphoribosyltransferase (HPRT) served as control housekeeping gene. 10 µl of each sample was removed every 3 cycles from 27 to 50 cycles in each reaction to amplify Epac1, Epac2 and HPRT fragments. Amplified PCR products were separated on 2 % agarose gels with Tris-borate EDTA buffer and stained with SYBR Green (Invitrogen). Gels were captured as a digital image and quantified by densitometry. Primer sequences for Epac1 were: forward 5’- GTTGTCGACCCACAGGAAGT-3’ and reverse 5’-ACCCAGTACTGCAGCTCGTT-3’, for Epac 2 were: forward 5’-CATGAGGGGAACAAGACGTT-3’ and reverse 5’ GGCC- TTCGAGGCTCTAATCT 3’ and for HPRT forward primer 5’-CCTGCTGGATTACATT- AAAGCACTG-3’ and reverse 5’-CCTGAAGTACTCATTATAGTCAAGG-3’.

In vivo siRNA transfection Mice injected i.h. with 50 ng siGLO Green (25 ng/hippocampus; D-001630-01-05, Dharmacon, Inc. Lafayette, CO, USA), were sacrificed 6, 24 or 48 hr post injection. The brain hemispheres were placed in a 4 % PFA solution for 24 hr, followed by 48 hr 30 % sucrose immersion. Afterwards, 30 µm thick coronal sections were stained with DAPI (1:5000) in PBS 0.01 M. After a quick washing step in PBS 0.01 M, sections were mounted, dried and analyzed under a Leica fluorescent microscope. ON-TARGET plus SMART pool mouse RAPGEF3 (Epac1 siRNA) and RAPGEF4 (Epac2 siRNA) probes were purchased from Dharmacon, (Dharmacon, USA). The target sequences for the mouse-specific Epac1 siRNAs mixture were as follows: sense: CCAGGCAGGAACCGGUAUAUU (J-057800-09); sense: GAUCUUUGUUCACGGCC- AAUU (057800-10); sense: GGUCAAUUCUGCCGGUGAUUU (057800-11) and sense: CCACCAUCAUCCUUCGAGAUU (057800-12). The target sequences for the mouse- specific Epac2 siRNAs mixture were: sense: CGAAAGACCUGGCGUACCAUU (J- 057784-05); sense: CAAGUUAGCUCUAGUGAACUU (J-057784-06); sense: GACAGA-

126 Epac2 in memory retrieval

AAGUACCACCUAAUU (J-057784-07) and sense: GGAGGAACUGUGUUGUUUAUU. ON-TARGETplus Non-targeting Poll siRNA (D-001810-10) was used as control (Dharmacon, USA). siRNAs were resuspended in RNAse free water. In vivo siRNA brain delivery was performed using jetSI 10 mM cationic polymer transfection reagent (Polyplus transfection Inc., New York) according to the transfection protocol of the manufacturer. 50 ng siRNA was injected i.h. on three consecutive days 3, 24 and 48 hr after training or on the three days prior to the second retention test.

Statistics Statistical comparisons were made by analysis of variance (ANOVA). For each significant F ratio, Fisher's protected least significant difference (PLSD) test was used to analyze the statistical significance of appropriate multiple comparisons. Data were expressed as mean ± s.e.m. Significance was determined at the level of p < 0.05.

Results

Intrahippocampal Epac activation facilitates memory retrieval in contextual fear conditioning The role of Epac in the different stages of the memory process was investigated using one trial contextual fear conditioning. Contextual fear conditioning is a hippocampus-dependent form of associative learning in which animals learn to fear a new environment because of its temporal association with an aversive mild electrical footshock. When injected intrahippocampally (i.h.) 20 min before training, the specific Epac activator 8-pCPT-2’O- Me-cAMP caused no significant change in freezing behavior during the retention test 24 hr after training in comparison to vehicle-injected and untreated mice (one-way ANOVA: F(2,25) = 1.110; p = 0.312, Fig. 1A). Injection of 8-pCPT-2’O-Me-cAMP or vehicle did not result in changes in mean activity or shock reactivity during training (data not shown). Moreover, no significant difference in freezing behavior was observed between groups during the retention test when 8-pCPT-2’O-Me-cAMP was injected immediately after training (one-way ANOVA: F(2,18) = 0.032; p = 0.969, Fig. 1B).

127 Chapter 6

To determine the effect of Epac activation on the retrieval of fear memory, mice were injected i.h. with 8-pCPT-2’O-Me-cAMP or vehicle 20 min before the retention test. Injection of 8-pCPT-2’O-Me-cAMP resulted in a significant increase in freezing behavior during the retention when compared to vehicle-injected and untreated animals (one-way ANOVA: F(2,24) = 5.550, p = 0.010; Fig. 1C).

ABC Injection Injection Injection Training 2” 24hr retention Training 2” 24hr retention Training 2” 24hr retention

180” 30” 180” 180” 30” 180” 180” 30” 180” * 80 80 80

(%)

(%) 60 (%) 60 60

40 40 40 20 20 20

Freezing

Freezing 0 0 0

ehicle ehicle ehicle V V V Untreated Untreated Untreated

-2’O-Me-cAMP -2’O-Me-cAMP -2’O-Me-cAMP

8-pCPT 8-pCPT 8-pCPT

Fig. 1. Intrahippocampal injection of Epac activator 8-pCPT-2’O-Me-cAMP (1 mM) facilitates the retrieval of contextual fear memory. Mice were injected either 20 min before training (A, immediately after training (B), or 20 min before retention (C) with 8-pCPT-2’O-Me-cAMP (1 mM) or vehicle. Untreated and vehicle-injected mice served as controls. Freezing behaviour was measured in the memory test 24 h after training. Error bars indicate standard error of the mean. Statistically significant differences: *p < 0.05 versus control groups.

Taken together, these data show that Epac activation in the hippocampus modulated the retrieval of contextual fear memory, but not acquisition or consolidation.

Intrahippocampal Epac activation facilitates memory retrieval in passive avoidance The effect of i.h. 8-pCPT-2’O-Me-cAMP injection on memory retrieval was also tested in the passive avoidance task. In this one trial fear-motivated avoidance task the animal learns to refrain from stepping through a door to an apparently safer but previously punished dark compartment. It is considered to be more complex than fear conditioning due to the combination of classical Pavlovian conditioning with the manifestation of an active response. Mice were habituated to the experimental set-up during three sessions prior to the

128 Epac2 in memory retrieval training session. We did not observe any difference between groups in their latencies to enter the dark compartment during the training session (one-way ANOVA: F(2,23) = 0.917, p = 0.414, Fig. 2A). The next day, one group of mice was injected i.h. with 8-pCPT-2’O- Me-cAMP (1 mM) 20 min before the retention test. Untreated and vehicle injected mice served as controls. Mice injected with 8-pCPT-2’O-Me-cAMP showed a significantly longer latency to enter the dark compartment when compared to the control groups (one- way ANOVA: F(2,23) = 4.650, p = 0.020, Fig. 2B). Overall, the memory retrieval enhancing effect of 8-pCPT-2’O-Me-cAMP in the passive avoidance paradigm was even more prominent than in fear conditioning.

A B

* 400 400

(sec)

(sec) 300 300

200 200

training

retention

100 100

Latency 0 Latency 0

ehicle ehicle V V Untreated Untreated -2’O-Me-cAMP -2’O-Me-cAMP

8-pCPT 8-pCPT Fig. 2. Intrahippocampal injection of the Epac activator 8-pCPT-2’O-Me-cAMP (1 mM) facilitates memory retrieval in the passive avoidance paradigm. Mice were habituated to the experimental set-up during three sessions. Mice were injected with 8-pCPT-2’O-Me-cAMP or vehicle 20 min before the retention test. Untreated and vehicle-injected mice served as controls. Latency to enter the dark compartment during training (A) and the retention test (B). Error bars indicate standard error of the mean. Statistically significant differences: *p < 0.05 versus control groups.

Intrahippocampal Epac activation does not affect anxiety The performance of the mice in the retention tests may be influenced by the level of anxiety the animal experiences. Therefore, we tested the effect of 8-pCPT-2’O-Me-cAMP on anxiety behavior in an elevated plus maze. Intrahippocampal 8-pCPT-2’O-Me-cAMP (1 mM) injection 20 min before exposure to the elevated plus maze test did not specifically affect anxiety behavior (one-way ANOVA: F(2,19) = 1.741; p = 0.202, Fig 3). Cannulated

129 Chapter 6 animals, i.e. 8-pCPT-2’O-Me-cAMP-injected and vehicle-injected mice, did show slightly, but not significantly higher levels of anxiety, which can be explained by the surgery procedure these animals underwent 6-7 days prior to testing in the elevated plus maze.

Fig. 3. Intrahippocampal injection of 8-pCPT-2’O-Me-cAMP (1

%) ( 80 mM) does not affect anxiety. Mice were injected with 8-pCPT-

time 60 2’O-Me-cAMP or vehicle 20 min before the test. Untreated and

otal vehicle-injected mice served as controls. Time spent in the 40 different compartments of the maze was measured during 480

arms/T

20 sec and ratio between time in open arms and total time in maze

open

in was taken as a measure of anxiety. Error bars indicate standard 0

i error of the mean.

ehicle V Untreated

-2’O-Me-cAMP

8-pCPT Since injection of 8-pCPT-2’O-Me-cAMP did not affect anxiety in the elevated plus maze test, the effect of Epac activation by 8-pCPT-2’O-Me-cAMP in the fear-motivated learning tasks can be ascribed to enhanced memory retrieval of the association between the electric footshock and the context.

Epac expression in the mouse brain Next we determined the distribution of Epac1 and Epac2 in the mouse brain. Epac2 was shown to be abundantly expressed throughout the entire mouse brain (Fig. 4A). High levels were detected in the cortex, hippocampus and thalamus. In the hippocampal cellular layers such as the stratum pyramidale (Py) and the granule layer (GCL) immunoreactivity was rather low whereas the basal and apical dendrites [stratum oriens (SO), stratum radiatum (SR), the stratum lacunosum moleculare (SLM) and molecular layer (ML)] showed a high Epac2 expression (Fig. 4B). For Epac1 staining, four different Epac1 antibodies (up to antibody saturation levels) were tested. Although positive Epac1 staining was observed with these antibodies in lung and heart tissue (M. Schmidt, unpublished data), no positive Epac1 staining could be detected in the brain (data not shown). Thus, it appears that Epac1 expression is very low in mouse brain. Moreover, semi-quantitative RT-PCR for

130 Epac2 in memory retrieval

Epac1 and Epac2 with mRNA isolated from the hippocampus, showed that Epac2 mRNA could be detected much earlier as Epac1 mRNA (Fig. 4C).

A Epac1 Fig. 4. Detection of Epac1 and Epac2 in mouse brain. A.

Epac2 Semi-quantitative RT-PCR experiment for Epac1 and Epac2.

Cycle 25 28 31 34 37 Epac2 mRNA could be detected after a lower amount of cycles as Epac1 mRNA. A representative experiment is B presented. B. Epac2 staining in the mouse brain: caudate

H putamen (Cpu), hippocampus (H), ventrolateral thalamic nucleus (VL), anterior hypothalamic area (AH). C. Epac2 VL Cpu Immunoreactivity in the hippocampus: stratum pyramidale (Py), stratum oriens (SO), stratum radiatum (SR), stratum AH lacunosum moleculare (SLM) and molecular layer (ML) granule layer (GCL), hilus (hil). C SO Py SR SLM

ML hil GCL

Our data are consistent with a previous study from Kawasaki and colleagues who also reported a high expression of Epac2 in the rat brain whereas Epac1 was barely detectable (Kawasaki et al., 1998).

Intrahippocampal Epac2 siRNA injection impairs fear memory retrieval To investigate the role of hippocampal Epac2 in memory retrieval, we specifically downregulated Epac2 expression before the memory test using in vivo lipid mediated siRNA gene silencing. A previous study already showed the efficient downregulation of Epac2 expression by these siRNA probes in in vitro neuronal cell cultures (Nijholt et al., 2008). To check for siRNA transfection efficiency in the in vivo mouse brain, we first injected mice i.h. with fluorescent siGLO green (Fig. 5A,B). A single bilateral injection of siGLO green resulted in a strong fluorescent signal in the pyramidal cell layer of the CA1

131 Chapter 6 area already as early as 6 hr after injection. The signal lasted at least up to 48 h after injection. Other brain areas were not affected by the treatment. Downregulation of Epac2 expression by i.h. injection of specific siRNA probes on three consecutive days was verified by semi-quantitative RT-PCR on the fourth day. Injection of Epac2 siRNA resulted in a 47 % reduction of hippocampal Epac2 mRNA (Fig 5C,D). The low level of Epac1 mRNA was not affected by the transfection with Epac2 siRNA.

A C

100

(%)

T 80 *

HPR 60

/ 40 20

Epac2 B 0

DAPI SiGLO Green Merge siRNA siRNA Untreated Epac2 Scrambled D

HPRT

Epac2 Epac1

Fig. 5. Efficient downregulation of hippocampal Epac2 expression by in vivo siRNA transfection. A. left panel: coronal brain section of the mouse brain atlas (Franklin and Paxinos, 1997), black dot indicates the injection site, right panel: photomicrograph showing a representative detail of a coronal brain section after hippocampal injection of methylene blue counterstained with nuclear fast red. DG, dentate gyrus; CA1, CA1 area of the hippocampus; CA3, CA3 area of the hippocampus. B. Representative fluorescent microphotographs showing siGLO green transfection into the pyramidal neurons of the CA1 area indicated by an arrow. A DAPI staining was used to identify nuclear staining. C. Bar graphs show the ratio of Epac2 mRNA band intensities verified to be within the linear range of product accumulation, divided by those of the coamplified HPRT product after 34 cycles. Statistically significant difference: *p < 0.05 versus control groups. D. Bands reflect the levels of Epac1 and Epac2 mRNA expression after 34 cycles.

