The Role of Stress, Glucocorticoids & Β-Adrenergic

University of Pennsylvania ScholarlyCommons

Publicly Accessible Penn Dissertations

Spring 2011

The Role of Stress, Glucocorticoids & β-Adrenergic Signaling in Aversive, Hippocampus-Dependent Memory Retrieval

Keith B. Schutsky University of Pennsylvania, [email protected]

Follow this and additional works at: https://repository.upenn.edu/edissertations

Part of the Pharmacy and Pharmaceutical Sciences Commons

Recommended Citation Schutsky, Keith B., "The Role of Stress, Glucocorticoids & β-Adrenergic Signaling in Aversive, Hippocampus-Dependent Memory Retrieval" (2011). Publicly Accessible Penn Dissertations. 990. https://repository.upenn.edu/edissertations/990

This paper is posted at ScholarlyCommons. https://repository.upenn.edu/edissertations/990 For more information, please contact [email protected]. The Role of Stress, Glucocorticoids & β-Adrenergic Signaling in Aversive, Hippocampus-Dependent Memory Retrieval

Abstract Studies in wild-type rats and mice, and mutant mice lacking norepinephrine (NE) demonstrate that NE, β1-adrenergic, cAMP/PKA and Epac signaling are required for intermediate-term hippocampus-dependent memory retrieval. Although it has been proposed that glucocorticoids (GCs) facilitate β-adrenergic signaling to impair retrieval, neither a specific equirr ement for NE nor a definitive mechanism as to how this interaction occurs has been established. This thesis provides compelling evidence that 1) GCs do not require (but can synergize with) the adrenergic system to impair hippocampus-dependent memory retrieval; 2) GCs and xamoterol interact with the β2 receptor to impair retrieval; 3) β2 receptors signal through Gi/o in the hippocampus to impair retrieval, acting to negate the facilitation of hippocampus- dependent memory retrieval mediated by β1-Gs signaling; 4) GCs impair retrieval through possibly direct interaction with the β2 receptor; 5) β2 signaling is required for retrieval and may be required for short-term maintenance of fear memory; 6) β3 signaling influences etrier val; 7) interactions between β receptors are important and net effects of β receptor stimulation (or antagonism) on hippocampal cAMP signaling likely determine the degree of retrieval impairment expressed, as individual β receptors can have similar or opposing effects on cAMP levels; and 8) adrenergic and glucocorticoid influences on etrier val are limited by the age of the memory.

Degree Type Dissertation

Degree Name Doctor of Philosophy (PhD)

Graduate Group Pharmacology

First Advisor Steven A. Thomas

Keywords stress, glucocorticoids, norepinephrine, beta-adrenergic receptor, fear memory retrieval, hippocampus

Subject Categories Medicine and Health Sciences | Pharmacy and Pharmaceutical Sciences

This dissertation is available at ScholarlyCommons: https://repository.upenn.edu/edissertations/990 THE ROLE OF STRESS, GLUCOCORTICOIDS & -ADRENERGIC

SIGNALING IN AVERSIVE, HIPPOCAMPUS-DEPENDENT MEMORY

RETRIEVAL

Keith Schutsky

A Dissertation in Pharmacology Presented to the Faculties of the University of Pennsylvania in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy 2011

Supervisor of Dissertation Supervisor Signature ______

Dr. Steven A. Thomas, M.D., Ph.D., Associate Professor of Pharmacology

Graduate Group Chairperson Signature______

Dr. Vladimir Muzykantov, M.D., Ph.D., Associate Professor of Pharmacology and Medicine

Dissertation Committee

Dr. Judith Meinkoth, Ph.D., Associate Professor of Pharmacology

Dr. Edwin Abel, III, Ph.D., Brush Family Professor of Biology

Dr. Julie Blendy, Ph.D., Associate Professor of Pharmacology

Dr. Benoit Giasson, Ph.D., Assistant Professor of Pharmacology DEDICATION

To my wife, Tina, for her love, strength and support.

ACKNOWLEDGEMENTS

I am extremely grateful to Dr. Steven Thomas for giving me the opportunity join his research group, and I am privileged to have studied under his guidance. He has supported me in every aspect of the scientific process - in forming hypotheses, designing experiments, analyzing results, writing and submitting papers for publication. His mentorship and tireless efforts have enabled me to prosper as an independent-thinking scientist and prepared me to face obstacles that may lie ahead in my career. I want to give special thanks to Ming Ouyang, Sung-Ha Jin and Hong-Mei Luo for their technical expertise and for taking the time to train me. I must acknowledge current lab members,

Christina Castelino and Lei Zhang for assistance with experiments and past lab members for their friendship and support. I want to thank my thesis committee members, Dr. Judith

Meinkoth, Dr. Ted Abel, Dr. Julie Blendy and Dr. Benoit Giasson for their helpful suggestions to maintain progress in my research and for strengthening my ability to answer questions concerning my research, its implications and its limitations. I must acknowledge my family and friends for their encouragement and wisdom. Most importantly, I want to thank my wife, Tina and my two children, Jake and Logan for their unconditional love and for helping me realize the things in life that are truly important.

iii

ABSTRACT

THE ROLE OF STRESS, GLUCOCORTICOIDS & -ADRENERGIC SIGNALING IN AVERSIVE, HIPPOCAMPUS-DEPENDENT MEMORY RETRIEVAL

Keith Schutsky Steven A. Thomas

Studies in wild-type rats and mice, and mutant mice lacking norepinephrine (NE) demonstrate that NE, 1-adrenergic, cAMP/PKA and Epac signaling are required for intermediate-term hippocampus-dependent memory retrieval. Although it has been proposed that glucocorticoids (GCs) facilitate-adrenergic signaling to impair retrieval, neither a specific requirement for NE nor a definitive mechanism as to how this interaction occurs has been established. This thesis provides compelling evidence that 1)

GCs do not require (but can synergize with) the adrenergic system to impair hippocampus-dependent memory retrieval; 2) GCs and xamoterol interact with the 2 receptor to impair retrieval; 3) 2 receptors signal through Gi/o in the hippocampus to impair retrieval, acting to negate the facilitation of hippocampus-dependent memory retrieval mediated by 1-Gs signaling; 4) GCs impair retrieval through possibly direct interaction with the 2 receptor; 5)  signaling is required for retrieval and may be required for short-term maintenance of fear memory; 6)  signaling influences retrieval;

7) interactions between  receptors are important and net effects of receptor stimulation

(or antagonism) on hippocampal cAMP signaling likely determine the degree of retrieval impairment expressed, as individual receptors can have similar or opposing effects on

iv cAMP levels; and 8) adrenergic and glucocorticoid influences on retrieval are limited by the age of the memory.

TABLE OF CONTENTS

Dedication ...... ii Acknowledgements...... iii Abstract ...... iv Table of Contents ...... vi List of Tables ...... xii List of Figures ...... xiii Abbreviations ...... xv

Chapter 1: Introduction 1.1 Memory Types & Importance of Hippocampus in Episodic Memory ...... 2 1.2 Strength of Memory is Influenced by Environment and Emotional State ...... 3 1.3 Animal Models of Fear Memory & Brain Structures Involved 1.3.1 Auditory (Cued) Fear Conditioning ...... 4 1.3.2 Contextual Fear Conditioning ...... 5 1.3.3 Inhibitory (Passive Avoidance) ...... 7 1.3.4 Spatial Navigation ...... 9

1.4 Role of NE/E & -adrenergic Signaling in Fear Memory 1.4.1 Epinephrine & Enhanced Consolidation ...... 10 1.4.2 Norepinephrine & Enhanced Consolidation ...... 11 1.4.3 NE/E is not Required for Fear Memory Consolidation ...... 12

1.4.4  Signaling is not Required for Fear Memory Consolidation ...... 13 1.4.5 NE/E is Required for Intermediate-term Memory Retrieval

1.4.5.1 NE/E and Spatial Retrieval ...... 14 1.4.5.2 NE/E and Contextual Memory Retrieval ...... 14

1.4.6  Signaling is Required for Intermediate-term Memory Retrieval

1.4.6.1  Signaling and Spatial Retrieval ...... 15

1.4.6.2 1 Signaling and Contextual Memory Retrieval ...... 16

1.4.6.3 2 & 3 Receptors Influence Retrieval Independently & in Concert with 1 ...... 17 1.5 Role of Stress & Glucocorticoids in Fear Memory 1.5.1 Stress, Glucocorticoids (GCs) & Consolidation ...... 18 1.5.2 Stress, GCs & Retrieval ...... 18 1.6 Interactions between Adrenergic & GC Signaling in Fear Memory

1.6.1 NE/E,  Receptor & GC Interactions in Consolidation ...... 19

1.6.2 NE/E, Receptor & GC Interactions in Retrieval ...... 20

1.7 GCs &  Signaling can Influence cAMP-PKA Signaling 1.7.1 cAMP-PKA Signaling & Fear Memory Consolidation ...... 21 1.7.2 cAMP-PKA Signaling & Fear Memory Retrieval ...... 22 1.8 Scope of Thesis Research ...... 23 1.9 Clinical Significance of this Work 1.9.1 Adrenergic & GC Therapies in Post-traumatic Stress Disorder (PTSD) ...... 23

1.9.2 -adrenergic Antagonists & Agonists for PTSD ...... 24

1.9.3 -adrenergic Antagonists & Agonists for PTSD 1.9.3.1 Reducing Consolidation ...... 24 1.9.3.2 Reducing Retrieval...... 25 1.9.3.3 Reducing Reconsolidation ...... 26 1.9.4 Glucocorticoids for PTSD 1.9.4.1 Reducing Consolidation ...... 27 vii

1.9.4.2 Longer-term GC Administration can Impair Memories Unrelated to the Trauma ...... 28

1.9.4.3 Reducing Retrieval by Long-term, Low-dose GC Administration ...... 29 1.9.4.4 Reducing Reconsolidation ...... 29 1.10 Statement of Purpose ...... 30

Chapter 2: Stress and Glucocorticoids Impair Hippocampus- dependent Emotional Memory Retrieval via Gi/o-coupled 2- adrenergic Signaling 2.1 Abstract ...... 33 2.2 Introduction ...... 33 2.3 Results 2.3.1 Corticosterone and a GR Agonist Impair Hippocampus- dependent Memory Retrieval ...... 37

2.3.2 Impairment of Retrieval by GCs Depends on 2 Signaling but not NE/E ...... 40

2.3.3 Direct Stimulation of 2 Receptors Impairs Memory Retrieval ...... 43

2.3.4 Genetic and Interspecies Corroboration that 2 Signaling Mediates Impairment of Retrieval ...... 45 2.3.5 Impairment of Memory Retrieval by Stress Requires NE/E and 2 Signaling ...... 47

2.3.6 Corticosterone, 2 Signaling and Stress Impair Retrieval in a Time-limited Manner ...... 50

2.3.7 By Coupling to Gi/o Proteins, 2 Opposes the Role of 1 in Retrieval ...... 51 2.4 Discussion ...... 55 2.5 Experimental Procedures 2.5.1 Subjects ...... 63 2.5.2 Pavlovian Fear Conditioning ...... 64 viii

2.5.3 Instrumental Fear Conditioning ...... 65 2.5.4 Restraint Stress...... 65 2.5.5 Drugs ...... 65 2.5.6 Dorsal Hippocampus Infusion ...... 66 2.5.7 Corticosterone Levels ...... 67 2.5.8 cAMP Levels ...... 67 2.5.9 Statistics ...... 68 2.6 Acknowledgements ...... 69

Chapter 3: Xamoterol Impairs Hippocampus-dependent Emotional Memory Retrieval via Gi/o-coupled 2-adrenergic Signaling 3.1 Abstract ...... 71 3.2 Introduction ...... 72 3.3 Materials & Methods 3.3.1 Subjects ...... 76 3.3.2 Pavlovian Fear Conditioning ...... 77 3.3.3 Instrumental Fear Conditioning ...... 77 3.3.4 Drugs ...... 78 3.3.5 Dorsal Hippocampus Infusion ...... 78 3.3.6 Statistics ...... 79 3.4 Results 3.4.1 Xamoterol Impairs Hippocampus-dependent Memory Retrieval .....80 3.4.2 Pharmacologic Evidence that Xamoterol Disrupts Retrieval through 2- and not 1 adrenergic Signaling ...... 81 3.4.3 Genetic Evidence that Xamoterol Disrupts Retrieval through 2- and not1-adrenergic Signaling ...... 83

3.4.4 Pertussis Toxin Blocks the Impairing Effects of Xamoterol on Retrieval ...... 85 3.4.5 Xamoterol Impairs Hippocampus-dependent Memory Retrieval in a Time-limited Manner ...... 86 3.5 Discussion ...... 88 3.6 Acknowledgements ...... 91

Chapter 4: Roles of 2- and 3-adrenergic Signaling in Aversive, Hippocampus-dependent Memory Retrieval 4.1 Abstract ...... 93 4.2 Introduction ...... 93 4.3 Materials & Methods ...... 94 4.4 Results & Discussion

4.4.1 2 Signaling and Fear Memory ...... 94 4.4.2 Evidence for Coincident Signaling or Direct Molecular Interaction between GR and 2-adrenergic Signaling in Aversive, Hippocampus-dependent Memory Retrieval ...... 101

4.4.3 Putative Role of 3-adrenergic Signaling in Aversive, Hippocampus-dependent Memory Retrieval ...... 103

4.4.4 Combined Inactivation of 1, 2 and/or 3-adrenergic Signaling Influences Aversive, Hippocampus-dependent Memory Retrieval ...... 105

Chapter 5: Summary & Future Directions 5.1 Summary ...... 109 5.2 Future Directions

5.2.1 Ex-Vivo Application of  and GR Agents in Control and AR KO Mice to Determine Effects on DH cAMP Signaling ...... 118 5.2.2 In-Vivo Microdialysis to Measure Extracellular cAMP in Naïve and Conditioned Animals Following DH Infusion AR & GR Compounds ...... 119 x

5.2.3 Use of GRfl/fl X CaMKII-Cre+ Mice to Distinguish Between Direct Interaction or Coincident Signaling Between GR & 2-AR ...... 121

5.2.4 Use of Floxed -AR Mice to Examine Hippocampal Subfield-specific Contributions of  Signaling in Retrieval...... 123 5.2.5 Intracellular Recording from CA1 Pyramidal Neurons to Examine if -cAMP & GR Signaling Mediate Voltage-Gated Calcium Channel Effects on the Slow-afterhyperpolarization (sAHP) ...... 124

References ...... 126

LIST OF TABLES

Chapter 1: Introduction 1.1 Brain Structures Involved in Several Tests of Fear Memory ...... 10

1.2 -adrenergic Receptors, their Pharmacology and Signaling ...... 17

xii

LIST OF FIGURES

Chapter 1: Introduction 1.1 Catecholamine Biosynthesis Pathway ...... 13

Chapter 2: Stress and Glucocorticoids Impair Hippocampus-dependent Emotional Memory Retrieval via Gi/o-coupled 2-adrenergic Signaling 2.1 Corticosterone and a GR Agonist Impair Hippocampus-dependent Emotional Memory Retrieval ...... 39

2.2 Corticosterone Impairment of Memory Retrieval Depends on 2 and not 1 Signaling or NE/E ...... 42

2.3 Direct Stimulation of 2 Receptors Disrupts Memory Retrieval ...... 44

2.4 Genetic and Interspecies Corroboration that 2 Signaling Mediates Impairment of Retrieval ...... 46

2.5. Impairment of Memory Retrieval by Stress Depends on GR and 2 Signaling ...... 49

2.6. Corticosterone, 2 Signaling and Stress Inhibit Hippocampus-dependent Memory Retrieval in a Time-limited Manner ...... 51