In the behavioral experiments, mice were injected i.h. with Epac2 siRNA (50 ng/brain) 3 hr, 24 hr and 48 hr after training in a contextual fear conditioning paradigm (Fig. 6A). Epac2 siRNA injection completely abolished the 8-pCPT-2’O-Me-cAMP-induced enhancement of retrieval and already caused a significant decrease in freezing behavior by

132 Epac2 in memory retrieval itself during the first retention test (one-way ANOVA: F(4,37) = 9.187; p = 0.001, Fig. 6B). In the scrambled siRNA injected animals 8-pCPT-2’O-Me-cAMP injection again improved memory retrieval. Interestingly, when the Epac2 siRNA injected animals were re-exposed to the conditioning box 14 days after the first retention test, they showed high freezing levels that were comparable to untreated or control siRNA injected mice (one-way ANOVA: F(4,35) = 0.862; p = 0.496, Fig. 6C). Injection of Epac2 siRNA on three consecutive days prior to this delayed retention test also did not affect freezing in any of the groups (Fig. 6C).

A Training Retention 1 Retention 2

Day 1 234...... 14 15 16 17

siRNA injection1 siRNA injection2

8-pCPT-2’O-Me-cAMP injection

B Retention 1 C Retention 2 100 * 80 80

60 *

(%) 60

(%) *

40 40

Freezing

Freezing 20 20

0 0

siRNA Untreated Epac2 Scrambled siRNA Untreated Epac2 Scrambled injection1 siRNA siRNA injection1 siRNA siRNA

8-pCPT-2’- - - + - + siRNA injection2 - - + - + O-Me-cAMP (Epac2 siRNA)

Fig. 6. Intrahippocampal injection of Epac2 siRNA impairs memory retrieval in contextual fear conditioning. Mice were injected with Epac2 or control siRNA 3 hr, 24 hr and 48 hr after training. Untreated mice served as additional controls. A. Experimental protocol B. Freezing behaviour was measured in the memory test 24 h after the last siRNA injection. C. Freezing behavior assessed during the second retention test 14 days after the first retention test. Where indicated this retention test was preceded by Epac2 siRNA injections on three consecutive days. Error bars represent standard error of the mean. Statistically significant differences: *p < 0.05 versus control groups.

133 Chapter 6

Discussion

In contrast to memory formation, the knowledge about the molecular mechanisms of memory retrieval is surprisingly limited due to the lack of tools to study this phase of the memory process. Most studies on memory using brain lesion and/or gene manipulation techniques cannot distinguish between effects on the molecular mechanisms of acquisition or consolidation of memories and those responsible for their retrieval from storage. Using the specific Epac activator 8-pCPT-2’O-Me-cAMP, we observed a surprisingly specific role of Epac signaling in associative fear memory retrieval whereas acquisition and consolidation were not affected. Ouyang and colleagues also recently reported a role for Epac signaling in memory retrieval (Ouyang et al., 2008). However, their design did not allow the investigation of Epac signaling in the different phases of the memory process. In their study, the memory retrieval impairment observed in dopamine-beta-hydroxylase deficient mice could be rescued by i.h. injection of a selective PKA activator together with a selective Epac activator whereas injection of one of the activators alone did not overcome the retrieval deficit. From these data they concluded that cAMP signaling via both Epac and PKA is required for retrieval (Ouyang et al., 2008). We report here that Epac activation alone can significantly improve memory retrieval in contextual fear conditioning. This retrieval- enhancing effect was even stronger in a passive avoidance paradigm.

Since 8-pCPT-2’O-Me-cAMP activates both Epac1 and Epac2 (Enserink et al., 2002), it was not possible to distinguish between the contribution of both Epac variants to the facilitation of memory retrieval. However, the finding in our and other studies that Epac2 is abundantly expressed in mouse brain whereas Epac1 is hardly detectable (Kawasaki et al., 1998) together with our finding that downregulation of Epac2 expression in the hippocampus impairs memory retrieval, strongly suggests a role for Epac2 in memory retrieval. Interestingly, Epac2 silencing only led to impaired memory retrieval 3 days after conditioning whereas Epac2 silencing during the retention test 17 days after conditioning had no effect on memory retrieval indicating a time-limited function of Epac2 signaling after conditioning. These data are consistent with earlier data showing that signaling by

134 Epac2 in memory retrieval norepinephrine through the beta1-adrenergic receptor is also only required for an intermediate term of memory retrieval (Murchison et al., 2004). Since beta1-adrenergic receptors couple to cAMP via Gs, it is likely that this result is at least in part mediated by Epac2. In line with these findings activation of beta adrenergic receptors and 8-pCPT-2’O- Me-cAMP were shown to recruit similar mechanisms to facilitate long-lasting hippocampal LTP (Gelinas et al., 2008). Also in several other cell systems such as the heart and vascular smooth muscle cells, a strong connection between beta-adrenergic signaling and Epac has already been established (Jensen, 2007; Métrich et al., 2008). From our finding that freezing was low in the Epac2 siRNA injected animals 3 days after conditioning but comparable to untreated and control siRNA injected animals 17 days after conditioning it can be concluded that Epac2 silencing in the hippocampus transiently affects memory retrieval instead of having a long-lasting effect on consolidation. Overall our data strengthen the hypothesis that retrieval may become independent of the hippocampus over time (McClelland et al., 1995; Squire et al., 2001; Wiltgen et al., 2004; Morris, 2006). However we cannot completely exclude the possibility that hippocampal signaling mechanisms other than Epac2 are involved in the delayed memory retrieval.

The subcellular mechanism by which Epac2 modulates the retrieval of fear memory still remains to be elucidated. Although little information is available on downstream molecules of Epac signaling in the hippocampus, Epacs are known to function as cAMP-mediated guanine nucleotide-exchange factors (GEFs) activating the small GTP-ase proteins Rap1 and Rap2 (Bos, 2006). Indeed, Ouyang and colleagues reported on unpublished data that the expression of a dominant-negative Rap construct in the dorsal hippocampus impairs memory retrieval in a manner identical to antagonists of beta1-adrenergic receptors, cAMP and PKA (Ouyang et al., 2008). Epac-Rap signaling has been reported to activate p42/p44 mitogen-activated protein kinases (p42/p44 MAPK; ERK1/2) in cultured rat hippocampal neurons (Lin et al., 2003). Moreover, application of 8-pCPT-2’O-Me-cAMP leads to a transient increase in p42/44 MAPK immunoreactivity in hippocampal slices of the CA1 area (Gelinas et al., 2008). Overall, MAPKs may have an important contribution in the memory enhancing effect of the Epac activator. Indeed, MAPK activation was observed to be increased in the hippocampus during memory retrieval (Szapiro et al., 2000), whereas

135 Chapter 6

MAPK inhibition by intrahippocampal injection of the MAPK kinase inhibitor PD098059 was shown to impair retrieval of an one-trial step-down avoidance task (Izquierdo et al., 2000). Impaired retrieval is generally a sensitive measure of memory impairment in age-associated memory impairment (AAMI) and the early stages of Alzheimer’s disease. From our data it can be speculated that enhancing Epac2 signaling might at least in part overcome the memory retrieval deficits reported. In this respect it is interesting to note that the Epac2 expression is reduced in brains showing Alzheimer’s pathology when compared to non- diseased control brains. These changes were restricted to those regions of the brain associated with Alzheimer’s disease such as the frontal cortex and the hippocampus but not in the cerebellum, a region resistant to this pathology (McPhee et al., 2005). On the contrary, post-traumatic stress disorder (PTSD) is characterized by traumatic memories that can manifest as daytime recollections, traumatic nightmares, or flashbacks in which components of the event are relived. These symptoms reflect excessive retrieval of traumatic memories that often retain their vividness and power to evoke distress for decades or even a lifetime. It can be hypothesized that such conditions may benefit from reduced Epac2 signaling.

Considering both the lack and the need of drugs proven to be effective in modulating memory retrieval, the specific effect of hippocampal Epac signaling on retrieval we observed is of particular interest and warrants further research into the role of Epac signaling in cognitive processes under physiological and pathological conditions.

Acknowledgements

We thank Wouter Scheper, Janne Papma and Martijn Clausen for excellent technical support. We would like to acknowledge J. Bos, University of Utrecht for kindly providing the Epac1 and Epac2 antibodies. U.E. is supported by the European Union’s FP6 funding, NeuroproMiSe, LSHM-CT-2005-018637. This work reflects only the author’s views. The European Community is not liable for any use that may be made of the information herein.

136 Epac2 in memory retrieval

References

Abel, T., & Nguyen, P.V. (2008). Regulation of hippocampus-dependent memory by cyclic AMP- dependent protein kinase. Progress in Brain Research 169, 97-115.

Bos J.L. (2006). Epac proteins: multi-purpose cAMP targets. Trends in Biochemical Sciences 31, 680- 686.

de Rooij J., Zwartkruis, F.J., Verheijen, M.H., Cool, R.H., Nijman, S.M., Wittinghofer, A., & Bos J.L. (1998). Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 396, 474-477.

Enserink, J.M., Christensen, A.E., de Rooij, J., van Triest, M., Schwede, F., Genieser, H.G., Doskeland, S.O., Blank, J.L., & Bos, J.L. (2002). A novel Epac-specific cAMP analogue demonstrates independent regulation of Rap1 and ERK. Nature Cell Biology 4, 901-906.

Franklin, K.B.J., & Paxinos, G. (1997). The mouse brain in stereotaxic coordinates. Academic Press, San Diego.

Gekel, I., & Neher, E. (2008). Application of an Epac activator enhances neurotransmitter release at excitatory central synapses. Journal of Neuroscience 28, 7991-8002.

Hucho, T.B., Dina, O.A., & Levine, J.D. (2005). Epac mediates a cAMP-to-PKC signaling in inflammatory pain: an isolectin B4(+) neuron-specific mechanism. Journal of Neuroscience 25, 6119- 6126.

Izquierdo, L.A., Barros, D.M., Ardenghi, P.G., Pereira, P., Rodrigues, C., Choi, H., Medina, J.H., & Izquierdo, I. (2000). Different hippocampal molecular requirements for short- and long-term retrieval of one-trial avoidance learning. Behavioral Brain Research 111, 93-98.

Jensen, J. (2007). More PKA independent beta-adrenergic signaling via cAMP: is Rap1-mediated glucose uptake in vascular smooth cells physiologically important? British Journal of Pharmacology 151, 423-425.

137 Chapter 6

Lin, S.L., Johnson-Farley, N.N., Lubinsky, D.R., & Cowen, D.S. (2003). Coupling of neuronal 5-HT7 receptors to activation of extracellular-regulated kinase through a protein kinase A-independent pathway that can utilize Epac. Journal of Neurochemistry 87, 1076-1085.

McClelland, J.L., McNaughton, B.L., & O’Reilly, R.C. (1995). Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. Psychological Review 102, 419-457.

McPhee, I., Gibson, L.C., Kewney, J., Darroch, C., Stevens, P.A., Spinks, D., Cooreman, A., & MacKenzie, S.J. (2005) Cyclic nucleotide signaling: a molecular approach to drug discovery for Alzheimer's disease. Biochemical Society Transactions 33, 1330-1332.

Métrich, M., Lucas, A., Gastineau, M., Samuel, J.L., Heymes, C., Morel, E., & Lezoualc’h, F. (2008). Epac mediates beta-adrenergic receptor-induced cardiomyocyte hypertrophy. Circulation Research 102, 959-965.

Morris, R.G. (2006). Elements of a neurobiological theory of hippocampal function: The role of synaptic plasticity, synaptic tagging and schemas. European Journal of Neuroscience 23, 2829-2846.

Murchison, C.F., Zhang, X.Y., Zhang, W.P., Ouyang, M., Lee, A., & Thomas, S.A. (2004). A distinct role for norepinephrine in memory retrieval. Cell 117, 131-143.

Murray, A.J., & Shewan, D.A. (2008). Epac mediates cyclic AMP-dependent axon growth, guidance and regeneration. Molecular and Cellular Neuroscience 38, 578-588.

Nijholt, I.M., Dolga, A.M., Ostroveanu, A., Luiten, P.G., Schmidt, M., & Eisel, U.L. (2008). Neuronal AKAP150 coordinates PKA and Epac mediated PKB/AKT phosphorylation. Cellular Signaling 10, 1715-1724

Nijholt, I.M., Farchi, N., Kye, M., Sklan, E.H., Shoham, S., Verbeure, B., Owen, D., Hochner, B., Spiess, J., Soreq, H., & Blank, T. (2004). Stress-induced alternative splicing of acetylcholinesterase results in enhanced fear memory and long-term potentiation. Molecular Psychiatry 9, 174-183.

138 Epac2 in memory retrieval

Nijholt, I.M., Ostroveanu, A., Scheper, W.A., Penke, B., Luiten, P.G., Van der Zee, E.A., & Eisel, U.L. (2008). Inhibition of PKA anchoring to A-kinase anchoring proteins impairs consolidation and facilitates extinction of contextual fear memories. Neurobiology of Learning and Memory 87, 693- 696.

Roscioni, S.S., Elzinga, C.R., & Schmidt, M. (2008). Epac: effectors and biological functions. Naunyn-Schmiedeberg's Archives of Pharmacology 377, 345-357.

Sakaba, T., & Neher, E. (2003). Direct modulation of synaptic vesicle priming by GABA(B) receptor activation at a glutamatergic synapse. Nature 424, 775-778.

Squire, L.R., Clark, R.E., & Knowlton, B.J. (2001). Retrograde amnesia. Hippocampus 11, 50-55.

Szapiro, G., Izquierdo, L.A., Alonso, M., Barros, D., Paratcha, G., Ardenghi, P., Pereira, P., Medina, J.H., & Izquierdo, I. (2000). Participation of hippocampal metabotropic glutamate receptors, protein kinase A and mitogen-activated protein kinases in memory retrieval. Neuroscience 99, 1-5.

Wiltgen, B.J., Brown, R.A., Talton, L.E., & Silva, A.J. (2004). New circuits for old memories: The role of the neocortex in consolidation. Neuron 44, 101-108.

139

Chapter 7

General Discussion Chapter 7

Content

1. The role of A-kinase anchoring protein 150 in learning and memory 2. PKA anchoring to AKAPs in learning and memory 3. Tools to investigate the function of AKAP(s) signalosome 4. mAKAP in the mouse brain 5. AKAPs as new therapeutic targets 6. The role of Epac in learning and memory 7. Overall conclusion and future perspectives 8. Reference

142 General discussion

1. The role of A-kinase anchoring protein 150 (AKAP150) in learning and memory

Recently, A-kinase anchoring proteins (AKAP) emerged as prototypic coordinators of cyclic adenosine monophosphate (cAMP) activated protein kinase (PKA) signaling providing high specificity to PKA signaling. The most investigated member of this family of proteins is AKAP79/150, a protein expressed in e.g. rodent (AKAP150) and human (AKAP79) brain. We showed that AKAP150 is highly expressed in the mouse striatum, cerebral cortex, and forebrain regions, while a relatively high expression was found in the hippocampus and olfactory bulb of the mouse brain. Low/no expression was observed in the cerebellum, hypothalamus, thalamus, and brain stem (Ostroveanu et al., 2007). When the AKAP150 staining in the mouse brain is compared to the previously reported AKAP150 immunoreactivity pattern in rats (Glantz et al., 1992, Lilly et al., 2005), it becomes obvious that the AKAP150 distribution in rat brain is much more homogeneous than in the mouse brain. For example, in mice the CA3 area of the hippocampus is very densely stained compared to the remainder of the hippocampus (Ostroveanu et al., 2007), while the rat hippocampus shows a homogenous staining. Overall, the expression patterns of AKAP150 in the mouse and rat brain reveal an abundant presence in brain regions involved in learning and memory such as the cortex and hippocampus. Although it is difficult to speculate on the role of AKAP150 in the mouse brain on the basis of its distribution, these data suggest that AKAP150 may well play a role in learning and memory.

In a study focusing on the distribution of AKAP150 at rat CA1 pyramidal cell postsynaptic densities (PSD), it was shown that AKAP150 interacts with components of the excitatory PSD, whereas AKAP150 immunoreactivity (IR) was not associated with inhibitory synapses (Lilly et al., 2005). In parallel to these findings, AKAP79/150 was identified as a constituent of postsynaptic densities (PSD) of excitatory synapses in human cerebral cortex (Carr et al., 1992). Moreover, AKAP79/150 is linked to the N- methyl D-aspartate receptor (NMDAR) and α-amino-3-hydroxyl-5-methyl-4-isoxazole-propionate receptor (AMPAR)

143 Chapter 7 in the PSD through binding with postsynaptic density-95, discs large, zona occludens-1 (PDZ) domain, membrane-associated guanylate kinase (MAGUK), scaffold proteins postsynaptic density-95 (PSD-95) and synapse-associated protein-97 (SAP97) (Colledge et al., 2000). In addition to PKA, AKAP79/150 can also bind protein phosphatase 2B (PP2B) and protein kinase C (PKC) (Carr et al., 1992b; Coghlan et al., 1995; Klauck et al., 1996; Sik et al., 2000). It was found that this multi-enzyme signaling complex plays an important role in coordinating changes in synaptic structure and receptor signaling functions underlying synaptic plasticity (Dell'Acqua et al., 2006). For example, AKAP79/150 bound PKA was found to be essential for the regulation of AMPA receptor surface expression and synaptic plasticity (Rosenmund et al., 1994). Additional evidence that AKAP79/150 could be implicated in synaptic transmission arises from the study of Genin and colleagues (2003) who showed that AKAP150 mRNA is upregulated during the maintenance phase of long- term potentiation (LTP), a mechanism of long-lasting enhancement in synaptic efficacy (Genin et al., 2003). A very recent study by Lu et al. (2007) also suggested that AKAP150 can be involved in synaptic plasticity. In transgenic mice expressing a truncated form of AKAP150 which cannot anchor PKA, hippocampal LTP was abolished in 7–12 but not 4- week-old mice. From these data they concluded that PKA anchored to AKAP150 critically contributes to LTP in the adult hippocampus (Lu et al, 2007). Changes in synaptic plasticity are suggested to be the mechanism underlying learning and memory processes (Martin et al., 2000; Micheau and Riedel 1999). The findings that AKAP79/150 plays an important role in synaptic plasticity, thus, provide additional evidence for the hypothesis that AKAP79/150 is important in learning and memory processes. However, knowledge on the role of AKAP79/150 in learning and memory remained so far rather limited.

Initial evidence for a role of AKAPs and PKA anchoring in learning and memory came from a study by Moita and colleagues. They blocked PKA anchoring in the lateral amygdala of rats and subjected the animals to auditory fear conditioning (Moita et al., 2002). Their data showed that inhibition of PKA binding to AKAPs in the rat lateral amygdala impairs memory consolidation during auditory fear conditioning (Moita et al., 2002). We provided the first direct evidence for a role of AKAP150 in learning and memory processes by assessing the expression of AKAP150 in the mouse hippocampus

144 General discussion after a single training session in a contextual fear conditioning paradigm. We could show for the first time that AKAP150 is upregulated in the mouse hippocampus 6 hours after training in this fear conditioning test (Nijholt et al., 2007). The time point of 6 h correspond to the late phase of memory consolidation and is critically regulated by cAMP signaling pathways in fear motivated associative learning. Moreover, mice exposed to a novel context also showed an AKAP150 upregulation in the mouse hippocampus (Nijholt et al., 2007). Although we can only speculate on the possible role of the increased expression of AKAP150, it might very well be that elevated AKAP150 levels result in a more efficient propagation of signals carried by locally generated cAMP (Colledge et al., 2000; Feliciello, et al., 1997), which in turn may contribute to processing the exposure to a novel context and the consolidation of associative memory. Since AKAP150 levels were only measured 6 h after training we cannot exclude that AKAP150 may also be important during other stages of the memory process. In general, the upregulation of AKAP150 may be the result of de novo protein synthesis or decreased protein degradation.

Recently several studies provided data on possible mechanism of how AKAP150 could modulate learning and memory processes. It was shown that constitutive loss of AKAP150 in mice modifies excitatory synaptic transmission (Tunquist et al., 2008). This is achieved by a deranged localization of PKA in AKAP150 null hippocampal neurons which in turn modulates the postsynaptic AMPA receptors. These AKAP150 KO mice also exhibit deficits in motor coordination and show memory retrieval impairment in the Morris water maze (Tunquist et al., 2008). Hall and colleagues (2007) showed that AKAP150 copurifies 2+ with L-type Ca channels Cav1.2 in the rat forebrain. Cav1.2 constitutes 80% of the L-type channels in the brain (Hell et al, 1993, Moosmang et al., 2005) and is concentrated at postsynaptic sites (Hell et al., 1996). It has been shown that hippocampal pyramidal neurons express predominantly the Cav1.2 channel (Hell et al., 1993, Davare et al., 2001,

Sinnegger-Brauns et al., 2004). Interestingly, conditional inactivation of the Cav1.2 gene in the mouse hippocampus and neocortex leads to a selective loss of protein synthesis- dependent NMDAR-independent Schaffer collateral/CA1 late-phase LTP and these mice show a severe impairment in a water-maze spatial-discrimination task (Moosmang et al,

2005). Although it was shown that Cav1.2 binds several AKAPs in the brain (MAP2B

145 Chapter 7

(Davare et al., 1999, Hall et al., 2007), AKAP15 (Hall et al., 2007)), AKAP150 proved to be critical for PKA-mediated regulation of this channel in the brain (Hall et al., 2007). Overall, these data suggest that AKAP150 could be important in learning and memory via 2+ the regulation of Ca signaling through Cav1.2. In addition, Chai et al. found evidence that AKAP150 co-immunoprecipitates with the acid- sensing ion channels ASIC1a and ASIC2a (Chai et al., 2007). It has been shown that these channels are critically regulated by AKAP150 and calcineurin/PP2B (Chai et al., 2007). Interestingly, ASIC1a is abundantly expressed in the amygdala complex and other brain regions known to be involved in fear memory, and has been implicated in LTP (Wemmie et al., 2002, Chai et al., 2007). Moreover, ASIC1 null mice display deficits in cued and contextual fear conditioning (Wemmie et al., 2003), whereas ASIC1a overexpression enhances fear-related freezing behavior (Wemmie et al., 2004). Furthermore, inhibition of PKA anchoring by Ht-31 induces a decrease in ASIC current amplitude in cultured mouse cortical neurons and Chinese hamster ovary (CHO) cells (Chai et al., 2007). Altogether these studies suggest that AKAP150 may be able to change memory performance via the regulation of ASIC channels.

Recently, several other AKAPs were reported to be important contributors to learning and memory processes. For example, it has been shown that AKAP Yu is involved in the formation of long term memory in Pavlovian olfactory learning of Drosophila (Lu et al., 2007) and WAVE-1 null-mice (the first AKAP null mouse model) are impaired in spatial (Morris water maze) and non-spatial (novel object recognition) memory formation but not in emotional learning and memory (passive avoidance) (Soderling et al., 2003).

In summary, these data provide evidence for an increasing number of AKAPs that are involved in learning and memory in different species. Our data, together with future findings will lead to the understanding of how AKAPs coordinate the action of signaling networks in cognitive processes under physiological and pathological conditions. Ultimately this knowledge on the coordination of signaling cascades may lead to the development of novel, more fine-tuned, innovative therapeutic strategies to treat cognitive dysfunction.

146 General discussion

2. PKA anchoring to AKAPs in learning and memory

Besides the role of particular AKAPs in learning and memory, it is at least as important to have a good understanding of the role of enzyme binding to these AKAPs. We addressed the question whether anchored PKA is critically involved in the different stages (acquisition, consolidation, retrieval, and extinction) of contextual fear memory in mice. Accordingly, we first injected mice intracerebroventricularly with St-Ht31 (a membrane permeable peptide that competes for PKA anchoring) at different time points during the memory process. Injecting mice before training with St-Ht31 to affect the acquisition phase resulted in impaired associative fear memory 24 h after the training. Similarly, injection of St-Ht31 immediately after training (consolidation phase) significantly attenuated conditioned fear. The learning deficit observed when St-Ht31 was injected immediately after training was similar to the effect of St-Ht31 injected before training. To discriminate between acquisition and consolidation, we also performed the retention test 1 h after the training with mice that were injected before the training. Interestingly, the performance of St-Ht31 injected animals did not differ from the control groups when the retention test was performed 1 h after training. The finding that mice which received St-Ht31 before training, showed unimpaired freezing 1 h after training but attenuated freezing 24 h after training, suggests that PKA anchoring onto AKAPs plays a specific role in the consolidation of contextual fear memories but not in acquisition (Nijholt et al., 2008). Moreover, disrupting PKA anchoring before the retention test had no effect on memory performance in contextual fear conditioning (Nijholt et al., 2008). Next, we tested the sub-region specific contribution of PKA anchoring in the hippocampus to the consolidation phase of fear memory. Therefore we injected immediately after training St-Ht31 locally in the CA1 area of the mouse hippocampus. During the retention test mice showed reduced freezing behavior when compared to vehicle or non-injected animals, which indicates that disrupting PKA anchoring locally in the CA1 is instrumental in the consolidation of fear memory. Our finding that PKA anchoring is important in learning is supported by the findings of Nie and colleagues (Nie et al., 2007). Genetically modified mice conditionally expressing Ht31 in the forebrain regions show impairments in the spatial version of the Morris water maze, a hippocampus-dependent memory task (Nie et al, 2007).

147 Chapter 7

Although these data showed that PKA anchoring is crucial for memory consolidation, they did not provide information on whether type I and/or type II PKA is involved. However, most of the AKAPs in the brain tether PKA type II. To date, no RI AKAP was described in the brain and only a few dual RI/RII AKAPs were characterized (Huang et al, 1997). Since we cannot completely rule out a role for RI anchoring, we injected stearated Super-AKAP- IS, in the mouse CA1 area of hippocampus before training mice in contextual fear conditioning. Super-AKAP-IS is a PKA anchoring inhibitory peptide that selectively binds RIIβ subunits of PKA. This selective anchoring inhibitory peptide also impaired the freezing behavior when injected immediately after training. Thus, RIIβ anchoring to AKAPs in the CA1 of the mouse hippocampus is crucial and sufficient for the consolidation of contextual fear memory.