2.7 2 Signaling Opposes 2 Signaling in Retrieval and in cAMP Production through Gi/o ...... 54

Chapter 3: Xamoterol Impairs Hippocampus-dependent Emotional Memory Retrieval via Gi/o-coupled 2-adrenergic Signaling 3.1 Xamoterol Impairs Hippocampus-dependent Memory Retrieval ...... 80

3.2 Impairment of Retrieval by Xamoterol is Prevented by a 2 but not a 2 Antagonist ...... 82

3.3 Genetic Evidence that 2 Signaling Mediates the Impairment of Memory Retrieval by Xamoterol ...... 84

xiii

3.4 Pertussis Toxin Blocks the Impairing Effects of Xamoterol on Retrieval ...... 86 3.5 Xamoterol Impairs Hippocampus-dependent Memory Retrieval in a Time- limited Manner ...... 87

Chapter 4: Role of 2- and 3-adrenergic Signaling &  Receptor Combinations in Aversive, Hippocampus-dependent Memory Retrieval

4.1 2 Signaling & Fear Memory ...... 98

4.2 GR Antagonists Prevent 2 Agonist-Mediated Disruption of Hippocampus-Dependent Memory Retrieval ...... 102

4.3 3 Signaling Influences Hippocampus-dependent Memory Retrieval ...... 104

4.4 Signaling by Multiple 2-receptor Combinations Influence Hippocampus-dependent Memory Retrieval ...... 107

Chapter 5: Summary & Future Directions

5.1 Effectors of 2-coupled, Gi/o Signaling ...... 118

xiv

ABBREVIATIONS

ADP adenosine diphosphate AP action potential

-cAMKII alpha isoform of calcium-dependent kinase II AR adrenergic receptor ATP adenosine triphosphate

Bet / Betax betaxolol (1 antagonist)

1 beta-1 adrenoceptor

fl/fl 1 1 gene flanked by loxP sites

2 beta-2 adrenoceptor

3 beta-3 adrenoceptor BLA basolateral amygdaloid complex 8-Br-cAMPS cAMP activating analog CA1 Cornu Ammonius area 1 (hippocampus) CA3 Cornu Ammonius area 3 (hippocampus) cAMP cyclic adenosine monophosphate CeA central nucleus of the amygdala CF context fear CNS central nervous system Co-IP co-immunoprecipitation Cort corticosterone CR conditioned response Cre Cre recombinase CREB cAMP response element binding protein xv

CS conditioned stimulus CUP caudate-putamen d day(s)

DBH dopamine -hydroxylase enzyme

Dbh dopamine -hydroxylase gene DG dentate gyrus (hippocampus) DH dorsal hippocampus DMF dimethylformamide E epinephrine Epac exchange protein activated by cAMP ERK extracellular signal-regulated kinase

GABA -aminobutyric acid GAD67 enzyme expressed in inhibitory neurons GC glucocorticoid GFAP enzyme expressed in glial cells

Gi/o subunit of G-protein that inhibits adenylate cyclase GP globus pallidus GR glucocorticoid receptor GRfl/fl GR gene flanked by loxP sites h hour(s) HPA hypothalamo-pituitary-adrenal axis IA inhibitory avoidance

ICI ICI 118,551 (2 antagonist) ICV intracerebroventricular IEG immediate early gene IHC immunohistochemistry

xvi

ISH in situ hybridization

Iso isoproterenol ( agonist) LA lateral nucleus of the amygdala LC locus coeruleus L-DOPS L-threo-3,4-dihydroxyphenylserine LoxP DNA sequence recognized by Cre LTP long-term potentiation LTM long-term memory MR mineralocorticoid receptor MWM Morris water maze NAcc nucleus accumbens NBM nucleus basalis magnocellularis NE norepinephrine NTS nucleus of the solitary tract Org / Org 34850 GR antagonist PAG periacqueductal gray matter PBNC parabrachial nuclear complex PFC prefrontal cortex PI3K phosphoinositide 3-kinase PKA protein kinase A

Pro / Prop propranolol ( antagonist)

Proc procaterol (2 agonist) PTSD post-traumatic stress disorder PTx pertussis toxin Rap2 Ras-related protein Ras rat sarcoma protein

xvii

Res / Restraint restraint stress Rp-cAMPS PKA inhibitor RU28362 GR agonist RU486 mifepristone (GR and progesterone antagonist) s second(s) sAHP slow-afterhypolarization sal saline SAT scholastic aptitude test SN substantia nigra

SR59230A 3 antagonist STM short-term memory TTX tetrodotoxin US unconditioned stimulus veh vehicle VGCC voltage-gated calcium channels VH ventral hippocampus wk week(s) WT wild-type

Xam xamoterol (1 agonist)

Zint zinterol (2 agonist)

xviii

Chapter 1

Introduction

Keith Schutsky & Steven A. Thomas

Department of Pharmacology, University of Pennsylvania, Philadelphia, PA 19104

1.1 Memory Types & Importance of Hippocampus in Episodic Memory

In most circumstances, the ability to learn and remember possesses great adaptive value, beneficial for an organism’s long-term survival and reproductive success.

Recently, scientists have begun to unravel the molecular and cellular mechanisms underlying the formation, storage and subsequent recall of both implicit and explicit memories[1-3]. Implicit memory is a type of long-term memory that doesn't involve conscious thought. Improvements of penmanship or driving ability by repetition are examples. Conversely, explicit memory can be defined as the conscious, intentional recall of past experiences and information. Remembering historical facts or recounting intimate details from an event from years ago involve this type of memory.

Explicit memory can be further subdivided into semantic and episodic memory.

The former involves "textbook learning" or general factual knowledge about the world. It enables one to say, without knowing exactly when and where one learned, that George

Washington was our first president, or that Philadelphia is a major city in Pennsylvania.

In contrast, episodic memory involves memory of times, places, associated emotions, and other contextual knowledge. It allows one to remember what he ate for dinner last night, who told him that a family member passed away, and where it was he received his first kiss. Episodic memory is centered in the brain's hippocampus, at least preliminarily[4].

Substantial evidence indicates that the hippocampus is initially required for the acquisition, consolidation and retrieval of episodic information[4]. However, the necessity of this structure is thought to dissipate during ―systems consolidation‖, such that the hippocampus is no longer required for memory retrieval[4]. Support for this

2 comes from hippocampal lesion studies in animals and examination of individuals with traumatic brain injury, in which bilateral damage to the hippocampus eliminates the ability to form new episodic memories, as well as impairs the ability to recall recent but not remote episodic memories[5-7].

1.2 Strength of Memory is Influenced by Environment and Emotional State Importantly, the strength of long-term, episodic memory is influenced by the environment in which it is acquired. Situations of high consequence (i.e. strong positive or negative outcome for the organism) lead to enhanced emotional arousal and often result in more rapidly acquired, powerful and/or longer-lasting memories[8, 9]. Enhanced consolidation from such events is thought to be mediated by neuromodulatory systems, including the adrenergic & glucocorticoid systems, whose activity correlates with the valence (―saliency‖) of the situation[10, 11]. For example, individuals learn to associate with people, places and things that bring them pleasure and strive to avoid those that cause them pain. In fact, a person’s most vivid episodic memories usually stem from events filled with emotions ranging from joyful to horrifying. While most of these memories are beneficial and have adaptive significance, intense, psychologically terrifying experiences can lead to psychiatric illnesses including post-traumatic stress disorder (PTSD), a disorder characterized by both the excessive retrieval and enhanced reconsolidation of traumatic memories[12, 13]. In the past decade, the dramatic increase in the diagnosis and treatment of PTSD would, in and of itself, serve as a strong impetus to better understand the molecular and cellular mechanisms underlying episodic memory recall. However, episodic memory is also believed to be impaired in a multitude of other 3 diseases, including depression[14], dementia[15], autism[16], Alzheimer’s[17],

Korsakoff's syndrome[18] and Cushing’s disease[19]. Interestingly, many of these illnesses are believed to involve alterations in adrenergic signaling[20, 21].

1.3 Animal Models of Fear Memory & Brain Structures Involved 1.3.1 Auditory (Cued) Fear Conditioning Auditory fear conditioning is form of associative learning considered to be an example of Pavlovian (classical) fear conditioning. In this behavioral paradigm, rodents learn to respond defensively to an initially neutral stimulus (the conditioned stimulus;

CS) (e.g. tone) after it has been associated or paired with a harmful stimulus (the unconditioned stimulus; US) (e.g. footshock). In rodents, presentation of the CS after conditioning in a distinct context elicits defensive behavior (e.g. freezing). Freezing is a species-specific response to fear, which is defined as the absence of movement except for respiration. Intermittent freezing may last for seconds to minutes depending on US intensity, number of US presentations and the extent of learning achieved by the rodent.

Cued fear is considered to be an implicit form of memory that does not depend on the hippocampus[22, 23]. Thus, cued fear serves as an excellent paradigm to control for non-specific effects of drugs that might selectively influence hippocampus-dependent memory retrieval. The neural underpinnings of cued fear are thought to be localized in discrete brain regions, some of which are similarly involved in paradigms designed to test hippocampus-dependent fear memory (1.3.2-1.3.4).

Auditory fear conditioning involves transmission of auditory and somatosensory information (from the thalamus and cortex) to the lateral nucleus of the amygdala (LA)

(the LA is part of the basolateral amygdaloid complex, or BLA), an area that is critical for fear conditioning and thought to be essential in integrating information about the CS and US[24]. Electrolytic, excitotoxic or functional inactivation of the LA cause deficits in the acquisition and expression of auditory fear conditioning[25-27].

Receiving projections from the LA, the central nucleus of the amygdala (CeA) is required for the acquisition, consolidation and expression of cued fear, demonstrated by lesion and functional inactivation experiments[28-30]. Most scientists consider the CeA to be the major output nucleus of a coordinated fear system that acts to carry out a variety of instinctive, species-specific responses that underlie defensive behaviors[24] and ablation of the CeA impairs the expression of most CRs, including freezing[24, 31, 32].

1.3.2 Contextual Fear Conditioning

Contextual fear conditioning is a form of Pavlovian conditioning in which an animal is placed into a novel environment and allowed to explore its surroundings for a discrete period of time. Then, an aversive stimulus (e.g. US-shock) is administered and the animal removed. When the animal is placed in the same environment at a later time, it demonstrates freezing if it has formed a context-US association.

As in auditory fear conditioning, lesions or reversible inactivation of the BLA and

CeA impair the acquisition and expression of contextual fear (CF)[24, 27, 33-36].

Interestingly, the CeA projects to areas of the midbrain periacqueductal gray (PAG), hypothalamus, and brain stem that orchestrate behavioral, endocrine, and autonomic conditioned responses (CRs) associated with auditory and contextual fear learning[24].

Notably, the PAG has been shown to be critical for mediating defensive behavior.

Chemical or electrical stimulation of the dorsal PAG promotes ―flightlike‖ behavior, while damaging the dPAG attenuates this response. In contrast, chemical or electrical stimulation of the ventral PAG causes freezing, and pre-training lesions in the vPAG significantly impair both contextual and auditory fear conditioning [32, 35, 37].

In addition, the dorsal hippocampus (DH) plays a critical role in expression of contextual fear learning, supported by electrolytic and excitotoxic lesion experiments[22,

38, 39]. Interestingly, electrolytic but not neurotoxic DH lesions impair the acquisition of context fear[23, 39]. A critical aspect of context fear is that the effect of the DH damage is time-dependent[24, 31, 39]. If the DH is ablated 1-14 days post-conditioning, rodents exhibit dramatic deficits in CF expression[24, 31, 39]. On the other hand, if the damage occurs more than 28 days after training, CF impairment is minimal[24, 31, 39]. A time- limited role for the hippocampus in contextual fear expression mirrors the symptoms of retrograde amnesia seen in human patients who have experienced hippocampus damage

(e.g. traumatic or anoxic brain injury). Memories acquired just prior to the injury, for example, are severely impacted, while those acquired several years before are left essentially intact[4].

Receiving projections from the amygdala, the ventral hippocampus (VH) also plays an important role in contextual fear conditioning. Electrolytic ablation, excitotoxic lesions or functional inactivation of the VH inhibit the acquisition and expression of contextual fear[40-43]. Similarly, the nucleus accumbens (NAcc) and prefrontal cortex

(PFC) influence various aspects of CF[31].

1.3.3 Inhibitory (Passive) Avoidance

Inhibitory avoidance (IA) is a form of instrumental conditioning in which the animal learns to avoid an aversive US (e.g. footshock) by remaining in the well-lit side of a two-chamber apparatus, refraining from entering the dark side where it previously received the US. Because rodents are nocturnal and prefer darkness, the animal has to suppress its innate tendency to enter the dark side by forming a learned association between the dark side and US. Animals that form such an association will cross over much later (have longer entrance latencies) than animals that remember poorly.

One important difference between Pavlovian fear conditioning and instrumental fear conditioning is the extent of amygdala involvement. For example, lesions or pharmacologic inactivation of the BLA impair IA whereas disrupting CeA function does not[44, 45] [As mentioned, the CeA is required for both contextual and auditory

Pavlovian conditioning][24]. In addition, the amygdala plays a direct role in consolidating Pavlovian fear memory, in contrast to operant conditioning where amygdalar influence is brief and less well defined. Furthermore, the number brain regions impacting IA exceeds those implicated in fear conditioning[44] (see below).

To date, most inactivation and lesion studies have interfered with IA consolidation, so this phase of mnemonic processing will be emphasized in the discussion that follows. In short, lesions and reversible inactivation by GABA receptor activation or sodium channel inactivation have demonstrated IA consolidation occurs over several hours and involves a multitude of brain regions in a coordinated and sequential fashion.

For example, functional integrity of the hippocampus, BLA, nucleus of the solitary tract

(NTS), and striatum is required during early memory consolidation (initially and up to 7

1.5 hours after training), whereas the integrity of brain regions including the nucleus basalis (NBM), substantia nigra (SN), and parabrachial nucleus (PBNC)—is required for several hours to days after training[44]. Remarkably, the entorhinal and parietal cortices are required for at least 1-2 months after acquisition[46]. Notably, the speed of consolidation (how long each region is necessary) depends on stimulus intensity, number of training trials and number of shocks administered[44]. The role of the hippocampus, amygdala and striatum in IA are discussed in detail below.

The dorsal and ventral hippocampus (DH & VH) plays a role in the initial phase of IA consolidation. Dorsal hippocampus integrity is required for 1.5 h post-conditioning while ventral activation is needed for 15 min[47, 48]. Functional differences between these two structures may account for different temporal requirements [Distinct roles for the DH & VH are also found in spatial navigation[49] (1.3.4)]. Likewise, the amygdala, particularly the BLA, is required for IA consolidation 0-1.5 h post-training. Infusion of lidocaine or tetrodotoxin (TTX) into the BLA, for example, reduces IA retention 1.5 but not 6 h after acquisition[50, 51]. Notably, functional inactivation of CeA does not impair consolidation of IA[51].

Several structures of the corpus striatum play a critical role in IA acquisition and consolidation. For example, permanent lesions of the caudate-putamen (CU)[52-54], nucleus accumbens (NAcc)[52, 55] and globus pallidus (GP)[56] impair the acquisition of IA. Likewise, TTX blockade of these regions 0.25-1.5 h post-training results in subsequent IA impairment[52]. Interestingly, functional inactivation of the substantia nigra (SN) impairs IA consolidation for a longer time period, up to 24 h post- conditioning[44]. 8

1.3.4 Spatial Navigation

`The Morris water maze (MWM) is the most commonly used task for assessing hippocampus-dependent spatial memory in rodents. It consists of placing a rodent in a pool filled with opaque water. To escape the pool, the animal learns to swim to a hidden platform located just below the surface of the water. Over several training trials the rodent learns to find the platform by using visual and spatial cues, including those strategically located on the walls outside of the maze. Distance swam, latency to reach the platform, time spent in each quadrant, number of platform crossings and swim speed are common measurements of this test.

Lesions and functional inactivation studies have implicated the necessity of several brain structures in various aspects of MWM learning or performance[57].

Notably, lesions of the amygdala do not impair water maze learning, but influence the degree of impairment exhibited when other brain regions are concurrently ablated or transiently inactivated[58-60]. It is well established that functional integrity of the hippocampus (particularly the DH) is critical for the acquisition, consolidation and retrieval of spatial information, as supported by temporary inactivation and neurotoxin lesion studies[61, 62]. Notably, the extent of memory impairment displayed by hippocampus-lesioned animals is directly proportional the volume of damaged tissue. DH lesions larger than 20% of the total hippocampal volume dramatically reduce spatial reference memory. In contrast, lesions of the VH do not significantly impair spatial navigation[49, 63]. In addition, lesion and inactivation experiments have implicated the striatum, basal forebrain, cerebellum and neocortex in different aspects of MWM

9 learning and sensorimotor function[57]. However, distinguishing between mnemonic and non-mnemonic contributions from these regions has been extremely difficult. A consensus is emerging that spatial learning in general and MWM performance in particular appear to depend upon the coordinated action of many brain regions and neurotransmitter systems constituting a functionally integrated neural network[57]. Table

1.1 summarizes brain regions involved in each of the paradigms described:

Cued Fear Contextual Fear IA MWM BLA BLA BLA DH CeA CeA DH Striatum* PAG* PAG* VH Basal Forebrain* DH NTS Cerebellum* VH NBM Cortex* NAcc PBNC mPFC SN Striatum Cortex

Table 1.1 Brain Structures Involved in Specific Tests of Aversive Memory * Involved in Memory and/or Motor Function

1.4 Role of NE/E &-adrenergic signaling in Fear Memory 1.4.1 Epinephrine & Enhanced Consolidation

The adrenergic catecholamine epinephrine (E) is released into the blood from the adrenal medulla following arousing or stressful situations. Systemic injection of E immediately after training produces dose-dependent effects on memory consolidation in rodents: Moderate doses enhance retention, whereas higher doses impair retention (e.g. an inverted U-shaped dose-response)[64, 65]. In addition, E increases consolidation for emotionally arousing material in human subjects[66]. Several theories attempt to explain 10

E-mediated effects on cognition. For example, evidence suggests that E can stimulate receptors in the liver to release glucose, and increased glucose levels can enhance memory consolidation in rats, mice and humans[64]. Furthermore, blocking associated rises in blood glucose levels prevent E-mediated enhancement of consolidation[67, 68].

Hence, glycogenolysis may play a role in mediating epinephrine influences on memory.

It is also known that E crosses the blood-brain barrier very poorly[69], and E’s effects on consolidation may be due, in part, to the activation of peripheral β-adrenergic receptors

(e.g. β receptors) on vagal afferents leading to the brainstem[11, 64, 70, 71]. Although there is limited data that β receptors are located on the vagus nerve, indirect evidence for this assertion comes from the finding that systemic administration of hydrophobic β blockers prevents the memory-enhancing effects of E[72]. In addition, E-mediated activation of the vagus nerve may increase NE centrally by stimulation of brainstem noradrenergic cell groups in the nucleus of the solitary tract (NTS) and locus coeruleus

(LC)[11, 73]. It is hypothesized that increased synthesis and release of NE in the CNS leads to the activation of β receptors within the BLA and other brain regions, and that stimulation of receptors within these structures is what leads to enhanced consolidation[11, 73]. Support for this comes from the observation infusion of β receptor antagonists into the BLA prevent enhancement of consolidation by systemically administered E[74].

1.4.2 Norepinephrine & Enhanced Consolidation

In the CNS, NE/E (adrenergic) neurons are located in 7 brainstem nuclei. The largest, the LC, provides NE input to most of the forebrain, including the amygdala and

11 hippocampus[75]. The adrenergic system is activated by unanticipated or emotional events[76], and multiple experimental outcomes demonstrate that moderate activation of this system can enhance the consolidation of fear memory[10, 77]. For example, infusing

NE directly into BLA, NTS or hippocampus before or immediately after training promotes memory consolidation[11, 73, 78]. However, few studies demonstrate that adrenergic signaling is actually required for the consolidation of fear memory[79-81].

1.4.3 NE/E is not Required for Fear Memory Consolidation

Pharmacologic and surgical approaches used to reduce or eliminate adrenergic signaling to study memory have been traditionally limited by two factors: their efficacy and/or their selectivity for NE/E. NE/E synthesis inhibitors, for example, either impair dopamine (DA) signaling (e.g. -methyl-p-tyrosine, a tyrosine hydroxylase inhibitor)(Figure 1.1) and/or inhibit other Cu2+/Zn2+-dependent enzymes in addition to dopamine -hydroxylase (DBH)[82-84]. Similarly, electrolytic and neurotoxic lesions of brainstem adrenergic neurons or their projections are incomplete, nonspecific or transient because of recovery mechanisms that occur in the remaining adrenergic neurons and terminals[85-89]. To circumvent these issues, mice with a targeted disruption of the gene for DBH (Dbh) were created (Figure 1.1). Knocking out this enzyme results in mutant

(Dbh-/-) mice, which lack the capacity to produce both NE and E[90, 91]. Multiple experimental outcomes from these animals demonstrate that NE/E is not required for the consolidation of fear memories. For example, Dbh-/- mice, compared to controls, exhibit equivalent short and long-term retention of IA and cued fear[92, 93]. Further, deficits exhibited by NE-deficient mice in contextual and spatial memory fully dissipate 5 or

12 more days after training[92]. Hence, consolidation processes have occurred in the absence of NE/E, leaving long-term fear memory intact.

Figure 1.1 Catecholamine Biosynthesis Pathway

1.4.4  Signaling is not Required for Fear Memory Consolidation

Analogous to increasing endogenous NE in the BLA by microinfusion, directly stimulating -adrenergic receptors in the BLA enhances fear memory consolidation[73].

Despite studies demonstrating that enhanced signaling can promote memory consolidation, there is a paucity of evidence showing that  signaling is required for fear memory consolidation. In fact, multiple experimental outcomes demonstrate the contrary, summarized as follows: 1) WT mice and rats treated with a non-selective  antagonist before or immediately after training exhibit normal contextual and spatial reference memory 5 days later[92]; 2) 1KO and 2KO mice express no deficit in IA (Chapter 2);

3) 1KO, 2KO and 12 DKO animals display no deficit in cued fear[92]; (Chapters 2,4)

1KO and 2KO mice express unimpaired contextual fear 5 or more days after training[92](Chapter 4); and 5) 12 DKO mice display unimpaired long-term memory of contextual fear, similar to their WT counterparts (Chapter 2). In sum, these results are inconsistent with a requirement for  signaling in fear memory consolidation.

1.4.5 NE/E is Required for Intermediate-term Memory Retrieval

1.4.5.1 NE/E and Spatial Retrieval

Dbh-/- mice exhibit impaired spatial reference memory 2 d (but not 2 h) after the last training trial, initially suggesting that NE/E is necessary for consolidation rather than the acquisition of this task[92]. Complicating interpretation of spatial navigation results is the extended training that is usually necessary for mice to acquire the task. Acquisition, consolidation, retrieval and reconsolidation occur repeatedly during such training, so it becomes difficult to assign deficits to a single stage of learning.

1.4.5.2 NE/E and Contextual Memory Retrieval

A major advantage of examining contextual fear, in addition to spatial navigation, is that robust acquisition can occur in a single trial, and this allows examination of acquisition, consolidation or retrieval in isolation by manipulating any single stage of learning at a particular time. Interestingly, contextual memory retrieval was impaired from two hours to several days after training, which was initially consistent with a possible role for NE in hippocampus-dependent memory consolidation, but a more thorough analysis revealed that this was not the case.

An important component of the Dbh-/- model is the ability to rapidly restore NE in the mutant mice either before or after training (or prior to testing) by administering a synthetic amino acid precursor of NE, L-threo-3,4-dihydroxyphenylserine (L-DOPS,

Figure 1.1)[90]. L-DOPS restores NE tissue content to near normal levels for ~12 h, with

NE levels peaking approximately 5 h after injection[90]. Remarkably, restoring NE before or immediately after training did not rescue memory consolidation because 2 days after L-DOPS administration, Dbh-/- mice continued to express deficits in contextual fear.

In contrast, administering L-DOPS before testing 1 or 2 days after training fully reversed deficits exhibited by these animals[92]. In addition, Dbh-/- mice exhibited normal contextual memory 5 d after training without any treatment, arguing against a requirement for NE in fear memory consolidation[92]. In sum, experimental outcomes from both paradigms demonstrate a critical, transient role for NE in the retrieval of fear memory.

1.4.6  Signaling is Required for Intermediate-term Memory Retrieval

1.4.6.1  Signaling and Spatial Retrieval 15

Many studies exploring the role of NE and -adrenergic signaling in fear memory have utilized rats, so experiments were conducted to test whether spatial navigation deficits displayed in NE/E-deficient mice also applied to rats whose adrenergic signaling was disrupted by administration of a non-selective  antagonist. [Using rats in the water maze possessed an added advantage because rats, unlike mice, can acquire this task rapidly (e.g. 4 blocks of 4 trials over 90 min)[94], so specific stages of learning can be better manipulated and dissected]. Systemic injection of the  antagonist, propranolol, produced no effect on spatial reference memory in rats when given shortly before training, or shortly before testing 1 h or 1 wk after training[92]. However, rats exhibited reduced spatial bias 1 d after training, when the  antagonist was administered shortly before testing, providing evidence that -adrenergic signaling is required for intermediate-term retrieval of spatial reference memory across species[92].

1.4.6.2  Signaling and Contextual Memory Retrieval

Administration of propranolol or either of several 1-adrenergic receptor antagonists to control mice before testing contextual fear reproduces retrieval deficits

-/- -/- expressed by Dbh mice[92]. In addition, treating Dbh mice with a 1 agonist just before testing completely reverses contextual freezing impairments[92]. These observations also hold true in rats, as 1 antagonists yield similar results[92].

Interestingly, these findings suggest that 1-signaling is both necessary and sufficient for hippocampus-dependent memory retrieval mediated by NE. However, it does not exclude the possibility that constitutive signaling by other adrenergic receptors in the absence of

NE can influence retrieval (Chapter 2), nor does it exclude the possibility that other adrenergic receptors can signal in similar or opposing ways to produce affects on retrieval in addition to 1 signaling (Chapters 2, 4).

1.4.6.3  & Receptors Influence Retrieval Independent & in Concert with 

Chapters 2-4 provide support that NE/E exerts some effects on the retrieval of hippocampus-dependent fear memory by activating other  receptors, alone or in conjunction with 1. In general,  receptors can be classified into 3 subtypes based on pharmacologic specificity and the intracellular signaling pathways they modulate[95]

(Table 2.1). IHC, ISH and radioligand binding studies have demonstrated that  receptors have overlapping, but somewhat unique expression patterns throughout the CNS[96-100].

Subtype 1   Agonists Xamoterol Procaterol, Zinterol CL 316243 Antagonists Betaxolol, CGP ICI 118551 SR 59230A

G protein Gs Gs, Gi/o Gs, Gi/o Effectors Adenylyl Cyclase Adenylyl Cyclase Adenylyl Cyclase cAMP cAMP cAMP

Table 1.2 -adrenergic Receptors, their Pharmacology and Signaling

1.5 Role of Stress & Glucocorticoids in Fear Memory

1.5.1 Stress, Glucocorticoids (GCs) & Consolidation

GCs are stress hormones released from the adrenal cortex that penetrate the blood-brain barrier and bind to two types of adrenal steroid receptors, glucocorticoid receptors (GRs) and mineralocorticoid receptors (MRs), the latter of which have higher

17 affinity for glucocorticoids[101]. Under stressful circumstances, increases in circulating

GCs cause low-affinity GRs to become increasingly saturated, and it is these receptors that are believed to be primarily involved in mediating GCs effects on memory [59, 102-

107], although some studies suggest a role for MR as well[108]. Like NE and  signaling, stress and GCs have been shown to enhance long-term memory consolidation in both humans and animals when administered shortly before or after training [109-112].

Moreover, CNS infusion of GR agonists and antagonists in rodents produce divergent effects on consolidation. For instance, systemic injection or direct infusion of GCs (or

GR agonists) into the NTS, BLA or hippocampus augments consolidation while administration of GR antagonists impairs consolidation[59, 103, 113-115]. Traditionally, it was believed that GCs influence consolidation by binding to intracellular and intranuclear receptors which affect gene transcription by binding of receptor homodimers to DNA or via protein-protein interactions with other transcription factors[116-118].

Alternatively, it been demonstrated that GCs can interact with membrane-associated steroid receptors[119-121] and these receptors may influence consolidation by rapidly altering cAMP-PKA signaling[11, 122]. For instance, infusion of GCs into the BLA is believed to facilitate -cAMP-PKA signaling, and infusion enhances the consolidation of fear memory[11, 122].

1.5.2 Stress, GCs & Retrieval

Stress and GCs have been shown to impair short-term and long-term memory retrieval in humans and animals[123-125]. For example, administration of stress or GCs shortly before testing impairs the retrieval of spatial reference memory[123], contextual

18 fear (Chapter 2), IA[126] and human declarative memory[127]. Effects of GCs on retrieval are often transient, waning several hours after stress exposure or hormone injection[128]. Similarly, experiments in Chapter 2 demonstrate that the influence of acute stress and GCs on retrieval dissipates several days after acquisition, even if GCs are administered shortly before testing (Chapter 2). In contrast, prolonged exposure to high levels of GCs, chronic stress and corresponding HPA dysfunction can cause long-lasting impairments of declarative/episodic memory[129]. However, these deficits are probably not retrieval-based, but reflect impaired consolidation processes. As opposed to processes underlying GC-mediated consolidation, GC processes pertaining to retrieval may be regulated solely through rapid, non-genomic mechanisms[130, 131].

1.6 Interactions between Adrenergic & GC Signaling in Fear Memory

1.6.1 NE/E,  Receptor & GC Interactions in Consolidation

Blocking the synthesis of E and GCs by removing or suppressing the adrenals both cause deficits in fear memory consolidation[132, 133]. Also, the consolidation effects of E are critically influenced by GCs. For example, the memory-enhancing effects of E (or stress) are blocked by metyrapone, a drug that prevents the synthesis and release of glucocorticoids[132, 134]. Glucocorticoid-based effects on consolidation likely involve concurrent noradrenergic activation within the BLA[128]. For example, propranolol-mediated block of β-adrenoceptors within the BLA prevents the memory- enhancing effects of systemic or intra-NTS GC administration[115, 135]. Similarly, infusions of receptor antagonist into the BLA shortly before IA training blocks the enhancement of memory consolidation brought about by systemic administration of

19 dexamethasone, a GR agonist[136]. Likewise, infusion of the antagonist atenolol into the BLA prevents the memory enhancing effects of a GR agonist infused into the hippocampus[104]. Moreover, the enhancement in IA consolidation mediated by stimulation of GRs in the NTS is prevented by infusion of atenolol into the BLA[115].