Recently Isiegas and colleagues showed that neuronal PKA facilitates the extinction of contextual fear (Isiegas et al., 2006). This was found in transgenic mice expressing from birth a dominant-negative form of PKA and in mice in which the same PKA dominant- negative form was temporarily regulated in brain regions thought to be involved in extinction (Isiegas et al., 2006). In our lab, we assessed for the first time the role of PKA anchoring in fear extinction. Mice were injected in the CA1 area of the hippocampus with stearated Super-AKAP-IS immediately after each extinction trial of contextual fear memory. We observed that inhibiting PKA anchoring in the hippocampus promotes the extinction of contextual fear conditioning (Nijholt et al., 2008). We thus were the first to show that disrupting RIIβ anchoring only in the hippocampus is sufficient to facilitate fear extinction. Our data are supported by the finding that temporally transgenic inhibition of PKA activity also promotes extinction (Isiegas et al., 2006). More evidence for a role of PKA anchoring in memory formation was provided by studies in Drosophila where AKAP- bound PKA is required for aversive memory during appetitive memory, a Pavlovian olfactory learning task in which an electric shock is used as an aversive unconditioned stimulus (Schwaerzel et al., 2007). Unfortunately, St-Ht31 and Super-AKAP-IS do not specifically block PKA anchoring to a particular AKAP, and therefore it cannot be concluded which AKAP(s) was involved in the impaired consolidation of contextual fear memory. We thus determined the amount of PKA anchored to AKAP150 in the dorsal

148 General discussion hippocampus after intrahippocampal injection of St-superAKAP-IS. St-superAKAP-IS was shown to reduce the amount of PKA anchored to AKAP150 in the dorsal hippocampus when compared to vehicle-injected mice. It was reported that PKA bound to AKAP150 is crucial for the phosphorylation of GluR1 AMPA receptor subunit at Ser845 (Colledge et al., 2000, Tavalin et al, 2002). This phosphorylation stimulates postsynaptic accumulation of GluR1-containing AMPARs during LTP (Esteban et al, 2003; Oh et al, 2006). Previously, it has been shown that treatment of hippocampal slices with membrane permeable Ht31 impairs PKA dependent hippocampal late-phase LTP (L-LTP) (Huang et al., 2006). Recently, Nie and colleagues showed in transgenic mice expressing Ht31 in the hippocampus that a disruption of PKA- AKAP interactions does not alter PKA activity or basal synaptic transmission but that only long lasting forms of hippocampal LTP are impaired (Nie et al, 2007). Interestingly, disruption of PKA anchoring or inhibition of PKA activity in hippocampal neurons leads to a PP2B dependent, long-term depression (LTD) -like downregulation of AMPAR currents and loss of AMPAR surface expression that likely involves AKAP79/150 (Hoshi et al., 2005; Rosenmund et al., 1994; Snyder et al., 2005; Tavalin et al., 2002). Overall, these data suggest that AKAP79/150 regulates the balance between PKA and PP2B signaling to control AMPAR phosphorylation underlying LTP and LTD in synaptic plasticity. However, so far the dynamics of this process are poorly understood. NMDA treatments that induce LTD at many synapses simultaneously (chem-LTD) in hippocampal neurons activate PP2B/CaN signaling and disrupt the association of AKAP79/150 with PSD-95 and cadherins. This leads to a loss of the AKAP from spines and coincides with the removal of AMPA receptors from synapses (Gomez et al., 2002; Gorski et al., 2005). In hippocampal slices, chem-LTD activation of PP2B/CaN is followed by a persistent redistribution of both AKAP79/150 and PKA-RII from postsynaptic membranes to the cytoplasm, without PP2B/CaN translocation (Smith et al., 2006). Furthermore, using fluorescence resonance energy transfer microscopy in hippocampal neurons, Smith et al. showed that PKA anchoring to AKAP79/150 is required for an NMDA receptor-dependent regulated cytoplasmic translocation of PKA and AKAP79/150 (Smith et al., 2006). In summary, PKA anchoring in the hippocampus is required for both long-lasting forms of hippocampal synaptic plasticity, and learning and memory.

149 Chapter 7

3. Tools to investigate the function of AKAP(s) signalosome

Besides PKA, there are more AKAP binding partners that participate in learning and memory processes. For example, protein phosphatases are already widely acknowledged as key molecules in synaptic plasticity and learning and memory. However, the role of protein phosphatase anchoring to AKAPs in learning and memory has not been investigated yet. The most interesting way to explore the function of any AKAP signalosome is by pharmacological studies and/or by genetic manipulation. Although the use of pharmacological compounds that block PKA anchoring already yielded fundamental insights into the role of AKAP signaling in various physiological systems, inhibitory peptides for any other AKAP binding partners are still not available to date. This is most likely due to the difficulty of designing such agents and of intracellular delivery, and to the expected unspecific action of such peptides. In future experiments the impact of the RI isoform-selective anchoring on learning and memory processes could be assessed using the RI anchoring disruptor (RIAD) (Carlson et al., 2006). To study in greater detail which specific AKAP is involved, it would be necessary to develop inhibitors that disrupt the interaction of PKA with one particular AKAP or to disrupt the interaction of PKA by introducing site-specific mutations in the PKA binding domain of a specific AKAP. Recently, genetically modified mice were generated in the AKAP field. It is not surprising that the first complete knockdown of an AKAP was AKAP150 (Hall et al., 2007). In addition to the AKAP150 KO mouse, mice in which the PKA binding site was removed from the AKAP150 gene were also generated (Hall et al., 2007). However, these mice are still not perfect “tools” for investigating the role of AKAP150 since the modification occurs also during development and compensatory mechanisms cannot be overcome. To define a role for a specific AKAP in learning and memory, precise control on the expression of the transgene in space and time is required. Another advantage of this spatial inducible system, compared to constitutive deletion of a gene or expression of a transgene, is a reduced chance of animal lethality. This is especially relevant when the gene of interest plays a crucial role during development or encodes for a toxic protein. Already, the tetracycline-dependent system is widely used as a tool for spatio- temporal control of transgene expression. Using this system, Isiegas and coworkers

150 General discussion generated genetically modified mice that conditionally express Ht31 in forebrain regions. Behavioral studies with these Ht31 transgenic mice revealed impairments in spatial learning and memory in the Morris water maze as well impairments in long-lasting forms of hippocampal LTP (Isiegas et al., 2008).

In our lab we aimed to generate mice that inducibly express AKAP150 or AKAP150 mutants only in forebrain regions. In order to explore both the in vitro and in vivo relevance of AKAP150 binding to PKA and PP2B, particularly in learning and memory, a strategy was developed involving removal of the PKA and PP2B binding site from AKAP150. Initially, using a polymerase chain reaction (PCR) based technique we were able to clone the mouse AKAP150 protein. Later on, by comparing the mouse AKAP150 sequence with the human AKAP79 and rat AKAP150 homologues we could identify the PKA and PP2B binding site on mouse AKAP150 and generate several deletion mutants: the PKA binding site deletion mutant, the PP2B binding site deletion mutant, and the PKA+PP2B binding site deletion mutant (Fig. 1).

Fig. 1. Left panel. Schematic diagram of the AKAP150 molecule. The PKA and PP2B binding site are located near the C- terminal end of the protein. The N- terminal end of the protein contains a binding site for PKC and the membrane-targeting domain. Right panel. Several mutants were generated: upper - binding site-deleting mutant for PKA, middle - binding site-deleting mutant for PP2B, lower - binding site-deleting mutant for both PKA and PP2B

To generate mice with an AKAP150 overexpression or an AKAP150 mutant overexpression we opted for the use of lentiviral vectors to transduce embryonic stem cells. These stem cells can then be injected into a mouse blastocyst to create a chimerical transgenic animal. Later on, by cross breeding of chimeric transgenic animals, a pure transgenic animal can be obtained. These transgenic animals can then be used for functional testing to determine the consequences of their altered genetic material (Fig. 2).

151 Chapter 7

Moreover, by using the lentiviral expression system, various cell lines could be infected in addition to test the viability and function of AKAP150 and its mutants in vitro. We managed to successfully infect embryonic stem cells with AKAP150 or AKAP150 binding site deletion mutants (unpublished data). These results will allow us to move forward in creating a transgenic animal expressing inducible AKAP150 binding site deletion mutants in the near future. Unquestionably, these transgenic animals can be used in future experiments for a better understanding of the role of AKAP150 and anchoring of PKA and PP2B to AKAP150 in learning and memory.

GENE OF INTEREST

Lentiviral vector In vitro experiments Electrophysiology + Gene of interest Overexpress gene of interest by infecting the cell line of interest

Producing viral particle Select for cells expressing desired gene Infecting embryonic stem cells Virus producer cell line Inject transformed ES Transgenic animal cells into blastocyst

Behavioral testing Fig. 2. Flowchart of the lentiviral expression system. The genes of interested were cloned in lentiviral expression vectors. Viral particles containing the transgene can be use to infect various cell lines. This will allow for in vitro experiment to test the function of the wildtype and mutants. Viral particles can also be used to infect embryonic stem cells that can be used to generate transgenic animals. The physiological relevance of mutant AKAP150 in transgenic animals can be assessed by behavioral testing, in vitro experiments, and electrophysiology.

4. mAKAP in the mouse brain

In the last few years, the number of AKAPs that were identified and characterized increased considerably. One of these, mAKAP (m of muscular), was found to be predominantly

152 General discussion expressed in cardiac and skeletal muscle, but also detected to some extent in the brain (Kapiloff et al., 1999). Two alternatively spliced variants of mAKAP were subsequently characterized: mAKAPα (preferentially expressed in the brain) and mAKAPβ (abundant in cardiac myocytes and skeletal muscle). In addition to their tissue-specific expression, mAKAPα and mAKAPβ considerably differ in size: the longer form mAKAPα contains an additional 244 amino acid residue N-terminal extension (Michel et al., 2005). Detailed knowledge on the localization of mAKAP in the brain remained so far limited. We showed that mAKAP is abundantly expressed throughout the entire mouse brain (Ostroveanu et al., 2009 submitted). Both immunohistochemistry and Western blotting revealed a high expression of mAKAP in the cortex, cerebellum, thalamus, hypothalamus, and hippocampus, while the lowest mAKAP expression was detected in the brain stem. Moreover, at subcellular levels, mAKAP was abundantly expressed both in perikarya (e.g: pyramidal neurons, Purkinje neurons), dendrites (e.g.: apical and basal dendrites of hippocampus pyramidal neurons) and to some extent in fibers (e.g: amygdala, cortex). Our results are in good agreement with the reported mRNA expression of AKAP100/mAKAP in several human brain regions (Kapiloff et. al., 1999; McCartney et al., 1995). The function of mAKAP in the brain still remains rather unexplored. Conversely, in the heart the function of mAKAP was extensively studied. In the heart, mAKAP is targeted to the nuclear envelope through the binding of three spectrin repeat domains, where it forms a large macromolecular signaling complex containing several signal transduction molecules, including PKA (Kapiloff et al, 1999, Kapiloff et al, 2001), the phosphodiesterase PDE4D3 (Dodge et al., 2001), the protein phosphatases PP2A and calcineurin (Kapiloff et al, 2001, Pare et al., 2005), ryanodine receptors (RyR2) (Kapiloff et al, 2001, Marx et al., 2000, Ruehr et al., 2003), the small GTPase Rap1, the guanine exchange factor Epac1 (Dodge- Kafka et al., 2005), and the mitogen-activated protein kinases (MAPK) MEK5 and ERK5 (Dodge-Kafka et al., 2005). Both in the brain and heart, mAKAP can also bind 3’- phosphoinositide-dependent kinase 1 (PDK1) and p90 ribosomal S6 kinase (RSK3) (Michel et al., 2005). Although in the heart 12 mAKAPβ signalosome components were reported, there are probably many more mAKAPβ binding partners. The mAKAPα signalosome in the brain has not been characterized yet. It can be expected that mAKAPα and mAKAPβ form more or less the same signalosome. Preliminary data

153 Chapter 7 from our laboratory show that mAKAPα co-immunoprecipitates with PDE4D3 and the exchange protein directly activated by cAMP 2 (Epac2) from hippocampal lysates (unpublished data). Whether Epac1 is also included in the mAKAPα macromolecular complex, still needs to be determined. In addition, preliminary colocalization studies in primary cortical neurons confirmed that mAKAP co-localizes with Epac2 and PKA-RIIβ (Fig. 3; unpublished data).

mAKAP Epac2 Merge

mAKAP PKA-RIIβ Merge

Fig. 3. mAKAP colocalizes with Epac2 and PKA-RIIβ in primary cortical neurons. Primary cortical neurons from E14 were probed with antibodies against mAKAP, Epac2 and PKA-RIIβ.

Since at the nuclear membrane of cardiomyocytes the mAKAP complex coordinates three different cAMP effectors (PKA, PDE4D3 and Epac1), and also calcium and MAPK kinase signalling pathways, it is likely that in the brain it has the same function. Since mAKAP forms such a large signalosome, it is currently unclear whether there is a single, large complex that contains all of the mAKAP binding partners or, alternatively, whether there are different mAKAP signalosomes that contain varying combinations of the signaling molecules. We detected the highest mAKAP expression presynaptically in brain regions such as the central amygdaloid nucleus, supraoptic nucleus, paraventricular hypothalamic

154 General discussion nucleus, and suprachiasmatic nucleus (Ostroveanu et al., 2009 submitted). Our finding that mAKAP is abundantly expressed, particularly presynaptically, suggests an important role for mAKAP in coordinating signal transduction pathways involved in adequate brain functioning. This presynaptic expression pattern overlaps remarkably well with the expression pattern for corticotrophin releasing factor (CRF) (Pilcher & Joseph, 1984). The paralleled distribution of mAKAP with the CRF system suggests that mAKAP may also be involved in a variety of functions such as emotion and autonomic responses but also memory processes. Furthermore, we found a dramatic decrease in mAKAP expression in all brain regions of aged mice. It has been shown that the expression levels of many genes related to neuronal signaling, plasticity, and structure are changed in the hypothalamus and cortex of the aged mouse brain (Jiang et al., 2001). Several proteins are downregulated in the aged brain. For example, the level of PKC β1 is significantly decreased in aged rats as assessed by RT-PCR and Western blotting (Yao et al, 1998). Moreover, in brain tissue from elderly patients diagnosed with mild cognitive impairments several synaptic proteins (e.g. synaptophysin) in the frontal and temporal cortex were shown to be downregulated (Counts et al., 2006). Although it is difficult to speculate on the role of mAKAP in the brain on the basis of its distribution, it may very well be that mAKAP contributes to changes in brain functioning in aged mice and that it plays a role in cognitive processes. A growing body of work on the role of AKAPs in the brain and other tissues provides evidence that these multifunctional scaffolding proteins facilitate the fidelity of cAMP signaling. This tight control of cAMP signaling is clearly important in the maintenance of a healthy state, whereas the loss of this regulation might initiate diseases. How mAKAP mediates the crosstalk and the integration of different signaling pathways via its multimolecular protein complexes and what function it serves in the brain remains, however, to be determined.