A common hypothesis is that GCs enhance memory consolidation by increasing the synthesis and release of NE within the BLA and other brain regions[11, 137]. Some indirect evidence supports this observation. For example, systemic administration of corticosterone after IA training increases NE levels in the BLA[138]. GCs have also been shown to block a specific catecholamine-transporter in glial cells, known to rapidly remove extracellular NE[139].

1.6.2 NE/E, Receptor & GC Interactions in Retrieval

Co-administration of a antagonist prevents GC-mediated retrieval impairment of IA[140]. Similarly, infusion of a  antagonist into the hippocampus or BLA mitigates the impairing effects of concurrent intra-hippocampal administration of GR agonist on spatial reference memory[141]. Interestingly,  antagonists infused into the BLA also block the memory retrieval impairing effects of a GR agonist infused into the hippocampus[141]. Further, it has been reported that DH infusion of a  antagonist blocks GR-agonist mediated deficits in spatial navigation and that systemic administration of a partial  agonist, xamoterol, mimics GC-mediated effects on memory retrieval[141]. However, evidence from our lab suggests that GC effects on retrieval actually depend on  rather than  signaling (Chapter 2).

1.7 GCs &  Signaling can Influence cAMP-PKA Signaling

1.7.1 cAMP-PKA & Fear Memory Consolidation

It has been suggested that the consolidation of hippocampus-dependent memory modulated by NE and GCs involves common neurobiologic substrates and signal transduction pathways, including cAMP and protein kinase A (PKA) signaling. Three decades of research across species illustrate that agents that elevate cAMP usually enhance memory consolidation, whereas agents that block cAMP signaling impair consolidation[142-146]. In addition, pharmacologic and genetic results in several species illustrate that cAMP-mediated PKA signaling is necessary for the consolidation of long- term fear memory[3]. For example, transgenic mice expressing a cAMP-insensitive form of a regulatory subunit of PKA exhibit deficits in spatial and contextual memory[147]. In addition, ICV or intrahippocampal administration of a PKA inhibitor disrupts IA consolidation [3, 144, 148, 149]. A major effector of PKA is cAMP response element binding (CREB) protein and transgenic mice lacking specific isoforms of CREB, dominant-negative CREB mutants and mice expressing truncated forms of CREB- binding protein (CBP) display impairments in spatial, contextual and auditory fear memory[3, 150-152].

GCs appear to promote cAMP-PKA signaling in order to enhance memory consolidation[122]. For example, BLA infusion of the cAMP-dependent PKA inhibitor,

Rp-cAMPS prevents GR agonist-mediated enhancement of IA[122]. However, infusion of a GR antagonist into the BLA does not prevent the augmentation of memory consolidation produced by concurrent injections of the cAMP activating analog, 8-Br-

21 cAMPS[122]. Hence, cAMP-PKA mediated enhancement of consolidation could be either 1) independent of GC signaling and/or 2) GCs effects occur upstream of cAMP, perhaps on NE release and/or activation of adrenoceptors[11, 122].

1.7.2 cAMP-PKA & Fear Memory Retrieval

Recent evidence suggests PKA signaling is involved in hippocampus-dependent memory retrieval. For example, activation of both PKA and exchange protein activated by cAMP (Epac) is required for CF retrieval[153]. Likewise, infusion of PKA inhibitors into the DH shortly before testing transiently impairs the retrieval of contextual information[153]. Further, CA1 infusion of a PKA inhibitor (Rp-cAMPS) impairs retrieval of IA[154]. Interestingly, restoring PKA-Epac signaling fully reverses contextual retrieval deficits exhibited by mice lacking NE/E, indicating the PKA-Epac signaling is necessary and sufficient for NE’s role in hippocampus-dependent retrieval[153].

It should be noted, however, that cAMP effects on consolidation and/or retrieval may not always be dependent on PKA signaling. Most of these agents described inhibit or restore cAMP signaling (e.g. effects are not-specific to PKA) and other targets of cAMP could possibly contribute (e.g. ion channels).

It has been suggested that GCs facilitate -adrenergic signaling to inhibit retrieval, but how this occurs is inconclusive. My studies demonstrate that stress and GCs

enhance  signaling in order to impair retrieval, which, in turn, leads to a reduction in cAMP-PKA signaling that is necessary for hippocampus-dependent recall (Chapter 2).

1.8 Scope of Thesis Research

The major aim of this thesis work is to examine interactions between stress, GCs,

NE/E and -adrenergic signaling in aversive hippocampus-dependent memory retrieval.

Pharmacological and genetic manipulations of NE and  adrenoceptors were utilized to access adrenergic contributions to stress- and GC-mediated retrieval impairment in vivo, using behavioral paradigms specifically designed to test both hippocampus-dependent

(e.g. CF, IA) and –independent (e.g. cued) fear memory. Corticosterone and cAMP radioimmunoassays were employed that provided molecular analysis of GC, stress and  receptor effects that correlated well with behavioral results. Systemic and/or intrahippocampal injections using a wide range of doses produced consistent results across preparations in both mice and rats. Conclusions justified from the data will significantly expand scientific knowledge about the complex interplay between  and GC signaling in retrieval while casting doubt on two prevalent but seldom tested hypotheses

(e.g. that NE and  signaling are required for GC-mediated effects on retrieval). In addition, initial results begin to define a molecular mechanism for how GCs, GRs, and 2-

ARs intersect to impair retrieval. Furthermore, evidence is presented that characterizes novel and potentially important roles for 2 and 3 receptors in aversive learning and memory, whose functions can be dependent and independent of 1 signaling.

1.9 Clinical Significance of this Work

1.9.1 Adrenergic & GC Therapies in Post-traumatic Stress Disorder (PTSD)

One of the cardinal symptoms of PTSD is the excessive retrieval and reconsolidation of traumatic memories. During the last decade, a growing number of clinical studies have administered adrenergic and GC-related compounds in attempts to treat the disease. This section highlights some of this work and attempts to explain why outcomes have been inconsistent. Furthermore, limitations of adrenergic- and GC- based therapies in reducing the frequency of traumatic memories are discussed.

1.9.2 -adrenergic Antagonists & Agonists for PTSD

Evidence suggests that -adrenergic antagonists, such as prazosin, may be beneficial in treating specific symptoms of PTSD, such as nightmares and other difficulties sleeping[155-157]. In addition, clonidine, an -adrenergic agonist, has been shown to be effective in reducing symptoms of hyperarousal and startle associated with the disease[158]. However, these compounds are largely ineffective in treating most other symptoms of PTSD, such as the recurrent, intrusive retrieval of traumatic memories.

1.9.3 -adrenergic Antagonists & Agonists for PTSD

1.9.3.1 Reducing Consolidation

Horrific events often leave indelible marks in human consciousness, permanent scars that may never heal. Heightened arousal and release of stress hormones after such harrowing experiences can result in over-consolidated memories which are highly resistant to degradation. Further, cues reminiscent of the trauma are believed to increase the release of stress hormones which cause further over-consolidation, promoting a vicious cycle that contributes to the development/expression of PTSD[159]. In rodents,

-adrenergic blockade immediately after aversive experiences reduces the enhancement of consolidation caused by stress (1.6.1). Based on this finding, studies were conducted in humans to investigate whether  blockers (e.g. propranolol) might assist in the secondary prevention of PTSD by interfering with consolidation processes believed to occur hours to days after the event. Initial studies supported this hypothesis, as individuals treated with propranolol 0-4 h after a motor vehicle accident show reduced PTSD symptoms weeks to months after the event[160]. Likewise, adults who recently experienced one of two forms of trauma (e.g. a motor vehicle accident or physical abuse) also found long- lasting benefit of propranolol administration 2-20 h after the trauma[161]. Notably, propranolol was continually administered for several days after initial dosing, so improvements due to receptor blockade cannot be firmly attributed to interfering with consolidation.

Interestingly, recent studies consistently fail to show a benefit for propranolol soon after experiencing any one of several forms of trauma (e.g. burns, severe injury)[162-164]. These results agree with rodent data produced by our lab and others demonstrating that  blockade does not impair the consolidation (e.g. ―strength‖) of fear memories per se[165] (also see 1.4.4). For example, consolidation of cued fear is unaffected by propranolol or NE/E deficiency and cues associated with the trauma often trigger the strongest recall for the event.

1.9.3.2 Reducing Retrieval

Interestingly, propranolol impairs the retrieval of hippocampus-dependent fear memory in rodents[92]. Conversely, evidence from humans demonstrates that 25 administration of propranolol shortly before testing improves SAT exam performance in anxiety-ridden students and increases cognitive flexibility and problem-solving ability in nervous examinees[166-168]. It should be noted that in the animal study, propranolol is administered before re-exposure to a stressful environment, while in the human study, propranolol is administered to subjects that are already stressed. It is theorized that

receptors become increasingly activated during stress, and propranolol-mediated blockade of -ARs could prevent stress-induced impairment of retrieval, similar to effects of antagonists in rodents (Chapter 2). In contrast, propranolol has been demonstrated to block  signaling that is necessary for retrieval[92]. One explanation for conflicting results is that  signaling can oppose the role of  signaling in retrieval

(Chapter 2), so utilizing a non-selective  antagonist might lead to inconsistent results.

Instead, administration of a selective 1 antagonist or 2 agonist might be more efficacious in reducing PTSD memories, given knowledge about the molecular mechanisms underlying the retrieval of hippocampus-based fear memory[92] (Chapter 2).

However, this too might be unproductive as our lab has demonstrated that propranolol’s effectiveness in reducing retrieval is limited by the age of the memory[92]. Similarly, the utility of  antagonists might wane over time because hippocampus-dependent retrieval becomes independent of adrenergic signaling[92].

1.9.3.3 Reducing Reconsolidation

Instead, the utility of propranolol for PTSD might be in interfering with mnemonic processes that occur following retrieval, such as reducing the reconsolidation of fear memories (or promoting extinction) [169]. Initial evidence suggests that 26 medicating patients with  blockers for short stretches when symptoms are already prevalent (or after reexposure therapy) might be beneficial because memories may become labile again, i.e. they may be strengthened or weakened. Animal studies support this hypothesis. For example, systemic or BLA blockade of  receptors following reactivation of fear memory disrupts reconsolidation and impairs long-term auditory fear memory in rats[165, 170]. Moreover, administration of systemic propranolol following reactivation produces long-term impairment of IA[171]. Interestingly, such studies suggest that propranolol can impair the reconsolidation of both hippocampus-dependent and -independent fear memories; translationally, this might lead to better efficacy, given that PTSD populations are heterogeneous. Notably, -adrenergic blockade following retrieval in humans reduces memory of emotionally enriched words (declarative memory) for at least several days after administration[172]. In contrast, others have shown less benefit for propranolol in inhibiting the reconsolidation of traumatic memories. For example, a recent study in rodents demonstrates that propranolol may only block the reconsolidation of specific types of fear memory[173]. Moreover, administration of propranolol after reactivation of conditioned fear in humans produces a long-lasting reduction in the fear response but not in declarative memory for the fear association itself[174]. Thus, the utility of propranolol to inhibit reconsolidation of traumatic memories has yet to be determined.

1.9.4 Glucocorticoids for PTSD

1.9.4.1 Reducing Consolidation

Like  blockers, administering GCs soon after a traumatic experience may reduce the onset or severity of PTSD by blocking the consolidation of fear memories. For example, individuals receiving high doses of GCs after major surgery or septic shock display reduced recall of events associated with the trauma[175-177]. Notably, hydrocortisone administration was continued for several days after initial dosing, so improvements cannot fully be attributed to affecting any single stage of memory (e.g. consolidation). Benefits of GCs in PTSD patients is at odds with the well-replicated finding that GCs enhance the consolidation of fear memories in both animals and humans(see 1.5.1). One alternative explanation is that endogenous levels of GCs are lower in patients that develop PTSD[178, 179], and increasing cortisol after a horrific experience may restore signaling to normal levels. It is hypothesized that restoring cortisol function attenuates the exaggerated stress response (e.g. through negative feedback inhibition) in individuals who are prone to develop the disorder[180, 181].

1.9.4.2 Longer-term GC Administration can Impair Memories Unrelated to the Trauma

Long-term exposure to high levels of GCs can produce cognitive deficits in episodic memory that can impair everyday functioning[182] and may create impairments more debilitating than the disease itself. In fact, evidence suggests that long-term GC exposure or stress may cause permanent hippocampal damage via cumulative and long- lasting changes in synapse function and morphology which negatively affect declarative memory[183-185]. In addition, chronically elevated GC levels that are associated with either long-term treatment or HPA dysfunction can exacerbate existing psychiatric illnesses and increase the risk for major depression and psychosis[186]. 28

1.9.4.3 Reducing Retrieval by Long-term, Low-dose GC Administration

Clinical studies have attempted to circumvent adverse affects of chronic, high- dose GC administration by employing low doses of GCs to treat PTSD patients over extended times [12, 188] (such doses have not been shown to create cognitive deficits).

Interestingly, evidence from our lab and others support initial clinical evidence that low doses might be more efficacious in inhibiting the retrieval of unwanted memories[12,

188](Chapter 2). However, initial studies that chronically administer GCs have had limited success, only reducing select symptoms of the disease that vary between individuals (e.g. nightmares, hyperarousal, feelings of re-experiencing the event)[187].

Moreover, a significant treatment effect in reducing the retrieval of traumatic memories has not been reported. The latter result is consistent with findings by our lab demonstrating that GCs impair the retrieval of fear memories for a limited time in rodents(Chapter 2), so long-term clinical administration might be ineffective, because retrieval processes may, over time, become insensitive to GC manipulation.

1.9.4.4 Reducing Reconsolidation

Utilizing GCs to interfere with reconsolidation processes (or promote extinction) may be more beneficial that utilizing GCs to interfere with consolidation or retrieval. In theory, administering GCs during or immediately after reexposure (or after recreating the trauma) may weaken the memory trace and thus reduce PTSD symptoms beyond the treatment period. Preliminary studies in animals and humans suggest this might hold true.

In rodents, administration of a GC after memory reactivation disrupts subsequent recall of context fear 2 weeks later [188, 189]. Similarly, combat veterans shown images

29 reminiscent of wartime trauma immediately preceding GC administration showed reduced PTSD symptoms 1 wk later[190]. However, benefits of GCs following reactivation may wane over time (similar to affects on retrieval), as veterans showed little benefit when tested 1 month later[190].

Paradoxically, administration of GR antagonists into the BLA after retrieval also cause impairments in IA and cued fear that persist for at least several days[191, 192]. At first pass, it seems logical that blockade of GR signaling impairs reconsolidation, because

GR signaling is required for consolidation of fear memory. It is possible that deficits mediated by GCs and GR antagonists are the result of opposing effects on reconsolidation and extinction[169, 193], similar to opposing functions demonstrated in the consolidation and retrieval of aversive memories. In other words, GCs may promote extinction (or promote the consolidation of corrective responses) while GR antagonists may inhibit the over-consolidation of fear memories believed to occur after re-exposure to the trauma.