5. AKAPs as new therapeutic targets

Several studies have implicated the AKAP signalosome in ion channel modulation, G- protein-coupled receptor desensitization, vesicular secretion, actin cytoskeletal dynamics,

155 Chapter 7 cell division and gene transcription regulated by cAMP, Ca2+, and lipid second messengers (Michel & Scott, 2002). Since many of these cellular processes are altered in human disease, understanding the mechanisms of AKAP signaling pathways may aid in the development of novel therapeutics. So far traditional treatment in cognitive disorders has focused on the manipulation of cell surface protein function. However, pharmacological manipulation of cAMP signaling cascades in space and time may ultimately give us more control on where and when events occur within the cell. Through careful intervention in the coordination of signaling cascades we can gently nudge a cell into a certain direction. By targeting anchoring proteins, like AKAPs, we could achieve this subtle intervention.

6. The role of Epac in learning and memory

Fairly recently, a novel PKA-independent cAMP effector was discovered and named exchange protein directly activated by cAMP (Epac). In independent studies, two variants of the Epac protein, Epac1 (also called cAMP-GEF-I) and Epac2 (also called cAMP-GEF- II), were characterized (de Rooij et al., 1998; Kawasaki et al., 1998). Both Epacs function as guanine nucleotide-exchange factors (GEFs) that specifically activate Rap GTPases (Rap1 and Rap2) upon binding to cAMP. Although Epac proteins have been found to control key cellular processes, including cellular calcium handling, integrin-mediated cell adhesion, gene expression, cardiac hypertrophy, inflammation, and exocytosis (Pereira et al., 2007; Oestreich et al., 2007; Hucho et al., 2005; Rangarajan et al., 2003; Lotfi et al., 2006; Kang et al., 2003; Morel et al., 2005), their role in the brain just started to be investigated. Since cAMP signaling is established to be of critical importance in learning and memory, a potential role for PKA-independent cAMP signaling through Epac proteins in the process of learning and memory may be expected. Recently, Epac has been linked to synaptic transmission and neuronal excitability (Zhong & Zucker, 2005), was shown to modulate membrane potentials in cultured cerebellar neurons, thereby potentially regulating cellular excitability (Ster et al., 2007), and to enhance neurotransmitter release at excitatory synapses (Gekel & Neher, 2008). Very recently it has been shown that Epac is involved in both LTP (Gelinas et al, 2008) and LTD (Ster et al., 2009) in the mouse hippocampus.

156 General discussion

Since both changes in LTP and LTD that affect synaptic transmission in the hippocampus are thought to be involved in memory formation, these data suggest furthermore the involvement of Epac in learning and memory processes.

We showed that intrahippocampal delivery of 8-pCPT-2’O-Me-cAMP, a compound that specifically activates Epac but not PKA, improves fear memory retrieval in contextual fear conditioning whereas acquisition and consolidation were not affected (Ostroveanu et al., 2009 submitted). The retrieval enhancing effect of the Epac activator was even more prominent in the passive avoidance test, a behavioral paradigm based on associative emotional learning similar to fear conditioning. Since the performance of the mice in the retention tests may be influenced by the level of anxiety the animal experiences, we tested the effect of 8-pCPT-2’O-Me-cAMP on anxiety behavior in an elevated plus maze. Intrahippocampal 8-pCPT-2’O-Me-cAMP injection before the elevated plus maze had no effect on anxiety. Therefore, the effect of Epac activation in the fear-motivated learning tasks can be solely ascribed to enhanced memory retrieval of the association between the electric footshock and the context (Ostroveanu et al., 2009 submitted). Ouyang and colleagues also recently reported a role for Epac signaling in memory retrieval (Ouyang et al., 2008). In their study, the memory retrieval impairment observed in dopamine-beta- hydroxylase deficient mice could be rescued by intrahippocampal (i.h.) injection of a selective PKA activator together with a selective Epac activator, whereas injection of one of the activators alone did not overcome the retrieval deficit (Ouyang et al., 2008). Interestingly, when cAMP agonists were infused into the dorsal hippocampus of wild-type mice 30 min before testing retrieval, no significant effects on retrieval were observed when the agonists were infused separately or in combination (Ouyang et al., 2008). The lack of any effect on memory retrieval in the wild type mice is in contradiction with our findings and may depend on the time of agonist delivery. Ouyang and colleagues performed the injections 30 minutes before the retrieval, whereas in our experiments, Epac agonist was injected into the CA1 are of the hippocampus 10 minutes before the memory test.

It has been shown that 8-pCPT-2’O-Me-cAMP activates both Epac1 and Epac2. Therefore, the enhanced memory retrieval effect cannot be attributed to a particular Epac. We showed

157 Chapter 7 that Epac2 is abundantly expressed throughout the whole mouse brain, whereas only low levels of Epac1 were detected. Using semi-quantitative RT-PCR for Epac1 and Epac2 with mRNA isolated from the hippocampus, Epac2 mRNA could be detected much earlier as Epac1 mRNA, pointing towards much higher Epac2 than Epac1 mRNA levels in the hippocampus (Ostroveanu et al., 2009 submitted). Our data are consistent with a previous study from Kawasaki et al. who also reported a high expression of Epac2 in the rat brain whereas Epac1 was barely detectable (Kawasaki et al., 1998). To date, no Epac-specific inhibitors are available. Therefore, we established a protocol for in vivo lipid mediated siRNA gene silencing in the mouse brain to investigate the role of Epac2 in learning and memory. Local Epac2 silencing in the CA1 area of the hippocampus led to impaired memory retrieval 3 days after conditioning whereas Epac2 silencing during the retention test 17 days after conditioning had no effect on memory retrieval indicating a time-limited function of Epac2 signaling after conditioning. The finding that Epac2 is abundantly expressed in the mouse brain together with the finding that Epac2 silencing impairs memory retrieval, suggests an important role for Epac2 in learning and memory. In our experiments we could not completely exclude a role for Epac1 in these processes. Both specific-Epac inhibitors and/or transgenic or knock-out mice will provide further evidence on the role of Epac1 and Epac2 in cognitive processes.

Interestingly, expression levels of both Epac1 and Epac2 were reported to be altered in the frontal cortex and hippocampal regions of brains showing Alzheimer’s pathology (McPhee et al., 2005). Alzheimer disease (AD) is a neurodegenerative disease characterized by the formation of β-amyloid plaques and neurofibrillary tangles. It has been shown that amyloid-beta protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory (Shankar et al., 2008). Moreover, in AD, neural degradation occurs primarily in neurons involved in memory storage and retrieval (Arshavsky, 2006). So far the relationship between changes in Epac expression and Alzheimer’s pathology remain unclear, but it could very well be that Epac plays a role in the cognitive decline such as observed in AD. This is also supported by the finding that the Epac2 levels are decreased in AD. We have shown in our studies that activation of Epac in the mouse hippocampus significantly and specifically improves the retrieval of fear memory. Considering both the

158 General discussion lack and the need of drugs proven to be effective in improving memory retrieval, the specific facilitating effect of Epac activation on retrieval we observed, is of particular interest and deserves further research into the role of Epac signaling in cognitive processes under physiological and pathological conditions.

Recent findings in several physiological systems acknowledged Epac as coordinator of specific cAMP signals independent of PKA. Interestingly, it appears that in part, signals carried through Epac are spatially and temporally compartmentalized. For example, in the heart mAKAP is able to target Epac1 (Dodge-Kafka et al., 2005), while in the brain AKAP150 binds both Epac1 and Epac2 (Nijholt et al, 2008). Thus, the spatio-temporal integration of cAMP pathways via Epac, PKA, AKAPs and PDEs dramatically increases the complexity and, consequently, the possible readouts of cAMP signalling. Moreover, since both Epac and PKA are ubiquitously expressed, an increase in cAMP levels will lead to the activation of both effectors which in turn can execute independent responses or develop a cross-talk effect. For example, it has been reported that PKA and Epac can exert opposing effects on the regulation of the PKB/AKT pathway. While PKA suppresses PKB phosphorylation and activity, activation of Epac leads to increased PKB phosphorylation (Mei et al., 2002; Nijholt et al., 2008). Interestingly, we recently showed that PKA and Epac-mediated PKB/Akt phosphorylation is coordinated in neurons by AKAP150 (Nijholt et al., 2008). In addition to the opposing effects, PKA and Epac may also work synergistically in other systems. For example both the stimulation of neurotensin (Li et al., 2007) and the attenuation of cAMP signaling through phosphodiesterases depend on the activation of Epac and PKA (Dodge-Kafka et al., 2005).

Future research will lead to a better understanding of the physiological roles of both Epac isoforms. Eventually, this may lead to a more detailed knowledge of the distinct role that individual Epacs play in shaping cAMP signalling in the mammalian brain. This in turn could result in the development of new strategies for selective pharmacologic manipulation of the different cAMP signaling systems in cognitive processes under physiological and pathological conditions.

159 Chapter 7

7. Overall conclusion and future perspectives

In the current thesis we aimed to investigate the role of two recently discovered, cAMP signaling avenues, via A-kinase anchoring proteins (AKAPs) and via exchange protein activated by cAMP (Epac), in memory processes. A large number of publications established an instrumental role of cAMP signaling in learning and memory processes. With the discovery of both AKAPs and Epac, cAMP research has undergone a revival and there grew a general awareness that cAMP signaling is much more complex than was initially believed. Accordingly, it has become apparent that AKAPs and Epac might mediate a more fine-tuned level of organization for cAMP second messenger systems. Our results yield important insights in the role of AKAPs, compartmentalized cAMP signaling and Epac in the different stages of learning and memory.

It was previously believed that anchoring proteins in the brain were only scaffolds for the signaling enzymes involved in cognitive processes. Nevertheless, recent data, including our own, show a completely new role for AKAPs as dynamic and active contributors to the molecular machinery of learning and memory. With the increased number of anchoring proteins discovered in the brain, careful analysis on the individual role and contribution of AKAPs in learning and memory will continue to be an important part of future research. Unquestionably, further analysis of AKAP macromolecular complexes role in learning and memory will be forthcoming from genetic approaches involving transgenic mice. Genetically modified mice expressing truncated AKAP molecules that hinder tethering of signaling enzymes, or mice lacking a particular AKAP, will permit a more precise dissection of particular AKAP-PKA-signaling enzymes/complexes in vivo. These mice will not only provide insight into the mechanisms of AKAP functions but may yield genetic models of specific diseases.

In this thesis, we could also show that cAMP signals independent of PKA pathways are essential events in learning and memory processes. Consequently, compelling evidence points towards an important role for Epac in these processes. However, the mechanism by

160 General discussion which Epac activation affects learning and memory is still unknown. Understanding how exactly cAMP signals are coordinated will bring light on the elusive process of learning and memory and memory retrieval in particular. cAMP signaling was shown to be crucial in cognitive deficits and may consequently be a promising target for therapeutic agents. Clearly, the ultimate goal of research on learning and memory is translating knowledge from the laboratory to clinical applications. Thus, in addition to the fundamental significance of understanding cAMP signaling cascades, our research opens up new routes for the development of new therapeutic strategies for memory-associated disorders and pathologies.

8. Reference

Arshavsky, Y.I. (2006). Alzheimer's disease, brain immune privilege and memory: a hypothesis. Journal of Neural Transmission 11, 1697-1707.

Carr, D. W., Hausken, Z. E., Fraser, I. D. C., Stofko-Hahn, R. E., & Scott, J. D. (1992). Association of the Type I1 CAMP-dependent Protein Kinase with a Human Thyroid RII-anchoring Protein. Cloning and characterization of the RII-binding domain. Journal of Biological Chemistry 267, 13376- 13382.

Chai, S., Li, M., Lan, J., Xiong, Z.G., Saugstad, J.A., & Simon, R.P. (2007). A kinase-anchoring protein 150 and calcineurin are involved in regulation of acid-sensing ion channels ASIC1a and ASIC2a. Journal of Biological Chemistry 31, 22668-22677.

161 Chapter 7

Coghlan, V.M., Perrino, B.A., Howard, M., Langeberg, L.K., Hicks, J.B., Gallatin, W.M., & Scott, J.D. (1995). Association of protein kinase A and protein phosphatase 2B with a common anchoring protein. Science 5194, 108-111.

Colledge, M., Dean, R.A., Scott, G.K., Langeberg, L.K., Huganir, R.L., & Scott, J.D. (2000). Targeting of PKA to glutamate receptors through a MAGUK-AKAP complex. Neuron 27, 107-119.

Counts, S.E., Nadeem, M., Lad, S.P., Wuu, J., & Mufson, E.J. (2006). Differential expression of synaptic proteins in the frontal and temporal cortex of elderly subjects with mild cognitive impairment. Journal of Neuropathology and Experimental Neurology 6, 592-601.

Davare, M.A., Dong, F., Rubin, C.S., & Hell, J.W. (1999). The A-kinase anchor protein MAP2B and cAMP-dependent protein kinase are associated with class C L-type calcium channels in neurons. Journal of Biological Chemistry 42, 30280-30287.

Davare, M.A., Avdonin, V., Hall, D.D., Peden, E.M., Burette, A., Weinberg, R.J., Horne, M.C., Hoshi, T., & Hell, J.W. (2001). A beta2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science 5527, 98-101.