1.10 Statement of Purpose

A major goal of this thesis was to resolve the conflicting observations that NE and 1 signaling are required for contextual and spatial memory retrieval (in mice and rats) but also necessary for stress and glucocorticoid impairment of hippocampus-dependent emotional memory[92, 141]. Because GC deficits are completely blocked by a non- selective  antagonist[141], we examined whether another adrenergic receptor (2) was responsible for stress/GC inhibition of episodic memory (Chapter 2). A second goal was to confirm the validity of a study suggesting that 1 and 2 are both required for the 30 consolidation of fear memory[194]. Our results are consistent with most studies that have failed to demonstrate a requirement for either receptor in the consolidation of Pavlovian fear conditioning (e.g. propranolol does not impair auditory fear; cued fear and IA are normal in 1KO, 2KO, 12 DKO mice)[92, 165](Chapter 2). A third aim was to refute a common notion that all CNS  receptors function in a similar ways in aversive learning and memory. For example, it was reported that GC enhancement of consolidation can be prevented by either 1 or 2 blockade, suggesting a redundant role for these receptors in auditory fear learning[136]. In contrast, experiments presented in the chapters that follow demonstrate that  receptors play highly different roles in retrieval, which parallels mounting evidence for distinct roles for the two receptors in stages of fear memory such as consolidation (communications with M. Ouyang & S.A. Thomas). A final goal of this work was to demonstrate that divergent roles of receptors can be mediated through opposing effects on intracellular signaling pathways, such as cAMP.

Chapter 2

Stress and Glucocorticoids Impair Hippocampus-dependent Emotional Memory Retrieval via Gi/o-coupled2- adrenergic Signaling

Keith Schutsky, Ming Ouyang, Christina B. Castelino, Lei Zhang & Steven A. Thomas

Department of Pharmacology, University of Pennsylvania, Philadelphia, PA 19104

Author contributions: K.S. and S.A.T. designed research; K.S., M.O., C.C., L.Z and S.A.T. performed research; K.S., M.O., C.C. and S.A.T. analyzed data; K.S. and S.A.T. wrote the paper.

2.1 ABSTRACT

Stress and glucocorticoids impair the retrieval of hippocampus-dependent emotional memory. It has been proposed that glucocorticoids act by promoting the release of norepinephrine (NE), which impairs retrieval by stimulating 1-adrenergic receptors and cAMP signaling. In contrast, evidence indicates that NE, 1 and cAMP signaling are transiently required for the retrieval of hippocampus-dependent emotional memory. Here we report that glucocorticoids do not require NE to impair retrieval.

Despite this, stress- and glucocorticoid-induced impairment of retrieval depends on the activation of 2-adrenergic receptors (not 1). Importantly, the effects of stress, glucocorticoids and 2 agonists are blocked by pertussis toxin, suggesting that they act through Gi/o proteins. In support of this, 2 signaling in isolation decreases cAMP in the hippocampus, and when combined with 1 signaling, prevents the increase in cAMP elicited by 1 receptors. The results demonstrate that 1 and 2 receptors can have quite distinct roles in CNS signaling and in cognition.

2.2 INTRODUCTION

Stress and glucocorticoid hormones (corticosterone in animals, cortisol in humans) can have pronounced effects on cognition[127, 195, 196]. Glucocorticoids activate mineralocorticoid (MR) and glucocorticoid (GR) receptors, both of which are prevalent in the hippocampus[101, 197]. Glucocorticoids preferentially occupy MR under basal physiologic conditions; however, glucocorticoid levels rise significantly following a stressor, causing substantial additional activation of GR[101]. The impact of

33 glucocorticoids on mnemonic processes can vary considerably, depending upon when and to what extent levels of these hormones are altered and the specific stages of learning that are examined. When administered shortly after training, glucocorticoids enhance the consolidation of hippocampus-dependent emotional memory[103, 110], and a GR antagonist administered shortly after training impairs hippocampus-dependent emotional memory consolidation[102, 103, 198].

Conversely, administration of stress or glucocorticoids shortly before retention testing impairs the retrieval of hippocampus-dependent emotional memory in animals and humans[123, 124, 199]. Infusion of the GR-selective agonist RU 28362 into the dorsal hippocampus (DH) 60 min before retention testing impairs retrieval[107], and viral expression of a transdominant GR in the dentate gyrus prevents retrieval impairment by corticosterone (cort)[106], strongly suggesting that retrieval impairment is mediated by

GR. However, a role for MR is also possible. Intracerebroventricular administration of the MR antagonist spironolactone is reported to mitigate cort-induced impairment of spatial reference memory[108].

Accumulating evidence suggests that glucocorticoids and other steroid hormones can operate through non-genomic as well as genomic mechanisms[200, 201]. Generally, genomic mechanisms affect protein expression occurring over tens of minutes to hours.

In contrast, non-genomic mechanisms are much more rapid, occurring within seconds to minutes. Cort likely affects memory retrieval through non-genomic, membrane- delimited mechanisms. Intracerebroventricular administration of anisomycin, a protein synthesis inhibitor, does not alter cort-induced impairment of spatial navigation in rats[131], and DH infusion of protein-conjugated cort, which selectively activates 34 membrane-associated receptors[202, 203], mimics the effects of stress on serial memory retrieval[204].

Offering a potentially rapid, non-genomic mechanism for glucocorticoid action, results from some studies suggest that both the impairment of retrieval and the enhancement of consolidation by cort depend on signaling by the adrenergic neurotransmitters norepinephrine and epinephrine (NE/E). -adrenergic signaling, in particular, may be critical. Administration of the  receptor antagonist propranolol prevents glucocorticoid-induced impairment of retrieval in rats and humans[140, 141,

205]. Further, DH infusion of the 1-selective receptor antagonist atenolol blocks the impairing effects of the GR agonist RU 28362 in the Morris water maze, and systemic administration of the 1-selective agonist xamoterol mimics the effects of RU

28362[141].

The suggestion that 1 signaling is responsible for the impairing effects of glucocorticoids on retrieval is at odds with the recently described requirement for 1 signaling in hippocampus-dependent emotional memory retrieval. Targeted disruption of the dopamine -hydroxylase gene (Dbh-/-) in mice results in NE/E deficiency and deficits in the Morris water maze and contextual fear conditioning that reflect impaired memory retrieval[92, 93]. Administration of the 1-selective agonist xamoterol shortly before

-/- testing rescues retrieval in Dbh mice, suggesting that 1 signaling is sufficient for the role of NE/E in retrieval[92]. In support of this, mice with targeted disruption of the gene for the 1 receptor (1 KO) display deficits in contextual fear memory, and systemic administration of the 1-selective antagonist betaxolol produces similar retrieval deficits

35 in wild-type mice and rats[92]. Finally, these effects are mimicked by infusion of the 1- selective antagonist atenolol into the DH.

The goal of this study, then, was to resolve the seemingly conflicting observations that 1 signaling is both required for hippocampus-dependent emotional memory retrieval and responsible for the impairment of such retrieval by glucocorticoids. We hypothesized that the pharmacologic characterization of the glucocorticoid-induced retrieval impairment is incomplete, and that another adrenergic receptor may mediate this effect. We further hypothesized that the intracellular signaling activated by this receptor may oppose the intracellular signaling activated by 1 receptors. Toward this goal, animals were tested in both Pavlovian and instrumental fear conditioning paradigms.

Pharmacologic reagents previously untested in these paradigms were administered to wild-type rats and mice, as well as to mice genetically modified to lack either NE/E, 1- or 2-adrenergic receptors. Because glucocorticoids are thought to underlie the impairment of hippocampus-dependent memory retrieval by stress, additional experiments examined the effects of stress on retrieval. The results provide substantial and unexpected support for the idea that signaling by 2-adrenergic receptors in the hippocampus opposes signaling mediated by 1 receptors at both the molecular and behavioral levels. This finding is inconsistent with the classic view that both  receptors couple to Gs proteins and elevate levels of cAMP, but is consistent with the divergent signaling mechanisms and roles for these two receptors in the heart[206, 207].

2.3 RESULTS

2.3.1 Corticosterone and a GR Agonist Impair Hippocampus-dependent Memory Retrieval

Most previous animal studies examined the impact of stress or glucocorticoids on the retrieval of instrumental fear memory (inhibitory avoidance) or spatial reference memory (Morris water maze). For comparison to previous studies, inhibitory avoidance was used in the current study. In addition, we asked whether the retrieval of Pavlovian contextual fear memory, like spatial reference memory, would be sensitive to glucocorticoids. Potential advantages of studying contextual fear memory include its robustness, rapid acquisition, and the use of dorsal hippocampus-independent cued fear memory as a control for potential effects on performance.

When cort was administered subcutaneously 30 min prior to testing retrieval one day after training, mice had lower avoidance latencies and reduced contextual freezing in these two paradigms (Figures 2.1A and 2.1B). In both, deficits produced by cort followed a U-shaped dose-response over a 10-fold dose range, with 1 mg/kg cort producing maximum impairment. To test whether cort might act through a GR-like mechanism, the selective GR agonist fluticasone was administered before testing contextual fear[208]. Like cort, fluticasone impaired contextual freezing in a U-shaped manner (Figure 2.1C).

To corroborate the finding that glucocorticoids act in the hippocampus to impair retrieval, bilateral cannulae targeting the DH were implanted and mice were fear conditioned one week later. One day after training, either vehicle or cort was infused bilaterally into the DH 15 min before testing. There was a dose-dependent decrease in

37 contextual freezing that, at 50 ng, was similar to the magnitude observed after systemic treatment (Figure 2.1D). Unlike systemic treatment, however, the effect of cort over a

10-fold dose range in the DH was not U-shaped, with 100 ng producing a similar effect to that for 50 ng. In support of a non-genomic mechanism of action, DH infusion of 20 ng cort 7.5 min prior to testing contextual fear produced the same degree of impairment as

20 ng infused 15 min prior to testing (Figure 2.1E).

Finally, to determine whether glucocorticoid impairment of contextual fear was specific to hippocampus-dependent memory retrieval, cort was administered prior to testing cued fear. Neither systemic nor DH cort impaired cued fear expression, indicating that these treatments do not alter fear, freezing or hippocampus-independent memory retrieval (Figure 2.1F).

A Avoidance Context 8 B 80

6 60 * # # ^ 4 40 # ^ ^

2 20

Freezing (%) Freezing Latency (min) Latency 0 0 0 0.5 1 2 3 0 0.25 0.5 1 2 3 Cort (mg/kg) Cort (mg/kg) Context Context C 80 D 80 60 * 60 ^ # # # 40 40 # #

20 20

Freezing (%) Freezing Freezing (%) Freezing 0 0 0 0.2 0.5 1 2 0 10 20 50 100 Fluticasone (mg/kg) DH Cort (ng @ 15 Min) E Context F Cue 80 80 SC DH 60 60

40 ^ 40

20 20

Freezing (%) Freezing (%) Freezing 0 0 0 20 0 1 0 50 DH Cort (ng @ 7.5 Min) (mg/kg) Cort (ng)

Figure 2.1 Corticosterone and a GR Agonist Impair Hippocampus-dependent Emotional Memory Retrieval

(A and B) Systemic corticosterone (cort) disrupts retrieval of inhibitory avoidance (IA) and contextual fear (CF). P < 0.05 for the main effect of dose on avoidance (9-18/group), and P <

0.001 for the main effect of dose on freezing (4-7/group). For all figures: systemic injection occurred 30 min prior to testing one day after training unless stated otherwise; * P < 0.05, ^

P < 0.01, # P < 0.001 for post-hoc comparisons. (C) Fluticasone, a GR agonist, mimics the 39 impairment of CF by cort. P < 0.001 for the main effect of dose (5-6/group). (D) Infusion of cort into the dorsal hippocampus (DH) impairs CF. P < 0.001 for the main effect of dose (5-

6/group). For all figures, DH infusions occurred 15 min prior to testing one day after training unless stated otherwise. (E) Cort impairment of CF is rapid, suggesting a non-genomic mechanism. Cort was infused 7.5 min prior to testing one day after training (5/group). (F)

Neither subcutaneous (SC) injection nor DH infusion of cort affects cued fear (5-8/group).

2.3.2 Impairment of Retrieval by Glucocorticoids Depends on 2 Signaling but not NE/E

Several studies have suggested that glucocorticoid-induced impairment of memory retrieval may depend upon the release of NE in the CNS[115, 140, 141]. To test this idea, wild-type and NE/E-deficient Dbh-/- mice were administered cort systemically

30 min prior to testing retrieval one day after training. Instrumental fear conditioning was used because, unlike the retrieval of contextual fear that depends on NE/E, the retrieval of instrumental fear is normal in Dbh-/- mice[93]. Importantly, the degree of cort-induced impairment displayed in Dbh-/- mice was similar to that seen in control mice, indicating that NE/E are not required for the impairing effects of glucocorticoids on retrieval (Figure 2.2A).

Previous results using the 1-selective antagonist atenolol and the 1-selective agonist xamoterol suggest that disruption of memory retrieval by glucocorticoids depends on 1 signaling[141]. Those findings are in direct contrast to observations indicating that

1 signaling is necessary for the retrieval of an intermediate phase of contextual and spatial memory[92]. One possible resolution for this discrepancy is that atenolol and xamoterol affect receptors in addition to 1 at higher doses, and that this other receptor is responsible for the effects of these drugs on retrieval. A likely candidate for this receptor 40 is the 2-adrenergic receptor, which is closely related to the 1 receptor and is also relatively abundant in the CNS.

To distinguish between effects mediated by 1 and 2 receptors, we initially administered either the 1 antagonist betaxolol, the 2 antagonist ICI 118,551 (ICI), or a combination of cort and the 1 or 2 antagonist to wild-type mice shortly before testing instrumental or contextual fear. While having no effect on retention latencies for instrumental fear, betaxolol impaired contextual fear as expected, indicating that the dose used was sufficient to block 1 receptors (Figures 2.2B and 2.2C). At the same dose, the

2 antagonist ICI did not alter memory retrieval in either test. Notably, co-injection of cort with the 1 antagonist did not prevent cort-induced impairment of retrieval in either paradigm. In contrast, co-administration of cort with the 2 antagonist completely mitigated cort-induced disruption of memory retrieval in both paradigms.

To determine whether ICI non-specifically elevates freezing when freezing is relatively low, wild-type mice were conditioned using a less intense shock (0.35 mA) and saline or ICI was administered 30 min prior to testing. Contextual fear did not differ between the groups, indicating that ICI does not enhance fear or freezing per se (Figure

2.2D). Finally, because cort acts in the DH to impair retrieval, we predicted that ICI, when co-infused with cort into the DH, would also block the impairment of retrieval.

Indeed, when infused 15 min prior to testing contextual fear, ICI alone did not affect retrieval, but it substantially reduced the disruption of retrieval by cort (Figure 2.2E).

A Avoidance B Avoidance 8 ^ 8 WT -/- 6 6 Dbh 4 4 ^ ^ ^ ^ 2

2 (min) Latency Latency (min) Latency 0 0 0 1 0 1 1+1 1+1 1 1 Cort (mg/kg) Drug (mg/kg) C Context D Context 80 # 80

60 60 # 40 # # 40

20 20

Freezing (%) Freezing Freezing (%) Freezing

0 0 0 1 1+1 1+1 1 1 0 1 Drug (mg/kg) ICI (mg/kg)

E Context 80 # Saline 60 * Cort 40 # Cort + Betax 20

Freezing (%) Freezing Cort + ICI 0 Betaxolol Cort: 0 0.05 0.05 0 ICI: 0 0 2 2 ICI DH Drug (g)

Figure 2.2 Corticosterone Impairment of Memory Retrieval Depends on 2 and not 1 Signaling or NE/E

(A) Cort impairs IA to a similar extent in control and NE/E-deficient Dbh-/- mice. Only the main effect of treatment was significant (P < 0.001; 7-19/group). (B and C) Co-injection of cort with the 2 antagonist ICI 118,551 (ICI) prevents cort impairment of IA and CF, while co- 42 injection of cort with the 1 antagonist betaxolol (Betax) does not. P < 0.01 for the main effect of treatment on avoidance (6-18/group), and P < 0.001 for the main effect of treatment on freezing (5-7/group). (D) ICI does not enhance low levels of CF in the absence of cort (4-

5/group). Animals were trained with lower shock intensity (0.35 mA) to approximate cort- impaired freezing. (E) Co-infusion of ICI into the DH prevents disruption of CF by cort. P <

0.001 for the main effect of treatment (4-5/group).