Dodge, K. L., Khouangsathiene, S., Kapiloff, M. S., Mouton, R., Hill, E. V., Houslay, M. D., Langeberg, L. K., & Scott, J. D. (2001). mAKAP assembles a protein kinase A/PDE4 phosphodiesterase cAMP signaling module. EMBO Journal 20, 1921-1930.

Esteban, J.A., Shi, S.H., Wilson, C., Nuriya, M., Huganir, & R.L., Malinow, R. (2003). PKA phosphorylation of AMPAR subunits controls synaptic trafficking underlying plasticity. Nature

162 General discussion

Neuroscience 6, 136-143.

Gekel, I., & Neher, E. (2008). Application of an Epac activator enhances neurotransmitter release at excitatory central synapses. Journal of Neuroscience 28, 7991-8002.

Gomez, L.L., Alam, S., Smith, K.E., Horne, E., & Dell'Acqua, M.L. (2002). Regulation of A-kinase anchoring protein 79/150-cAMP-dependent protein kinase postsynaptic targeting by NMDA receptor activation of calcineurin and remodeling of dendritic actin. Journal of Neuroscience 16, 7027-7044.

Gorski, J.A., Gomez, L.L., Scott, J.D., & Dell'Acqua, M.L. (2005) Association of an A-kinase- anchoring protein signaling scaffold with cadherin adhesion molecules in neurons and epithelial cells. Molecular Biology of the Cell 8, 3574-3590.

Hall, D.D., Davare, M.A., Shi, M., Allen, M.L., Weisenhaus, M., McKnight, G.S., & Hell, J.W. (2007). Critical role of cAMP-dependent protein kinase anchoring to the L-type calcium channel Cav1.2 via A-kinase anchor protein 150 in neurons. Biochemistry 6, 1635-1646.

Hell, J.W., Westenbroek, R.E., Warner, C., Ahlijanian, M.K., Prystay, W., Gilbert, M.M., Snutch, T.P., & Catterall, W.A. (1993). Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel alpha 1 subunits. The Journal of Cell Biology 4, 949-962.

163 Chapter 7

Hell, J.W., Westenbroek, R.E., Breeze, L.J., Wang, K.K., Chavkin, C., & Catterall, W.A. (1996) N- methyl-D-aspartate receptor-induced proteolytic conversion of postsynaptic class C L-type calcium channels in hippocampal neurons. Proceedings of the National Academy of Sciences United States of America 8, 3362-3367.

Hoshi, N., Langeberg, L.K., & Scott, J.D., (2005). Distinct enzyme combinations in AKAP signaling complexes permit functional diversity. Nature Cell Biology 7, 1066-1073.

Huang, L.J., Durick, K., Weiner, J.A., Chun, J., & Taylor, S.S. (1997). Identification of a novel protein kinase A anchoring protein that binds both type I and type II regulatory subunits. Journal of Biological Chemistry 12, 8057-8064.

Huang, L.J., Durick, K., Weiner, J.A., Chun, J., & Taylor, S.S. (1997). D-AKAP2, a novel protein kinase A anchoring protein with a putative RGS domain. Proceedings of the National Academy of Sciences United States of America 21, 11184-11189.

Hucho, T.B., Dina, O.A., & Levine, J.D. (2005). Epac mediates a cAMP-to-PKC signaling in inflammatory pain: an isolectin B4(+) neuron-specific mechanism. Journal of Neuroscience 25, 6119- 6126.

Isiegas, C., Park, A., Kandel, E.R., Abel, T., & Lattal, K.M. (2006). Transgenic inhibition of neuronal protein kinase A activity facilitates fear extinction. Journal of Neuroscience 26, 12700-12707.

Jiang, C.H., Tsien, J.Z., Schultz, P.G.,& Hu, Y. (2001). The effects of aging on gene expression in the hypothalamus and cortex of mice. Proceedings of the National Academy of Sciences United States of America 4, 1930-1934.

Kang, G., Joseph, J.W., Chepurny, O.G., Monaco, M., Wheeler, M.B., Bos, J.L., Schwede, F., Genieser, H.G., & Holz, G.G. (2003). Epac-selective cAMP analog 8-pCPT-2'-O-Me-cAMP as a

164 General discussion stimulus for Ca2+-induced Ca2+ release and exocytosis in pancreatic beta-cells. Journal of Biological Chemistry 10, 8279-8285.

Kapiloff M.S., Schillace R.V., Westphal A.M., & Scott J.D. (1999). mAKAP: an A-kinase anchoring protein targeted to the nuclear membrane of differentiated myocytes. Journal of Cell Science 114, 3167-3176.

Klauck, T.M., Faux, M.C., Labudda, K., Langeberg, L.K., Jaken, S., & Scott, J.D. (1996). Coordination of three signaling enzymes by AKAP79, a mammalian scaffold protein. Science 271, 1589-1592.

Li, J., O'Connor, K.L., Cheng, X., Mei, F.C., Uchida, T., Townsend, C.M. Jr., & Evers, B.M. (2007). Cyclic adenosine 5'-monophosphate-stimulated neurotensin secretion is mediated through Rap1 downstream of both Epac and protein kinase A signaling pathways. Molecular Endocrinology 1, 159- 171.

Lotfi, S., Li, Z., Sun, J., Zuo, Y., Lam, P.P., Kang, Y., Rahimi, M., Islam, D., Wang, P., Gaisano, H.Y., & Jin, T. (2006). Role of the exchange protein directly activated by cyclic adenosine 5'- monophosphate (Epac) pathway in regulating proglucagon gene expression in intestinal endocrine L cells. Endocrinology 8, 3727-3736.

165 Chapter 7

Martin, S.J., Grimwood, P.D., & Morris, R.G. (2000). Synaptic plasticity and memory: an evaluation of the hypothesis. Annual Review of Neuroscience 2, 649-711.

Mei, F.C., Qiao, J., Tsygankova, O.M., Meinkoth, J.L., Quilliam, L.A., & Cheng, X. (2002). Differential signaling of cyclic AMP: opposing effects of exchange protein directly activated by cyclic AMP and cAMP-dependent protein kinase on protein kinase B activation. Journal of Biological Chemistry 13, 11497-11504.

Micheau, J., & Riedel, G. (1999). Protein kinases: which one is the memory molecule? Cellular and Molecular Life Sciences 4, 534-548.

Michel J., Townley, I.K., Dodge-Kafka, K.L., Zhang, F., Kapiloff, M.S., & Scott, J.D. (2005). Spatial restriction of PDK1 activation cascades by anchoring to mAKAPα. Molecular Cell 20, 661-672.

Michel, J.J., & Scott, J.D. (2002). AKAP mediated signal transduction. Annual Review of Pharmacology and Toxicology 42, 235-257.

Moita, M.A., Lamprecht, R., Nader, K., & LeDoux, J.E., (2002). A-kinase anchoring proteins in amygdala are involved in auditory fear memory. Nature Neuroscience 5, 837-838.

Moosmang, S., Haider, N., Klugbauer, N., Adelsberger, H., Langwieser, N., Müller, J., Stiess, M., Marais, E., Schulla, V., Lacinova, L., Goebbels, S., Nave, K.A., Storm, D.R., Hofmann, F., &

166 General discussion

Kleppisch, T.(2005). Role of hippocampal Cav1.2 Ca2+ channels in NMDA receptor-independent synaptic plasticity and spatial memory. Journal of Neuroscience 43, 9883-9892.

Nie, T., McDonough, C.B., Huang, T., Nguyen, P.V., & Abel, T. (2007). Genetic disruption of protein kinase A anchoring reveals a role for compartmentalized kinase signaling in theta-burst long- term potentiation and spatial memory. Journal of Neuroscience 27, 10278-10288.

Nijholt I., Ostroveanu A., Scheper W., Penke B., Luiten P., Van der Zee E., & Eisel U. (2008). Inhibition of PKA anchoring to A-kinase anchoring proteins impairs consolidation and facilitates extinction of contextual fear memories. Neurobiology of Learning and Memory 87, 693-696.

Nijholt I.M., Dolga A.M., Ostroveanu A., Luiten P.G., Schmidt M., & Eisel U.L. (2008). Neuronal AKAP150 coordinates PKA and Epac mediated PKB/AKT phosphorylation. Cellular Signaling 10, 1715-1724.

Oestreich, E.A., Wang, H., Malik, S., Kaproth-Joslin, K.A., Blaxall, B.C., Kelley, G.G., Dirksen, R.T., & Smrcka, A.V. (2007). Epac-mediated activation of phospholipase C(epsilon) plays a critical role in beta-adrenergic receptor-dependent enhancement of Ca2+ mobilization in cardiac myocytes. Journal of Biological Chemistry 8, 5488-5495.

Oh, M.C., Derkach, V.A., Guire, E.S., & Soderling, T.R. (2006). Extrasynaptic membrane trafficking regulated by GluR1 serine 845 phosphorylation primes AMPA receptors for long-term potentiation. Journal of Biological Chemistry 2, 752-758.

Ostroveanu, A., Van der Zee, E.A., Eisel, U.L. & Nijholt, I.M. (2009). Detailed analysis of mAKAP expression in the brain of young and old mice. Submitted

Ostroveanu, A., Van der Zee, E.A., Eisel, U.L., Schmidt, M., & Nijholt, I.M. (2009). Exchange protein activated by cyclic AMP 2 (Epac2) plays a specific and time-limited role in memory retrieval. Submitted

167 Chapter 7

Pereira, L., Métrich, M., Fernández-Velasco, M., Lucas, A., Leroy, J., Perrier, R., Morel, E., Fischmeister, R., Richard, S., Bénitah, J.P., Lezoualc'h, F., & Gómez, A.M. (2007). The cAMP binding protein Epac modulates Ca2+ sparks by a Ca2+/calmodulin kinase signalling pathway in rat cardiac myocytes. The Journal of Physiology 2, 685-694.

Pilcher, W.H., & Joseph, S.A. (1984). Co-localization of CRF-ir perikarya and ACTH-ir fibers in rat brain. Brain Research 299, 91-102.

Rangarajan, S., Enserink, J.M., Kuiperij, H.B., de Rooij, J., Price, L.S., Schwede, F., & Bos, J.L. (2003). Cyclic AMP induces integrin-mediated cell adhesion through Epac and Rap1 upon stimulation of the beta 2-adrenergic receptor. The Journal of Cell Biology 4, 487-493.

Schwaerzel, M., Jaeckel, A., & Mueller, U. (2007). Signaling at A-kinase anchoring proteins organizes anesthesia-sensitive memory in Drosophila. Journal of Neuroscience, 27, 1229-1233.

Shankar, G.M., Li, S., Mehta, T.H., Garcia-Munoz, A., Shepardson, N.E., Smith, I., Brett, F.M., Farrell, M.A., Rowan, M.J., Lemere, C.A., Regan, C.M., Walsh, D.M., Sabatini, B.L., & Selkoe, D.J. (2008). Amyloid-beta protein dimers isolated directly from Alzheimer's brains impair synaptic plasticity and memory. Nature Medicine 8, 837-842.

168 General discussion

Sinnegger-Brauns, M., Hetzenauer, A., Huber, I., Renström, E., Wietzorrek, G., Berjukov ,S., Cavalli, M., Walter, D., Koschak, A., Waldschütz, R., Hering, S., Bova, S., Rorsman, P., Pongs, O., Singewald, N., & Striessnig, J. (2004). Isoform-specific regulation of mood behavior and pancreatic βcell and cardiovascular function by L-type Ca 2+ channels. Journal of Clinical Investigation 10, 1430-1439.

Smith, K.E., Gibson, E.S., & Dell'Acqua, M.L. (2006). cAMP-dependent protein kinase postsynaptic localization regulated by NMDA receptor activation through translocation of an A-kinase anchoring protein scaffold protein. The Journal of Neuroscience 9, 2391-2402.

Snyder, E.M., Colledge, M., Crozier, R.A., Chen, W.S., Scott, J.D., & Bear, M.F. (2005). Role for A kinase-anchoring proteins (AKAPS) in glutamate receptor trafficking and long term synaptic depression. Journal of Biological Chemistry 280, 16962-16968.

Soderling, S.H., Langeberg, L.K., Soderling, J.A., Davee, S.M., Simerly, R., Raber, J., & Scott, J.D. (2003). Loss of WAVE-1 causes sensorimotor retardation and reduced learning and memory in mice. Proceedings of the National Academy of Sciences United States of America 100, 1723-1728.

Ster, J., De Bock, F., Guérineau, N.C., Janossy, A., Barrère-Lemaire, S., Bos, J.L., Bockaert, J., & Fagni, L.(2007). Exchange protein activated by cAMP (Epac) mediates cAMP activation of p38 MAPK and modulation of Ca2+-dependent K+ channels in cerebellar neurons. Proceedings of the National Academy of Sciences United States of America 104, 2519-2524.

Ster, J., de Bock, F., Bertaso, F., Abitbol, K., Daniel, H., Bockaert, J., & Fagni, L. (2009). Epac mediates PACAP-dependent LTD in the hippocampus. The Journal of Physiology 1, 101-113.

Tunquist, B.J., Hoshi, N., Guire, E.S., Zhang, F., Mullendorff, K., Langeberg, L.K., Raber, J., & Scott, J.D. (2008). Loss of AKAP150 perturbs distinct neuronal processes in mice. Proceedings of the National Academy of Sciences United States of America 34, 12557-1262.

169 Chapter 7

Wemmie, J.A., Chen, J., Askwith, C.C., Hruska-Hageman, A.M., Price, M.P., Nolan, B.C., Yoder, P.G., Lamani, E., Hoshi, T., Freeman, J.H. Jr, & Welsh, M.J. (2002).The acid-activated ion channel ASIC contributes to synaptic plasticity, learning, and memory. Neuron 3, 463-477.

Wemmie, J.A., Askwith, C.C., Lamani, E., Cassell, M.D., Freeman, J.H. Jr, & Welsh, M.J. (2003). Acid-sensing ion channel 1 is localized in brain regions with high synaptic density and contributes to fear conditioning. Journal of Neuroscience 13, 5496-5502.

Wemmie, J.A., Coryell, M.W., Askwith, C.C., Lamani, E., Leonard, A.S., Sigmund, C.D., & Welsh, M.J. (2004). Overexpression of acid-sensing ion channel 1a in transgenic mice increases acquired fear-related behavior. Proceedings of the National Academy of Sciences United States of America 10, 3621-3626.

Yao, J.-J., Huang, Z.-H., Masten, S.J., Mizutani, T., Nakashima, S., & Nozawa, Y. (1998). Changes in the expression of protein kinase C (PKC), phospholipases C (PLC) and D (PLD) isoforms in spleen, brain and kidney of the aged rat: RT-PCR and Western blot analysis Mechanisms of Ageing and Development 1, 151-172.

170 Summary of the thesis

Summary of the thesis

The objective of this thesis was to investigate the role of two, recently discovered, cAMP signaling avenues in memory processes. We addressed the role of cAMP-dependent protein kinase (PKA) anchoring to A-kinase anchoring proteins (AKAPs) and the newly discovered cAMP effector, exchange factor activated by cAMP (Epac) in learning and memory.

In Chapter 1, we reviewed how compartmentalization of cAMP signaling is achieved via tethering of PKA to a family of scaffolding proteins called A-kinase anchoring proteins (AKAPs). We discuss how several AKAPs and PKA anchoring to AKAPs could contribute to the molecular machinery of synaptic plasticity and learning and memory. Moreover, we address how exchange protein activated by cAMP (Epac) could contribute to learning and memory processes.

In Chapter 2, using both immunohistochemical and Western blot analysis, we presented a detailed distribution of AKAP150 protein in the mouse brain. Our results showed that AKAP150 is highly expressed in striatum, cerebral cortex and forebrain regions. A relatively high expression was found in the hippocampus and olfactory bulb. In contrast, low/no AKAP150 expression was observed in cerebellum, hypothalamus, thalamus and brain stem. At the cellular level, AKAP150 was detected both in perikarya and dendrites. Because AKAP150 was highly expressed in brain regions known to be involved in learning and memory, we suggested that AKAP150 could play a role in these processes.

In Chapter 3, we addressed the question whether AKAP150 protein levels in the mouse hippocampus change during a learning task. Several groups of mice were trained in a contextual fear conditioning paradigm, a hippocampus-dependent learning task. In order to learn the task (to associate the conditioning box with fear) one group of mice received an electric footshock 3 minutes after they were introduced in the conditioning box. To detect whether the electric footshock by itself or the novelty situation (exposure to the conditioning box) has an effect on AKAP150 expression, two control groups were included: one group which received the footshock immediately after mice were introduced

171 Summary of the thesis in the conditioning box and one control group in which mice were introduced in the box but they received no foot shock. Untreated (naïve) mice were used to detect the basal level of AKAP150 in the hippocampus. AKAP150 protein levels were observed to be upregulated 6 hours after mice were subjected to a contextual fear conditioning training trial, independent of the group, when compared to the untreated mice. To eliminate the novelty component, mice were habituated by pre-exposing them to the novel environment (conditioning box), before conditioning. Now 6 hours after training, only mice that learned to fear the box showed upregulated levels of AKAP150 protein in the hippocampus. Our data suggested an important role for AKAP150 in associative learning and novelty processing and showed for the first time that protein levels of an AKAP are actively regulated during a learning task.

In Chapter 4, we studied the importance of PKA anchoring to AKAPs during the different stages of fear learning: acquisition, consolidation, retrieval and extinction. Accordingly, PKA anchoring disruptors were injected in the CA1 area of the hippocampus at various time points during the learning task. We found that disrupting PKA anchoring to AKAPs in the hippocampus impairs the consolidation phase of learned fear whereas it has no effect on acquisition and retrieval. In addition, hippocampal delivery of peptides that inhibit PKA anchoring promoted the extinction of fear memory. Unfortunately, the peptides used are not specific for a particular AKAP. However, from the finding that injection of the PKA anchoring disruptor, St-superAKAP-IS, impairs PKA anchoring to AKAP150 and our finding that AKAP150 expression is upregulated during the consolidation phase of fear learning (Chapter 3), we concluded that inhibition of PKA anchoring to AKAP150 is at least in part responsible for impaired memory consolidation. Overall, our data suggested that the temporal and spatial specificity of the hippocampal PKA signaling pathway, mediated by AKAPs, is critical to consolidate long-term contextual fear memory whereas PKA anchoring to AKAPs may put a constraint on extinction.

In Chapter 5, we characterized the distribution of a newly discovered AKAP in the brain, named mAKAP. Although it was initially discovered in the heart, we could show that mAKAP is also abundantly expressed in the mouse brain. In general, mAKAP could be found throughout the whole mouse brain. mAKAP expression was particularly abundant in

172 Summary of the thesis the hippocampus, hypothalamus and olfactory bulb, whereas the cortex, cerebellum and striatum exhibited a more moderate mAKAP staining. The lowest expression level was detected in the brain stem. At subcellular levels, mAKAP is abundantly expressed both in perikarya (e.g: pyramidal neurons, Purkinje neurons), dendrites (e.g.: apical and basal dendrites of hippocampal pyramidal neurons) and to some extent in fibers (e.g: amygdala, cortex). Interestingly, the highest mAKAP expression was detected presynaptically in brain regions such as central amygdaloid nucleus, supraoptic nucleus, suprachiasmatic nucleus. Moreover, in aged mice, we found a dramatic decrease in mAKAP expression throughout the entire mouse brain. For example, perikarya in the cortex and hippocampus exhibited a low mAKAP expression, and also in the central amygdaloid nucleus and paraventricular hypothalamic nucleus mAKAP was now hardly detectable.

In Chapter 6, we focused on the function of exchanged protein activated by cAMP (Epac1/2), a PKA-independent cAMP effector, in the mouse brain. We showed that Epac 2 is abundantly expressed in the mouse brain, while Epac 1 can hardly be detected. Furthermore, we explored the role of Epac in learning and memory, particularly in different stages of contextual fear memory: acquisition, consolidation and retrieval. During contextual fear conditioning, 8-pCPT-2’O-Me-cAMP, a cAMP analogue which specifically activates Epac but not PKA, was delivered to the mouse hippocampus at different time points corresponding to acquisition, consolidation or retrieval, and memory performance was measured. We showed for the first time that Epac activation in the hippocampus improves fear memory retrieval and has no effect on acquisition and consolidation. The enhancement of memory retrieval was confirmed in passive avoidance, another one trial fear motivated task. Since Epac activation had no effect on anxiety we can exclude that the performance of the mice in the retention tests may be influenced by the level of anxiety the animal experiences. To date, no Epac inhibitors are available. Therefore, we established an in vivo lipid mediate gene silencing method in the brain, to investigate the role of Epac silencing in fear memory retrieval. Reduced Epac2 expression in the mouse hippocampus via Epac2 siRNA transfection, completely abolished the 8-pCPT-2’O-Me-cAMP-induced enhancement of retrieval and already caused a significant decrease in freezing behavior by itself during the retention test 72 h after training. Interestingly, downregulation of Epac2

173 Summary of the thesis expression before a second retention test, 17 days after the training trial, had no effect on memory performance. In summary, our results showed a time-limited role for Epac2 signaling in memory retrieval and opens new avenues to investigate the molecular mechanisms underlying memory retrieval.

Chapter 7 summarizes and discusses all findings of the thesis and provides an overall conclusion and future perspectives.

174 Nederlandse samenvatting

Nederlandse samenvatting:

Het hoofddoel van het onderzoek beschreven in deze dissertatie was het bestuderen van de rol van twee recent ontdekte cAMP signaalwegen in leer- en geheugenprocessen. We hebben gekeken naar de rol van binding van cAMP-afhankelijk eiwit kinase (PKA) aan A- kinase bindende eiwitten (AKAPs) en de rol van de cAMP effector, exchange factor geactiveerd door cAMP (Epac).

In Hoofdstuk 1 wordt achtergrond informatie gegeven over hoe lokale en precieze cAMP gemedieerde signalering bereikt kan worden via de binding van PKA aan een familie van scaffold eiwitten genaamd A-kinase bindende eiwitten (AKAPs). We bespreken hoe verschillende AKAPs en de binding van PKA aan AKAPs kan bijdragen aan de moleculaire processen die ten grondslag liggen aan synaptische plasticiteit en leren en geheugen. Daarnaast wordt er een hypothese opgesteld over de rol van Epac in leer- en geheugen processen.

In Hoofdstuk 2 geven we een gedetailleerde beschrijving van de lokalisatie van het eiwit AKAP150 in de hersenen van muizen zoals verkregen met behulp van immunohistochemische technieken en Western blotting. Onze resultaten laten zien dat AKAP150 in hoge mate gevonden kan worden in het striatum, de cerebrale cortex en andere gebieden in de voorhersenen. Een wat minder hoge expressie van AKAP150 werd gevonden in de hippocampus en de bulbus olfactorius. Daarentegen werd er bijna geen AKAP150 waargenomen in het cerebellum, de hypothalamus, de thalamus en de hersenstam. Op het niveau van de cel was AKAP150 zowel in het cellichaam als ook in de dendrieten zichtbaar. Omdat AKAP150 vooral in verhoogde mate werd gevonden in gebieden die betrokken zijn bij leren en geheugen, verwachtten wij dat AKAP150 een belangrijke rol speelt in deze processen.

De hoofdvraag die we in Hoofdstuk 3 wilden beantwoorden was of de AKAP150 eiwit niveaus in de hippocampus van muizen veranderen gedurende een leertaak. Verschillende groepen muizen werden getraind in een hippocampus-afhankelijke leertaak genaamd

175 Nederlandse samenvatting

contextuele angst conditionering. In deze test leren muizen om een associatie te maken tussen de conditioneringsbox en een milde voetschok. Tijdens de training worden de muizen gedurende 3 minuten in de conditioneringsbox geplaatst en daarna krijgen ze een milde voetschok. De volgende dag wordt het geheugen getest door de dieren opnieuw 3 minuten in dezelfde box te plaatsen. Als ze de associatie tussen de box en de voetschok hebben gemaakt zullen ze zogenaamd “freezing” gedrag (een bewegingsloze houding) laten zien. Om te kunnen bepalen of de voetschok zelf of het blootstellen aan een nieuwe omgeving al een verandering in AKAP150 expressie teweeg brengt, hebben we twee controle groepen geïntroduceerd: een groep die meteen een voetschok kreeg zodra ze in de box geplaatst werden en de tweede groep werd wel in de box geplaatst maar ontving geen voetschok. Het basale expressie niveau van AKAP150 werd bepaald in de hippocampus van naïeve (onbehandelde) muizen. AKAP150 eiwit niveaus bleken 6 uur na de trainingssessie verhoogd te zijn, onafhankelijk van het trainingsprotocol, en dus in alle groepen. Kennelijk leidt de plaatsing in een nieuwe omgeving al tot verhoogde AKAP150 niveaus tijdens de leerfase waarin informatie wordt opgeslagen. Om dit effect te kunnen onderscheiden van een effect specifiek door leren, werden de muizen in een volgend experiment eerst meerdere keren in de conditioneringsbox geplaatst voorafgaand aan de trainingssessie om ze te laten wennen aan de testomgeving. Nu zagen we alleen een verhoogde AKAP150 expressie in de hippocampus van dieren die daadwerkelijk geleerd hadden, de associatie tussen de box en de voetschok te maken. Deze data laten voor de eerste keer zien dat AKAP150 een belangrijke rol speelt bij associatief leren en dat het eiwit niveau van een AKAP actief kan variëren tijdens een leertaak.

In Hoofdstuk 4 hebben we het belang van PKA binding aan AKAPs gedurende de verschillende stadia van het leerproces (het vastleggen, opslaan, ophalen en uitdoven van informatie) bestudeerd. Hiertoe hebben we op verschillende tijdspunten tijdens het leerproces lokaal in de hersenen stoffen toegediend die de binding van PKA aan AKAPs kunnen blokkeren. Het verhinderen van de binding van PKA aan AKAPs bleek het opslaan van informatie te verslechteren terwijl het vastleggen en het ophalen van informatie er niet door werden beïnvloed. Daarnaast werd het uitdoven van de geleerde associatie bij herhaaldelijk terugplaatsen in de box zonder voetschok toediening versneld door toediening

176 Nederlandse samenvatting van de inhibitoren. Helaas zijn deze stoffen niet gericht tegen de binding van PKA aan een specifiek AKAP maar eerder tegen de binding aan meerdere AKAPs. We konden echter wel laten zien dat toediening van de inhibitoren de binding van PKA aan AKAP150 vermindert. In combinatie met de bevinding dat AKAP150 belangrijk is bij het opslaan van informatie (Hoofdstuk 3) kunnen we dus concluderen dat inhibitie van de binding van PKA aan AKAP150 tenminste ten dele verantwoordelijk is voor de verslechterde informatie opslag die we zagen na de toediening van de inhibitoren. Over het algemeen duiden onze data op een tijds- en plaats-afhankelijke rol van hippocampale PKA signaalwegen, welke gecoördineerd wordt door AKAPs, in de opslag van het lange termijn geheugen in een contextuele angst conditioneringstest. Daarnaast vertraagt de binding van PKA aan AKAPs het uitdoven van informatie.