2.3.3 Direct Stimulation of 2 Receptors Impairs Memory Retrieval

Given that a 2 antagonist blocks the effect of cort, we asked whether stimulating

2 receptors would disrupt memory retrieval. Toward this goal, the 2 agonist procaterol was administered systemically to wild-type mice prior to testing contextual fear. Like cort, procaterol reduced freezing with a U-shaped dose-response, with 10 g/kg producing maximum impairment (Figure 2.3A). Further, the disruption of memory retrieval by procaterol was blocked by ICI but not the 1 antagonist betaxolol (Figure

2.3B). Like cort, procaterol did not affect cued fear, indicating that the effect of procaterol is on contextual memory retrieval rather than fear or freezing per se (Figure

2.3C). Interestingly, at low doses that did not affect memory retrieval when given alone, combined cort and procaterol treatment significantly impaired retrieval (Figure 2.3D). In contrast, at doses that maximally affected memory retrieval when given alone, combined cort and procaterol treatment had no effect on retrieval. In other words, when the drugs were combined, a U-shaped dose-response was still observed, but with a leftward shift in potency. Further, contextual memory was also impaired when procaterol or a second 2 agonist, zinterol, was infused into the DH (Figure 2.3E). Finally, an impairment of

43 retrieval by procaterol and the block of this effect by ICI but not betaxolol also occurred in wild-type mice for instrumental fear memory (Figure 2.3F).

Context Context A B 80 80 ^ ^ 60 60 * # * 40 40 *

20 20

Freezing (%) Freezing Freezing (%) Freezing 0 0 0 5 10 20 50 250 0 30 100 300 1000 1000 Procaterol (g/kg) Proc plus: ICI (g/kg) Bet

Cue Context C 80 D 80 60 60 * 40 40 ^

20 20

Freezing (%) Freezing Freezing (%) Freezing 0 0 0 10 Proc: 0 5 5 10 Proc (g/kg) Cort: 0 100 250 1000 Drug (g/kg)

Context F Avoidance E 80 8 ^ 60 ^ 6 # ^ ^ 40 # # 4 # ^

20 2

Freezing (%) Freezing Latency (min) Latency 0 0 0 0.2 0.5 2 10 20 50 100 0 10 10 10 DH Procaterol (g) DH Zinterol (ng) +Bet +ICI Procaterol (g/kg)

Figure 2.3 Direct Stimulation of 2 Receptors Disrupts Memory Retrieval

(A and F) The 2 agonist procaterol disrupts CF (5-6/group) and IA (7-22/group) retrieval. P

< 0.01 for the main effect of treatment in both paradigms.B and F) The impairment of CF (5-

6/group) and IA (7-22/group) retrieval by procaterol (Proc) is blocked by co-administration of a 2 antagonist (ICI, 1 mg/kg for F), while co-administration of a 1 antagonist (Bet, 1 mg/kg) does not. P < 0.001 for the main effect of treatment on freezing (B). (C) Procaterol does not affect retrieval of cued fear (5-8/group). (D) When co-administered, cort and procaterol impair

CF retrieval at considerably lower doses than when either is given individually. P < 0.05 for the main effect of treatment (4-7/group). (E) Infusion of the 2 agonists procaterol or zinterol into the DH impairs CF retrieval. P < 0.001 for the main effect of treatment (4-5/group).

2.3.4 Genetic and Interspecies Corroboration that 2 Signaling Mediates Impairment of Retrieval

To corroborate the pharmacologic data, the effects of cort and procaterol in mice lacking either 1- or 2-adrenergic receptors were examined for instrumental fear.

Compared to wild-type, both KO lines displayed normal conditioned avoidance, allowing the examination of drug effects (Figure 2.4A). Cort and procaterol impaired avoidance similarly in wild-type and 1 KO mice, indicating that 1 signaling is not required for the disruption of retrieval. In contrast, neither cort nor procaterol impaired avoidance in 2

KOs, indicating that 2 receptors are necessary for impairment.

While our studies utilizing mice were driven in part by the existence of genetic models, most other animal studies investigating the impairment of memory retrieval by cort have utilized rats. To determine whether the results implicating 2 receptors in the impairment of retrieval might be species-specific, rats were treated with the same pharmacologic reagents that were used in mice. In rats, systemic injections of cort or procaterol disrupted conditioned avoidance one day after training (Figure 2.4B). While similar doses of cort (2 mg/kg) reduced retention latencies equally in mice and rats, somewhat higher doses of procaterol were needed to produce the same degree of 45 impairment. Differences in the pharmacologic properties of procaterol could account for this species-specific difference. In rats, the effect of cort was blocked by co- administration of a 2 but not a 1 antagonist, indicating that glucocorticoid disruption of retrieval via 2 signaling is shared across the two species.

A Avoidance: Mice WT 1 KO # 2 KO 10 8 ^ 6 ^ 4 # ^ #

2 Latency (min) Latency 0 Saline Cort 1 Proc 0.01 Drug (mg/kg)

B Avoidance: Rats Saline Cort Cort + Bet Cort + ICI 10 ^ ICI 8 Procaterol 6 4 2 ^ ^

Latency (min) Latency ^ 0 0 2 2+1 2+1 1 0.06 Drug (mg/kg)

Figure 2.4 Genetic and Interspecies Corroboration that 2 Signaling Mediates Impairment of Retrieval

(A) Cort or procaterol (Proc) disrupts retrieval of IA in 1 KO but not 2 KO mice. P < 0.001 for the main effect of treatment; P < 0.001 for the main effect of genotype; and P < 0.01 for the interaction of treatment and genotype (7-10/group). (B) Cort or procaterol impair IA in rats.

Cort-induced disruption of retrieval is prevented by co-administration of the 2 antagonist ICI but not the 1 antagonist betaxolol (Bet). P < 0.001 for the main effect of treatment (5-

10/group).

2.3.5 Impairment of Memory Retrieval by Stress Requires NE/E and 2 Signaling

Like cort, acute stress disrupts hippocampus-dependent emotional memory retrieval in rats[123]. To verify and extend these findings, we subjected wild-type mice to inescapable restraint of various durations 30 min prior to testing contextual fear.

Restraint stress reduced freezing in a U-shaped manner, and at durations of 2 and 5 min, impaired retrieval to a similar extent as that for treatment with cort or a 2 agonist (Figure

2.5A). To test whether restraint impairs retrieval through a mechanism consistent with

GR activation, mice were pretreated with the GR antagonist Org 34850 (Org) 15 min prior to administering restraint or cort. At 15 mg/kg, Org partially blocked the effects of restraint and treatment with 2 mg/kg cort, and completely blocked the effect of 1 mg/kg cort. At 30 mg/kg, Org fully blocked both restraint- and cort-induced retrieval deficits

(Figure 2.5B). The relatively high systemic doses of Org required are consistent with its limited CNS penetration[209]. To test whether stress-induced retrieval deficits depend on 2 signaling, ICI was administered 15 min before restraint. ICI completely blocked the effects of restraint on retrieval (Figure 2.5C). Further, the elevated levels of plasma cort observed 30 min following restraint were not reduced by pretreatment with ICI or

Org, indicating that these compounds do not impair the release of cort (Figure 5D).

Interestingly, plasma levels of cort following stress or cort administration did not correspond well with the impairment of retrieval. For example, restraint doubled cort over basal levels, and this level was comparable to that following injection of saline

(Figure 2.5D). Yet restraint impaired retrieval while saline injection did not. Further, injection of cort at doses that impair retrieval resulted in plasma levels that were 3- to 4- fold greater than those for restraint. Naturally, one would expect that restraint stress activates stress-response systems in addition to that for glucocorticoids. This expectation led us to test whether NE/E contribute to the impairing effects of restraint on retrieval, even though the impairing effects of exogenous cort are independent of NE/E. Restraint significantly impaired retrieval of instrumental fear in wild-type mice (Figure 2.5E). In contrast, restraint did not impair retrieval in Dbh-/- mice, although there was a trend toward a small effect. Further supporting of a role for 2 receptors, restraint also did not impair retrieval in 2 KO mice, although there was also a trend toward a small effect.

Context A 80 C Context 80 60 * 60 40 ^ # 40 20

Freezing (%) Freezing 20 Freezing (%) Freezing

0 0 0 0.5 2 5 10 Cont Res Restraint (min) Pretreatment: ICI

B Context Control Restraint Cort 1 Cort 2 80

60 # # * 40 #

20 Freezing (%) Freezing

0 Vehicle Org 15 Org 30 Pretreatment (mg/kg) WT D Plasma E Avoidance Dbh-/- # 2 KO 1200 8 1000 # 6 800 600 4 400 # # ^

Cort (ng/ml) Cort ^ 2 200 ^ (min) Latency 0 0 Sal Res Sal Res Sal Res

Saline Cort 1 Cort 2 ControlSal-ResOrg-ResICI-Res

Figure 2.5 Impairment of Memory Retrieval by Stress Depends on GR and 2 Signaling

(A) Inescapable restraint stress 30 min before testing disrupts CF retrieval, with maximal reduction at intermediate durations. P < 0.01 for the main effect of duration (3-7/group). (B)

Pretreatment with the GR antagonist, Org 34850 (Org, 30 mg/kg), 15 min prior to 2-min restraint stress or cort prevents disruption of CF retrieval. P < 0.001 for the main effect of

49 pretreatment; P < 0.01 for the main effect of treatment; and P < 0.001 for the interaction of pretreatment and treatment (4-9/group). (C) Pretreatment with the 2 antagonist ICI 15 min prior to 2-min restraint stress prevents disruption of CF retrieval (4-9/group). (D) Plasma cort levels rise significantly 30 min after restraint (Res) or the systemic injection of saline or cort

(1 or 2 mg/kg). Pretreatment with either Org (30 mg/kg) or ICI (1 mg/kg) 15 min prior to 2- min restraint stress does not prevent the rise in cort. P < 0.001 for the main effect of treatment (7-8/group). (E) Restraint stress (2 min) impairs IA retrieval in WT but not Dbh-/- or

2 KO mice. Mice were either injected with saline or restrained 30 min before testing. P <

0.01 for the main effect of treatment (7-19/group). The only significant post-hoc comparison between saline and restraint by genotype is for the WT mice.

2.3.6 Corticosterone, 2 Signaling and Stress Impair Retrieval in a Time-limited Manner

NE, 1, cAMP and PKA signaling all play a time-limited role in hippocampus- dependent emotional memory retrieval[92, 153]. For example, NE/E-deficient mice display deficits in contextual fear 0.1 - 4 days but not 5 or more days after training, and mice and rats treated with a antagonist show compromised contextual fear and spatial navigation over a similar time course. Therefore, we determined whether stress, glucocorticoids and 2 signaling might also have time-limited effects on retrieval. In vehicle-injected mice, contextual and instrumental fear did not differ between days 1, 3 and 5 (Figures 2.6A and 2.6B). However, the effects of cort, procaterol and restraint stress were all time-sensitive, being present on days 1 and 3 after training but absent on day 5. DH infusion of procaterol or cort 5 days after training also had no effect on retrieval (Figure 2.6C).

A Context 80 Saline 60 Stress # * 40 # # # ^ Cort

20 Proc Freezing (%) Freezing 0 1 3 5 Days After Training

Avoidance Context (D5) B 10 C 80 8 60 6 * 40 4 # # 20

2 (%) Freezing Latency (min) Latency 0 0 1 3 5 0 0.05 2 Days After Training DH Drug (g)

Figure 2.6 Corticosterone, 2 Signaling and Stress Inhibit Hippocampus-dependent Memory Retrieval in a Time-limited Manner

(A) Cort (1 mg/kg), procaterol (Proc, 10 g/kg) or restraint (Stress, 2 min) impairs CF retrieval at 1 and 3 but not 5 days after training. P < 0.001 for the main effect of day; P <

0.001 for the main effect of treatment; and P < 0.01 for the interaction of day and treatment

(4-12/group). (B) Cort (1 mg/kg) or procaterol (10 g/kg) impairs IA retrieval at 1 day, partially impairs retrieval at 3 days, and does not impair retrieval at 5 days after training. P <

0.001 for the main effect of day; P < 0.05 for the main effect of treatment; and P < 0.05 for the interaction of day and treatment (7-20/group). (C) Neither infusion of cort nor procaterol into the DH affects CF retrieval 5 days after training (5/group).

2.3.7 By Coupling to Gi/o Proteins, 2 Opposes the Role of 1 in Retrieval

To obtain more direct evidence that 1 and 2 receptors oppose each other in retrieval, we asked whether interfering with 2 signaling would rescue retrieval when 1

51 signaling was blocked or absent. As expected, retrieval was impaired when wild-type mice were treated with the 1 antagonist betaxolol shortly before testing contextual fear one day after training (Figure 2.7A). However, when ICI was combined with betaxolol, no effect on retrieval was observed, i.e. the 2 antagonist reversed the deficit caused by the 1 antagonist. Similarly, ICI reversed the retrieval deficits present in 1 KO and

-/- Dbh mice. Lastly, 1/2 double KO mice exhibited normal contextual fear, providing additional genetic support for the pharmacologic data.

Next, we sought to determine the mechanism by which 2 signaling impairs retrieval. Although the traditional view is that -adrenergic receptors couple to and signal via the adenylyl cyclase-stimulatory G protein Gs, there is evidence that 2 receptors can also signal through Gi/o[210-212]. Because cAMP signaling is necessary and sufficient for contextual memory retrieval mediated by NE and 1 receptors[153], we asked whether 2 signaling might oppose 1 signaling in retrieval by activating Gi/o and inhibiting adenylyl cyclase.

To test this idea, pertussis toxin (PTx) was employed because it uncouples Gi/o proteins from their receptors by ADP-ribosylation of Gi/o. Because PTx can take several days to be fully effective in the hippocampus in vivo[213, 214], we infused PTx into the

DH one day before fear conditioning to minimize potential effects on acquisition and consolidation. Mice were tested for retrieval two days after conditioning. Pretreatment with PTx did not affect retrieval in mice infused with vehicle 15 min before testing

(Figure 2.7B). Strikingly, however, PTx pretreatment completely prevented the impairment of retrieval by zinterol, cort and stress. A low dose of cort was also

52 examined to determine whether PTx pretreatment might shift the dose-response relationship to the left (i.e. more sensitive to cort), but no evidence for this was observed.

Given the results with PTx, we then asked what effect selectively stimulating either 1 or 2 receptors would have on cAMP levels in the DH. To avoid the potential confounds of anesthesia and sacrifice on cAMP levels following DH infusion, DH slices were used. Slices were incubated in aCSF with or without the non-selective  agonist isoproterenol (Iso). To selectively stimulate 1 receptors, slices from 3 KO mice were treated with Iso plus the 2 antagonist ICI. To selectively stimulate 2 receptors, slices from 1/3 double KO mice were treated with Iso. To stimulate both 1 and 2 receptors, slices from 3 KO mice were treated with Iso. As a control, slices from 1/2/3 triple

KO mice were treated with Iso. Iso had no effect on cAMP in the triple KO slices

(Figure 2.7C). However, Iso at 0.3 - 1 M increased cAMP when selectively stimulating

1 receptors (Figure 2.7E). In contrast, Iso at 0.1 – 1 M caused cAMP to decrease when selectively stimulating 2 receptor (Figure 2.7D). Finally, when 1 and 2 receptors were stimulated simultaneously, no increase in cAMP was observed (Figure 2.7F). Taken together, the results indicate that 1 and 2 receptors functionally oppose one another with respect to cAMP levels.

-/- A Context WT 1 KO Dbh  / DKO # 1 2 80 ^ # 60 ^ 40 ^ #

20 Freezing (%) Freezing

Sal Bet Sal ICI Sal ICI Sal Bet-ICI Treatment B Context DH Pretreatment: Saline PTx 80 ^ # ^ 60

40 # # #

20 Freezing (%) Freezing 0 Sal Zint 100 Cort 50 Cort 5 Stress Test Treatment (ng in DH or Stress)

 Recs C No  Recs D 2  TKO) 1,3 DKO) 250 1,2,3 250 200 200 150 150

100 100 cAMP (%) cAMP cAMP (%) cAMP 50 50 0 0 ^ 0 0.1 0.3 1 0 0.1* 0.3 1 Iso (M) Iso (M)

E  1 Recs F  1 +  2 Recs 3 KO + ICI) 3 KO) 250 250 200 200 150 ^ 150

100 * 100 cAMP (%) cAMP 50 (%) cAMP 50 0 0 0 0.1 0.3 1 0 0.1 0.3 1 Iso (M) Iso (M)

Figure 2.3.7 2 Signaling Opposes 1 Signaling in Retrieval and in cAMP Production through Gi/o

(A) The impairment of CF retrieval by betaxolol (Bet, 1 mg/kg, 30 min before testing one day after training) is prevented by concurrent administration of ICI (1 mg/kg). The deficits in CF

-/- retrieval exhibited by 1 KO and Dbh mice one day after training are reversed by treatment with ICI 30 min before testing. Finally, 1/2 double KO (DKO) mice do not exhibit a deficit in

CF. P < 0.001 for the main effect of group (5-16/group). (B) Infusion of pertussis toxin (PTx,

10 ng) into the DH one day prior to training prevents the impairment of CF retrieval by restraint, cort (50 ng) or zinterol (100 ng). Restraint was administered 30 min before and cort or zinterol was infused 15 min before testing 2 days after training. P < 0.001 for the main effect of pretreatment; P < 0.001 for the main effect of treatment; and P < 0.001 for the interaction of pretreatment and treatment (5/group). (C - F) Hippocampal slices were incubated in aCSF (0) or various concentrations of the non-selective  receptor agonist isoproterenol (Iso). (C) Non-specific effects of Iso on cAMP were not apparent in slices from

1/2/3 triple KO (TKO) mice (5-10/group). (D) Selective stimulation of 2 receptors by Iso decreased cAMP in slices from 1/ 3 DKO mice. P < 0.05 for the main effect of concentration

(8-18/group). (E) Selective stimulation of 1 receptors by Iso increased cAMP in slices from 3

KO mice treated with the 2 antagonist ICI. P < 0.05 for the main effect of concentration (7-

8/group). (F) Combined stimulation of 1 and 2 receptors by Iso did not alter cAMP in slices from  3 KO mice (8-10/group).

2.4 DISCUSSION

The goal of this study was to resolve the competing ideas of how NE is required for the retrieval of hippocampus-dependent emotional memory, yet is also responsible for the impaired retrieval of such memories by stress and glucocorticoids. Results from the current studies employing NE/E-deficient Dbh-/- mice indicate that NE/E are not required for the impairing effects of glucocorticoids per se (Figure 2.2A). This finding is contrary 55 to the hypothesis that glucocorticoids impair memory by enhancing the release of

NE[115, 140, 141]. However, NE/E do contribute to the impairing effects of stress, as evidenced by the attenuation of stress-induced retrieval impairment in Dbh-/- mice (Figure

2.5E). Consistent with this finding, relatively high plasma levels of cort were required to impair retrieval when cort was administered before testing (Figure 2.5D). In contrast, the levels of cort following restraint stress were only moderately elevated (about twice baseline), and were similar to the levels observed following injection of saline, a procedure that does not impair retrieval. We hypothesized that stress-response systems in addition to glucocorticoids are engaged following restraint, and that these other factors synergize with glucocorticoids to impair retrieval. Evidence from Dbh-/- mice indicates that NE/E are among these factors. Our data using agonists for the two systems indicate that NE/E and glucocorticoids released during stress may act in concert to impair retrieval at hormone levels that would be innocuous individually (Figure 2.3D).

Consistent with the above, extracellular levels of NE in the DH are significantly elevated

30 min after restraint stress, whereas they remain unchanged 30 min after intraperitoneal injection of saline[215, 216].

Despite confirming that glucocorticoids and NE/E contribute to the impairment of hippocampus-dependent emotional memory retrieval by stress, the mechanisms by which these factors impair retrieval have been unclear. Prior pharmacologic evidence suggested that glucocorticoids and NE/E impair retrieval by activating 1-adrenergic receptors[141].

However, additional pharmacologic and genetic evidence indicates that signaling by 1 receptors is required for contextual and spatial memory retrieval[92]. In the current

56 study, we found that glucocorticoids impair memory retrieval in the presence of a 1 antagonist (mice and rats) and in 1 KO mice (Figures 2.2 and 2.4), indicating that 1 signaling is not required for the impairment of retrieval. In contrast, we found that low doses of 2 receptor agonists impair retrieval in mice and rats (Figures 2.3 and 2.4). In support of a critical role for 2 receptors, the impairing effects of 2 agonists, cort and stress were greatly reduced both in the presence of a 2 antagonist (mice and rats) and in

2 KO mice (Figures 2.2 – 2.5). Consistent results between rats and mice are important because nearly all of the prior animal studies on stress, glucocorticoids and memory retrieval were performed in rats.

The data implicating 1 receptors in the impairment of retrieval were based in part on the use of the 1-selective agonist xamoterol[141]. In that study, doses of 3 and 10 mg/kg were used to demonstrate retrieval impairment. In contrast, xamoterol fully rescues contextual memory retrieval in Dbh-/- mice at 1 mg/kg. We hypothesized that higher doses of xamoterol activate 2 (as well as 1) receptors, and that the activation of

2 receptors impairs retrieval. In support of this, we have found that the impairment of retrieval by 6 mg/kg xamoterol is blocked in the presence of a 2 antagonist and in 2 KO mice, while impairment is unaffected in the presence of a 1 antagonist and in 1 KO mice (Chapter 3), analogous to the current results for the 2-selective agonist procaterol.

Extrapolating on the above, these results suggest that at the doses used previously[141], the 1-selective antagonist atenolol also blocks 2 receptors.

Classically, -adrenergic receptors couple to the adenylyl cyclase-stimulating G protein Gs, and their activation increases cAMP. Given that 1 receptors and cAMP 57 signaling in the hippocampus are required for hippocampus-dependent emotional memory retrieval[92, 153], it was not apparent how activation of 2 receptors in the hippocampus would impair retrieval. One possibility was that these two receptor subtypes are expressed on different cells or in distinct subcellular compartments, where increases in cAMP would have opposite effects on processes important for retrieval.

However, a second possibility was suggested by recent studies examining signaling by 2 receptors. Evidence suggests that 2 receptors can couple to the pertussis toxin-sensitive, adenylyl cyclase-inhibiting G protein Gi, and that such coupling can limit or prevent increases in cAMP in cardiac myocytes[210-212]. In support of such a mechanism, uncoupling Gi/o proteins from their receptors by PTx pretreatment in the DH blocked the impairing effects of stress, cort and a 2 agonist on retrieval (Figure 2.7B). Further, cAMP levels in DH slices were increased by selectively stimulating 1 receptors, but were decreased by selectively stimulating 2 receptors (Figure 2.7C). Finally, when 1 and 2 receptors were stimulated simultaneously, no changes in cAMP were observed.

Based on the above, the simplest explanation for the impairing effects of stress, glucocorticoids and 2 signaling is that they oppose the increase in cAMP that is necessary for retrieval. However, 2 signaling through Gi/o has been linked to effects that may be independent of a constraint on cAMP levels[217], and it is possible that some of these effects are also relevant to the impairment of retrieval. Additional studies will be needed to address this. Highlighting the contrasting roles of these two receptors, 2 signaling in the heart acts to protect cardiomyocytes from excessive 1 stimulation that can lead to apoptosis and heart failure if unchecked[206, 207].

Results from many studies have suggested that important interactions underlie the cognitive effects of cort and NE[195, 196, 218, 219]. Our results suggest that this occurs at least in part at the level of 2-adrenergic signaling. It is thus of interest to consider how glucocorticoid signaling might intersect with 2 signaling to impair retrieval.

Glucocorticoid signaling appears to be mediated by a GR-like receptor because fluticasone mimics the effect of cort and Org 34850 blocks the effect of cort (Figures

2.1C and 2.5A). This is consistent with other studies suggesting a role for GR or GR-like signaling in the impairment of retrieval[106, 107]. 2 signaling appears to be downstream of glucocorticoid signaling because the 2 antagonist ICI blocks the impairing effect of cort (Figures 2.2 and 2.4). However, it is not clear how glucocorticoid signaling might activate 2 signaling, given that the effect of cort does not require NE/E (Figure 2.2A). One possibility is that either cort or GR interact directly with the 2 receptor to cause activation. Another possibility is that cort and GR may signal through an intermediary that leads to the activation of 2 signaling. In each of these cases, the 2 receptor would be a key effector for at least some of the rapid, non- genomic effects of glucocorticoids. Alternatively, it is possible that the impairment of retrieval requires coincident signaling by 2 and GR, a mechanism that would also be sensitive to 2 blockade. This mechanism seems less likely because exogenous cort administration in the absence of NE/E impairs retrieval; however, the coincident signaling model is possible if there is sufficient constitutive signaling at 2 receptors in the absence of NE/E[220, 221]. The reversal of retrieval impairment by ICI in Dbh-/-

59 mice is consistent with the idea that there is some constitutive signaling at 2 receptors

(Figure 2.7A).

Interestingly, cort, fluticasone, procaterol and the combination of cort plus procaterol all exhibited U-shaped dose-response curves for retrieval impairment when given systemically (Figures 2.1 and 2.3). These observations are complementary to the inverted U-shaped dose-response curve reported for the GR agonist RU 28362 in the enhancement of emotional memory consolidation[136]. These findings suggest that moderate levels of stress can affect memory, but higher levels of stress actually have less of an effect. Indeed, we found that 10 min of restraint impaired retrieval less than 2 or 5 min of restraint (Figure 2.5A). The reason why impairment dissipates with higher doses has not been clear. Interestingly, specific stimulation of 2 receptors by isoproterenol decreased cAMP in a U-shaped manner, providing potential insight into the above. It is possible that higher levels of stimulation result in rapid desensitization of 2 signaling, or that the recruitment of additional signaling pathways by 2 receptors mitigate signaling induced by lower levels of stimulation. Future studies will be needed to address this.

In addition to the level of stress being an important determinant of whether effects on memory are observed, our data demonstrate that the duration between acquisition and retrieval is also an important factor (Figure 2.6). The time course for the effects of stress, cort and 2 signaling was essentially identical to that for the requirement of NE, 1, cAMP and PKA signaling in retrieval[92, 153], i.e. an intermediate phase of memory was susceptible. These findings are consistent with the idea that 2 signaling acts to oppose the role of 1 signaling in retrieval. The time course data also rule out the possibility that

60 stress, cort and 2 signaling affect contextual information processing per se, which would be independent of the interval between training and testing.

A potentially important aspect of -adrenergic receptor physiology is that NE, the adrenergic neurotransmitter in the DH, is about an order of magnitude more potent in activating 1 receptors than 2 receptors[222]. As a result, it seems likely that moderately arousing conditions would facilitate hippocampus-dependent emotional memory retrieval through the release of NE that would then activate 1 receptors. Extending this idea, strongly arousing conditions typical of ―stress‖ would impair memory retrieval through the release of glucocorticoids and additional NE that together would activate 2 receptors.

Because strongly arousing conditions enhance memory consolidation, this could temporarily facilitate a shift away from processes favoring retrieval toward those favoring consolidation.

With respect to consolidation, results from many studies indicate that stimulating glucocorticoid or adrenergic signaling by exogenous administration of agonists shortly after conditioning enhances memory retention[77, 195, 196]. These results have often been interpreted as indicating that glucocorticoid and adrenergic signaling are required for the enhanced consolidation of emotionally arousing events. In contrast, interfering with glucocorticoid signaling shortly after training leads to deficits in hippocampus- dependent memory consolidation, but does not impair emotional memory in general, such as cued fear[102, 198]. The current study confirms and extends observations indicating that adrenergic signaling is not uniquely required for the consolidation of emotional memory, regardless of whether it depends on the hippocampus. While NE/E-deficient

-/- Dbh and 1 KO mice exhibit deficits in hippocampus-dependent emotional memory

-/- retrieval, neither Dbh , 1 KO, 2 KO nor 1/2 double KO mice exhibit deficits in the consolidation of instrumental or Pavlovian fear[92, 93; current study and unpublished data]. Additional studies will be required to resolve the findings that -adrenergic signaling is not required for consolidation of these tasks, even though exogenous stimulation of  signaling is sufficient to enhance consolidation.

The majority of studies examining the roles of -adrenergic signaling in human cognition have employed the general  receptor antagonist propranolol. Given that 2 signaling can act to oppose 1 signaling, a potentially important corollary derived from the current studies is that the use of general  receptor agents (e.g. isoproterenol as an agonist and propranolol as an antagonist) may not readily reveal the distinct roles of individual  receptors, and the use of these agents may generate results that are complex or misleading. This idea is support by our pharmacologic and genetic data (Figure 2.7A).

Findings from the current study are likely relevant to common situations in which stress negatively impacts cognition. Public speaking and examinations testing knowledge often result in anticipatory arousal and stress that can interfere with cognitive performance. These tasks usually depend on the retrieval of recently learned or reviewed information that is likely susceptible to the impairing effects of stress. Interestingly, the non-selective  antagonist propranolol is reported to enhance performance on examinations and in cognitive flexibility tasks[166-168]. Our findings suggest that the effect of propranolol is due to its blockade of the 2 receptor, and that a selective 2 antagonist would be similarly efficacious.

Finally, one of the hallmarks of PTSD is the intrusive retrieval of memories for traumatic events. At first pass, results from the current study and other studies suggest that glucocorticoids, 2 agonists and 1 antagonists might reduce the retrieval of traumatic memories in PTSD. However, it is possible that these agents would often be without effect on traumatic memory retrieval because of the limited time window over which they would reduce retrieval[Figure 6; 92]. Further, hippocampus-independent (e.g. cued) traumatic memory is likely to be unaffected by these agents. Perhaps the most promising avenue for treating the intrusive retrieval of traumatic memories in PTSD will be to interfere with the reconsolidation of these memories following retrieval[169, 223].

Such an approach might be more efficacious than promoting extinction of traumatic memories because extinction learning is typically context-dependent. Indeed, results from some studies suggest that  blockers such as propranolol may interfere with the reconsolidation of aversive memory and reduce symptoms of PTSD[165, 171, 224, 225].