In Hoofdstuk 5 wordt de lokalisatie van een recentelijk ontdekt AKAP, genaamd mAKAP beschreven in de hersenen van muizen. Dit AKAP eiwit is voor het eerst ontdekt in het hart maar wij hebben kunnen laten zien dat het ook in hoge mate aangetroffen kan worden in de hersenen. Over het algemeen kan mAKAP in vrijwel alle delen van de hersenen gevonden worden. De hoogste expressie zagen we in de hippocampus, de hypothalamus en de bulbus olfactorius terwijl ook nog redelijk veel mAKAP te zien was in de cortex, het cerebellum en het striatum. In de hersenstam werd bijna geen mAKAP gevonden. mAKAP was zowel in celllichamen (bijv. pyramidale en Purkinje neuronen) als ook tot op bepaalde hoogte in celvezels (bijv. amygdala en cortex) zichtbaar. Opvallend was dat de hoogste mAKAP expressie waargenomen werd in presynaptische hersengebieden zoals in de centrale kern van de amygdala, de supraoptische kern en de suprachiasmatische kern. Daarnaast vonden we een dramatische verlaging van de mAKAP niveaus in oude muizen. In cellichamen van de cortex en de hippocampus maar ook in de centrale kern van de amygdala en de paraventriculaire hypothalamus kern van deze dieren werd bijvoorbeeld nog nauwelijks mAKAP waargenomen.

In Hoofdstuk 6 hebben we ons geconcentreerd op de rol van exchange protein geactiveerd door cAMP (Epac1/2), een PKA-onafhankelijke cAMP effector, in de hersenen van muizen. We hebben kunnen laten zien dat er voornamelijk Epac2 aanwezig is in de

177 Nederlandse samenvatting

hersenen terwijl Epac1 nauwelijks aanwezig is. Daarnaast hebben we de rol van Epac in leer- en geheugenprocessen bestudeerd met name in de verschillende stadia van het leerproces: het vastleggen, opslaan of ophalen van informatie. Om de verschillende stadia te beïnvloeden werd op verschillende tijdspunten tijdens de contextuele angst conditionering een specifieke activator van Epac, de cAMP analoog 8-pCPT-2’O-Me- cAMP, geïnjecteerd in de hippocampus en het effect op het leervermogen gemeten. In deze studie hebben we voor de eerste keer kunnen laten zien dat Epac activatie het ophalen van eerder vastgelegde informatie verbetert terwijl het geen effect heeft op het vastleggen en opslaan van informatie. De verbetering in het ophalen van informatie werd ook gezien in een andere leertaak genaamd passive avoidance. Omdat Epac activatie geen vermindering of versterking van angstgedrag bij muizen liet zien in een gedragstaak die specifiek de mate van angst bij muizen meet, kunnen we concluderen dat de resultaten in de twee leertaken te wijten zijn aan effecten op het leervermogen en niet op de mate van angst die het dier ervaart. Helaas zijn er vandaag de dag nog geen Epac inhibitors beschikbaar. Om toch te kunnen bepalen wat een inhibitie van Epac voor effect heeft op het leervermogen, hebben we met behulp van Epac2 specifieke siRNA constructen de expressie van Epac2 in de hippocampus sterk verminderd net voor de geheugentest. Verminderde hippocampale Epac2 expressie resulteerde in een verslechtering van het leervermogen in de geheugentest drie dagen na de trainingssessie en het verhinderde de 8-pCPT-2’O-Me-cAMP geïnduceerde verbetering in het ophalen van het geheugen. Opvallend was dat een verlaging van de Epac2 expressie 17 dagen na de training niet resulteerde in minder freezing gedrag tijdens de geheugentest. Dit duidt erop dat Epac2 alleen gedurende een bepaalde tijd (relatief kort na de trainingssessie) een rol speelt bij het ophalen van informatie. Dit is een van de eerste meldingen van een stof die alleen een effect heeft op het ophalen van het geheugen. Deze bevinding biedt dus heel veel nieuwe mogelijkheden om de mechanismen die ten grondslag liggen aan het ophalen van eerder opgeslagen informatie verder te onderzoeken.

In hoofdstuk 7 worden alle resultaten samengevat en verder bediscussieerd aan de hand van literatuurgegevens. Dit hoofdstuk eindigt met een algemene conclusie en suggesties voor toekomstig onderzoek.

178 Acknowledgements

Acknowledgements

I started my PhD project on a very rainy day in September. Nothing special you might think but that rainy day was followed by approximately 3 weeks of nasty Dutch weather, in which I forgot what the sun looks like and how is to go out without rain protection. However, since rain is considered a good omen, I thought this was a sign for the things to come. Now, more than four and a half years later, I can definitely say that my PhD journey caused a remarkable change both in my professional and private life. I find it difficult to express my gratitude and admiration to the people who have contributed to the making of this thesis. I just hope that all the names written here will receive the proper value of their influence. First of all, I would like to express my gratitude to my promoters, Paul Luiten, whose knowledge added considerably to my graduate experience, Eddy van der Zee, who introduced me to the wonderful world of brain anatomy and Ulrich Eisel who refreshed my memory of molecular biology. To all of you thank you and all my consideration for your supervision. I cannot express enough my gratitude to my copromoter Ingrid Nijholt. Not only that I had the fortune to work with you but I also had the possibility to learn… and learn… and learn more from your experience. I remember the overwhelming beginning, the long working hours, and finally the exciting results. It was a pleasure setting up new techniques, designing and performing experiments, and making order in the data chaos! I believe with all my heart that your supervision and assistance allowed me to extend my knowledge and to become the scientist that I am today. Moreover, thank you and Robert-Jan for your friendship and all your help. I thank the members of the reading committee: Martina Schmidt, Matthijs Verhage and Benno Roozendaal for their time, effort and timely response. I would also like to acknowledge the collaborators. Thank you Martina Schmidt for the helpful discussions and for kindly providing antibodies and the Epac activator, and Botond Penke for helping synthesize the peptide inhibitor. Likewise, thank you Csaba Nyakas for comprehensive discussions over immunostaining. I would also like to thank the master students I have “tortured” during the last four years: Janne Papma, Martijn Clausen, Marco de Bruyn, Anna Rybczyńska, Diane Jansen, Marcia

179 Acknowledgements

Peters, and Sabina Lukovac. To all, thank you for all your help. Special thanks to “Mister” Wouter Scheper, not only for being such a joy in the lab but also for becoming an amazing friend. Appreciation also goes out to Marije Lowik, Bert Venema and Jan Keijser for all of their technical assistance throughout my PhD training. Many thanks goes also to Joke Poelstra, Henk Visscher, and Jaap Bouwer for all the times in which their assistance helped me along the way. Furthermore, I would like to thank to all my ex-colleagues Ivica Granic, Timur Cetin, Nikoletta Dobos, Viktor Roman, Deepa Natarajan, Doretta Caramaschi, Alinde Wallinga, Girstaute Dagyte and Roelina Hagewoud. It was definitely fun having all of you around! I wish you all good luck with everything! Thank you Britta Kust, Janine Wieringa and Diana Koopmans for coordinating the BCN training programme. It was interesting to meet other PhD students, make new friends and maybe future collaborators. I would also like to acknowledge several people who have made a difference in my life. Initial, am fost foarte trist cand am plecat din tara dar am fost si mai trist cand am ajuns in Groningen. Cu toate acestea, cativa (putini dar importanti) romani m-au facut sa ma simt ca acasa departe de casa. Ii multmesc foarte mult Amaliei Dolga pentru explicatiile detaliate si ajutorul in laborator, pentru drumurile prin oras dupa acte, cinele in comunitate si nu numai. Totodata, multumesc Cristi Marocico si Lavinia Slabu pentru discutiile in comunitate, sprijinul moral acordat si mai ales pentru prietenia voastra. Inca o data multumesc pentru tot dragi romani! Draga Mihai Nedelcu (specialFX) iti multumesc pentru entuziasmul si prietenia ta, pentru drumurile de la aeroport acasa si vacantele petrecute impreuna. Recunosc, de la tine am invatat ca niciodata nu ai suficiente degete sa manevrezi nenumarate camere foto si ca nici o panoramare nu e suficient de mare. Doresc sa multumesc din toata inima familiei mele. Imi pare rau ca a trebuit sa plec de acasa pentru a-mi gasi propriul drum in viata dar apreciez faptul ca m-ati sustinut in alegerea luata. Dorul de casa imi va aminti intotdeauna de voi si ajutorul vostru. Dragii mei, mama, tata, Mari, Gheorghe si minunatii mei bunici va multumesc din suflet pentru tot! Draga mea Oana, iti multumesc foarte mult ca ai fost aproape de mine si ca ai gustat din experienta olandeza. Iti multumesc pentru dragostea ta si pentru ca ai pus ordine in haosul din jurul meu. Cu tine, viata este minunata.

180 List of publications

List of publications

Publications:

Ostroveanu A, Van der Zee EA, Eisel UL, Nijholt IM. Detailed analysis of mAKAP expression in the brain of young and old mice. Submitted

Ostroveanu A, Van der Zee EA, Eisel UL, Schmidt M, Nijholt IM. Exchange protein activated by cyclic AMP 2 (Epac2) plays a specific and time-limited role in memory retrieval. Submitted

Nijholt IM, Dolga AM, Ostroveanu A, Luiten PG, Schmidt M, Eisel UL. Neuronal AKAP150 coordinates PKA and Epac-mediated PKB/Akt phosphorylation. Cellular Signalling, 2008, 10:1715-2174.

Ostroveanu A, Nijholt IM, Scheper WA, Penke B, Luiten PG, Van der Zee EA, Eisel UL. Inhibition of PKA anchoring to A-kinase anchoring proteins impairs consolidation and facilitates extinction of contextual fear memories. Neurobiology of Learning and Memory, 2008, 1:223-229.

Dolga AM, Nijholt IM, Ostroveanu A, Ten Bosch Q, Luiten PG, Eisel UL. Lovastatin induces neuroprotection through tumor necrosis factor receptor 2 signalling pathways. Journal of Alzheimer Disease, 2008, 2:111-122.

Ostroveanu A, Van der Zee EA, Dolga AM, Luiten PG, Eisel UL, Nijholt IM. A-kinase anchoring protein 150 in the mouse brain is concentrated in areas involved in learning and memory. Brain Research, 2007, 11:97-107.

Nijholt IM, Ostroveanu A, de Bruyn M, Luiten PG, Eisel UL, Van der Zee EA. Both exposure to a novel context and associative learning induce an upregulation of AKAP150 protein in mouse hippocampus. Neurobiology of Learning and Memory, 2007, 874:693- 696.

181 List of publications

Abstracts:

Ostroveanu A, Van der Zee EA, Luiten PG, Eisel UL, Nijholt IM. 2008. Role of A-kinase anchoring proteins in contextual fear memory. Abstracts of the 6th Forum of European Neuroscience, Geneva, Switzerland.

Ostroveanu A, Van der Zee EA, Luiten PG, Eisel UL, Nijholt IM. 2008. Role of A-kinase anchoring proteins in contextual fear memory. Abstracts of the 3rd Molecular and Cellular Cognition Society Europe Meeting, Geneva, Switzerland.

Scheper WA, Ostroveanu A, Nijholt IM. Role of PKA anchoring to A-kinase anchoring proteins in fear memory. 2008. Abstracts of the 7th Endo-Neuro-Psycho Meeting, Doorwerth, the Netherlands.

Ostroveanu A, Van der Zee EA, Luiten PG, Eisel UL, Nijholt IM. Role of A-kinase anchoring proteins in fear memory. 2007. Abstracts of the 2nd International Meeting on Anchored cAMP Signalling Pathways, O.S.H.U Portland, Oregon, U.S.A.

Ostroveanu A, Van der Zee EA, Luiten PG, Eisel UL, Nijholt IM. Disruption of anchored cAMP-dependent protein kinase (PKA) signalling impairs the consolidation of contextual fear memory. 2007. Abstracts of the 6th Endo-Neuro-Psycho Meeting, Doorwerth, the Netherlands.

Dolga AM, Nijholt IM, Ostroveanu A, Luiten PG, Eisel UL. Novel lovastatin mediated neuroprotective mechanisms. 2007. Abstract of the 3rd Conference on Advances in Molecular Mechanisms of Neurological Disorders Meeting, Salamanca, Spain.

Ostroveanu A, de Bruyn M, Luiten PG, van der Zee EA, Eisel UL, Nijholt IM. A-Kinase Anchoring Protein150 in mouse brain. 2006. Abstract of the 5th Endo-Neuro-Psycho Meeting, Doorwerth, the Netherlands.

182 Curriculum vitae

Curriculum vitae

Anghelus Ostroveanu was born on January 25th, 1976 in Ciocanesti, a small village in the south of Romania. He finished secondary school in Calarasi, Romania in 1994. The same year he was admitted at the University of Bucharest, Faculty of Biology from where he received his Bachelor degree in 1998. He continued his studies at the University of Bucharest and in 2000 received his Master degree in human genetics under the supervision of Prof. Dr. Veronica Stoian. From 2000 to 2004 he was employed as Biologist and later on as Senior Biologist within the Pharmacological and Toxicological Laboratory for human- use drugs at the National Medicines Agency (Romanian Drug Regulatory Authority) in Bucharest. During 2004-2008 he followed a PhD project at the University of Groningen, Faculty of Mathematics and Natural Sciences in the Department of Molecular Neurobiology, the Netherlands, supported by an Ubbo Emmius scholarship. Here he conducted fundamental research on the role of cAMP signalling in learning and memory under the supervision of Prof. Dr. Paul Luiten, Dr. Ingrid Nijholt, Prof. Dr. Eddy Van Der Zee and Prof. Dr. Ulrich Eisel. Currently, Anghelus is working as a Post-doc fellow in the group of Prof. Dr. Jo W.M. Höppener at the Department of Metabolic and Endocrine Diseases (Division of Medical Genetics), University Medical Center Utrecht, Location Wilhelmina Children's Hospital.

183