2.5 EXPERIMENTAL PROCEDURES

2.5.1 Subjects

+/− −/− Wild-type, Dbh , Dbh , β1 KO and β2 KO mice were on a hybrid 129/Sv x

C57BL/6 background[91, 226, 227]. β3 KO, β1/β3 double KO and β1/β2/β3 triple KO mice were on a hybrid FVB/N x 129/Sv x C57BL/6 x DBA/2 background[228]. Mice were generated by mating either heterozygotes or homozygotes, and genotype was determined by PCR. The prenatal loss of Dbh−/− mice was rescued as previously described[229]. No significant differences in results were found by gender or parental genotype, so data were combined. Female Fischer 344 rats (Harlan, Indianapolis, IN)

63 were 3-4 weeks old upon arrival. Animals were maintained on ad lib food and water and a 12 h light:dark cycle, with lights on beginning at 07:00. Animals were housed in relatively small, quiet rooms for 3-4 weeks before studies began to minimize the stress associated with caretaking and colony management during the light phase. Mice were 3-

6 months old and rats were 7-11 weeks old when tested. Studies were performed during the light phase, with most experiments taking place between 08:00 - 15:00. Studies were in accordance with NIH guidelines and had the approval of the IACUC at the University of Pennsylvania.

2.5.2 Pavlovian Fear Conditioning

Adjacent to the training room, animals were placed in pairs (mice) or singly (rats) into opaque holding buckets (~12 cm diameter) with bedding and covers for 30-60 min before being manipulated further. Animals were given a 3-min handling session each day for 2 days in the training room. Saline was injected at the end of handling each day. On the following day, training consisted of placing the animal in the conditioning apparatus

(ENV-010MC with ENV-414S, Med Associates, St. Albans, VT) for 2 min, after which an 84 dB, 4.5 kHz tone was activated for 30 s. Two seconds before the end of the tone, a

2 s, 1 mA footshock was delivered. The animal was returned to its home cage 30 s after shock. Contextual fear was tested for 5 min in the conditioning apparatus in the absence of the tone. Cued fear was tested in a context containing distinct visual, tactile, and olfactory cues. After 2 min, the training tone was turned on for 3 min. Percent freezing was estimated by scoring the presence or absence of non-respiratory movement every 5 s.

Tests were conducted 1 day after training except where indicated.

2.5.3 Instrumental Fear Conditioning

Animals were handled as described above. Training consisted of placing a rodent in the lighted chamber of the apparatus used for Pavlovian conditioning and timing its latency to fully enter (except for the tail) the dark chamber. Once the animal entered the dark chamber, the retractable partition separating the two chambers was lowered and a footshock was delivered for 2 s (0.30 – 0.35 mA for mice, 0.75 mA for rats). After 15 s the animal was returned to its home cage. Animals that did not enter the dark chamber after 100 s during conditioning were excluded. Testing was identical to training except that no shock was delivered and the partition remained up. Latencies to enter the dark chamber were recorded. If an animal did not enter the dark chamber within 10 min, it was returned to its cage and assigned a latency of 10 min. Tests were conducted 1 day after training except where indicated.

2.5.4 Restraint Stress

Mice were handled as described above, except that they were placed individually into the holding buckets. Mice were immobilized by being placed into ventilated 50 ml plastic tubes for 2 min (unless noted otherwise), 30 min prior to testing retrieval.

2.5.5 Drugs

Fluticasone, betaxolol HCl, ICI 118,551 HCl, procaterol HCL (Tocris, Ellisville,

MO), corticosterone (Sigma, St Louis, MO) and Org 34850 (Organon, Newhouse, UK) were administered subcutaneously 30 min before testing for standard conditions, or 45 min before testing when restraint was administered 30 min before testing. Except for

65 cort, fluticasone and Org 34850, drugs were dissolved in 0.9% saline (procaterol also contained 0.1 mg/ml ascorbic acid, pH 7.4, Sigma). Cort and fluticasone were dissolved in DMF and diluted into saline (0.025-0.3% DMF), while Org 34850 was dissolved in

DMSO. Vehicle was either saline, saline containing 0.1 mg/ml ascorbic acid, saline containing 0.3% DMF, or 100% DMSO. Injection volumes were 10 l/g except for Org

34850 and DMSO, which were 2 l/g. Zinterol HCl and pertussis toxin (Tocris) were dissolved in 0.9% saline (zinterol also contained 0.1 mg/ml ascorbic acid, pH 7.4) and infused into the DH. For DH infusion, cort was dissolved in DMF and diluted into saline

(0.1-1% DMF).

2.5.6 Dorsal Hippocampus Infusion

A double guide cannula (C235 system, Plastics One, Roanoke, VA) was implanted under pentobarbital anesthesia (72.5 mg/kg ip) using a stereotax

(SAS75/EM40M, Cartesian Research, Sandy, OR). The guide was placed 1.7 mm posterior to Bregma and 1.5 mm bilateral for DH infusions. The guide projected 1.5 mm below the base and the dummy cannula extended 0.5 mm below the guide. One week after surgery, bilateral infusions were made into conscious mice while gently holding the nape of the neck. The dual injection cannula extended 0.9 mm below the guide.

Infusions were 0.5-1 μl/side at 0.4 μl/min, with the injection cannula being left in place for 30 s before the mouse was returned to its home cage. Because studies indicate that the effects of PTx are best evaluated 3 days after infusion, PTx was infused 3 days before testing (i.e. 1 day before training, which was 2 days before testing). A pilot study

66 indicated that 1 ng and 10 ng PTx (comparable to or lower than what has been used by others) were similarly effective.

2.5.7 Corticosterone Levels

Mice were handled for two days as described above. Between 08:00 and 11:00 on the third day, some mice were either left untreated or received an injection of vehicle or cort 30 min before retro-orbital blood collection using heparin-coated capillary tubes.

Other mice were subjected to 2 min of restraint stress 30 min before collection. These mice were pretreated with either vehicle, 1 mg/kg ICI or 15-30 mg/kg Org 34850 15 min before restraint. Blood was placed on ice and then spun at 2,000g at 4°C for 5 min and plasma was stored at -80°C. Cort was measured in duplicate by [125I]-radioimmunoassay

(MP Biomedicals, Orangeburg, NY) according to instructions.

2.5.8 cAMP Levels

Mice were decapitated without anesthesia and the brain was rapidly removed and chilled in ice-cold artificial cerebral spinal fluid (aCSF) containing sucrose as described[92]. Brain slices transverse to the DH were cut at 400 m and DH slices were surgically isolated and allowed to recover for at least 1 h at 37°C in oxygenated regular aCSF. Slices were then transferred to a second chamber containing isoproterenol (and

0.1 M ICI for selective stimulation of 1 receptors in 3 KO slices) in oxygenated aCSF at 37°C for 2 min, after which they were immediately placed on ice in 0.15 ml 6% TCA with ~4,000 cpm of [3H]-cAMP (adenosine 3′, 5′-cyclic phosphate, ammonium salt, [2,8-

3H]; PerkinElmer, Waltham, MA) and then homogenized using a Sonic Dismembrator

100 (Fisher, Pittsburgh, PA). Pilot experiments indicated that 5 min incubations provided similar results to those for 2 min incubations, while 1 min incubations provided similar results or did not affect cAMP. Pilot incubations with 10 nM isoproterenol were without effect on cAMP and are not shown. Extracts were centrifuged at 2,500g at 4°C for

15 min, and the pellet was stored at -80°C for subsequent Bradford assay to determine total protein. Supernatants were extracted four times with 0.6 ml water-saturated ether, evaporated, reconstituted in 1 ml assay buffer and stored at -20°C. Recovery was assessed by measuring the amount of [3H]-cAMP in each sample, which ranged from 60-

80%. To determine cAMP levels, a [125I]-radioimmunoassay (PerkinElmer) was employed using 100 l of undiluted sample in duplicate according to instructions for the non-acetylated procedure. On the basis of the Bradford assays, percent recoveries and standard curves, a cAMP level (pmol /mg protein) was calculated for each slice. For each genotype, the mean for the slices incubated in aCSF alone was normalized to 100%.

Actual values were (in pmol/mg protein): 102 ± 11 for 1/2/3 triple KO, 95 ± 15 for

1/3 double KO, 88 ± 7 for 3 KO, and 109 ± 18 for 3 KO with 0.1 M ICI.

2.5.9 Statistics

Data were analyzed with Statistica 9.0 (StatSoft, Tulsa, OK) using one- or two- way ANOVA. Post-hoc comparisons were made using Fisher’s LSD. Data are presented as mean ± standard error. For all figures, * indicates P < 0.05, ^ indicates P < 0.01, and # indicates P < 0.001. Comparisons marked as significant are to the reference group except where indicated.

2.6 ACKNOWLEDGMENTS

We thank Dainippon Sumitomo Pharma (Osaka, Japan) for their generous gift of

L-threo-3,4-dihyroxyphenylserine used to rescue Dbh-/- mice prenatally, Organon

(Newhouse, UK) for providing Org 34850, and B. Kobilka and B. Lowell for providing stock for the  receptor KO mouse lines. This work was supported by NIH grant

5R01MH063352 and DOD grant PT075099 to S.A.T, and NIH grant 5T32GM008076 to

R. Pittman.

Chapter 3

Xamoterol Impairs Hippocampus- dependent Emotional Memory Retrieval via Gi/o-coupled 2- adrenergic Signaling

Keith Schutsky, Ming Ouyang & Steven A. Thomas

Department of Pharmacology, University of Pennsylvania, Philadelphia, PA 19104

Author contributions: K.S. and S.A.T. designed research; K.S. and M.O. performed research; K.S., M.O. and S.A.T. analyzed data; K.S. and S.A.T. wrote the paper.

3.1 ABSTRACT

Xamoterol, a partial 1-adrenergic receptor agonist, has been reported to impair the retrieval of hippocampus-dependent spatial reference memory in rats. This finding is in contrast to the observation that xamoterol restores memory retrieval in gene-targeted mice lacking norepinephrine (NE) and in a transgenic mouse model of Down syndrome in which NE is reduced. Restoration of retrieval by xamoterol in these two models complements the observation that NE and 1 signaling are required for hippocampus- dependent retrieval of contextual and spatial reference memory in wild-type mice and rats. Additional evidence indicates that cAMP-mediated PKA and Epac signaling are required for the retrieval of hippocampus-dependent memory. As a result, we hypothesized that xamoterol has effects in addition to the stimulation of 1 receptors that, at higher doses, act to counter the effects of 1 signaling. Using Pavlovian and instrumental fear conditioning, as well as pharmacologic and genetic approaches to manipulate -adrenergic signaling, we report that xamoterol-induced disruption of memory retrieval depends on 2-adrenergic receptor signaling. Interestingly, the impairment of memory retrieval by xamoterol is blocked by pertussis toxin pretreatment, suggesting that 2 signaling opposes 1 signaling during memory retrieval at the level of

G protein and cAMP signaling. Finally, similar to the time-limited roles for NE, 1 and cAMP signaling in hippocampus-dependent memory retrieval, xamoterol only impairs retrieval for several days after training. We conclude that the disruption of memory retrieval by xamoterol is mediated by Gi/o-coupled 2 signaling, which opposes the Gs

71 coupled 1 signaling that is transiently required for hippocampus-dependent emotional memory retrieval.

3.2 INTRODUCTION

As the registrar of our personal experiences, episodic memory is one of the hallmarks of identity. Among the neural systems that underlie episodic memory, the hippocampus is central. Bilateral damage to the hippocampus can eliminate the ability to form new episodic memories, as well as impair the ability to recall recent but not remote episodic memories[230, 231]. Many of our most influential episodic memories are derived from events rich with emotion that can range from ecstatic to traumatic. While the vast majority of these memories are beneficial and/or prized, intense, psychologically traumatic experiences can lead to psychiatric problems such as post-traumatic stress disorder (PTSD). In fact, dysregulation of episodic memory is one of the cardinal symptoms in PTSD patients, who often suffer from recurrent, intrusive memories of traumatic events[232]. Frequent retrieval of such memories can be debilitating due to the fear and anxiety they invoke.

For this and other reasons, the mechanisms underlying the consolidation, retrieval and reconsolidation of episodic memory are of great interest. Many studies have contributed to our understanding of the molecular and cellular signaling pathways important for episodic memory consolidation[3, 233-235]. In addition, some studies have begun to illuminate signaling pathways required for the retrieval of episodic memory[236, 237]. As a result, it has been suggested that the consolidation and retrieval of hippocampus-dependent memory have common neurobiologic substrates and signal

72 transduction pathways. As an example, nearly four decades of research across species illustrate that agents that acutely elevate cAMP (cyclic 3’,5’-adenosine monophosphate) usually enhance memory consolidation, whereas agents that block cAMP signaling impair consolidation[3, 234].

Recent evidence suggests that cAMP signaling is also involved in hippocampus- dependent memory retrieval. Bilateral infusion of a cAMP antagonist into the dorsal hippocampus (DH) of rats 10 min before testing inhibitory avoidance one day after training reduces step-down latency[154]. Further, signaling by norepinephrine (NE) through the 1-adrenergic receptor is required for the retrieval of an intermediate phase of hippocampus-dependent memory[92]. Mice with targeted disruption of the gene for the

1 receptor (1 KO) display deficits in contextual fear, and systemic administration of the

1 antagonist betaxolol produces retrieval deficits in wild-type mice and rats. These effects are reproduced by infusion of the 1-selective antagonist atenolol into the DH. 1 receptors couple to the adenylyl cyclase stimulatory G protein, Gs, and their activation is predicted to elevate cAMP levels in the DH. Consistent with this idea, cAMP antagonists impair retrieval when infused into the DH shortly before testing contextual fear, and DH infusion of cAMP agonists rescues retrieval in gene-targeted mice lacking NE[153].

Further, activation of at least two targets of cAMP signaling, protein kinase A (PKA) and the exchange protein activated by cAMP (Epac), is also required for hippocampus- dependent memory retrieval. Finally, transgenic mice expressing an invertebrate, Gs- coupled octopamine receptor exhibit both octopamine-dependent increases in DH cAMP and enhancement of contextual fear memory retrieval[238].

In contrast to the above, it has been reported that stimulation of 1 receptors impairs memory retrieval. In rats, infusion of the 1-selective antagonist atenolol blocks the impairing effects of a glucocorticoid agonist on retrieval of spatial reference memory[141]. In support of this, injection of 3 or 10 mg/kg of the partial 1 agonist xamoterol decreases spatial reference memory retrieval. The authors conclude that the impairment of hippocampus-dependent memory retrieval by glucocorticoids and xamoterol involves the activation of 1-coupled cAMP signaling. This observation is surprising because administration of xamoterol at 1-3 mg/kg fully rescues contextual memory retrieval in gene-targeted NE-deficient mice, suggesting that 1 signaling is sufficient for the role of NE in retrieval[92].

Atenolol received approval for use in U.S. patients in 1981 and is currently used to treat congestive heart failure and hypertension. Clinical application of xamoterol began in the 1980’s and is prescribed today as a cardiac stimulant and hypotensive agent[244-

246]. In the late 1970s – early 1990s, radiolabeled agents shown to block receptors with high affinity and decent selectivity were used to first localize and quantify -AR subtypes in tissues including the heart, lung, spleen and cerebral cortex[247]. The ability of

agonists (xamoterol), agonists (procaterol, zinterol), antagonists (atenolol,

3 125 betaxolol) and antagonists (ICI 118551) to displace agents like H-DHA and I-CYP became a widely used method of determining specificity of these agents (125I-CYP also binds 5-HT receptors)[243, 248, 249].

Several studies have demonstrated modest specificity for xamoterol and atenolol in CNS ex vivo preparations[243, 248, 249]. For example, xamoterol shows 10-fold specificity for vs. receptors, which is all right considering that antagonists at receptors display 4-6 times greater selectivity[243, 248, 249]. Unfortunately, xamoterol is still amongst the most selective agonists available, limiting its ability to study selective effects pertaining to learning and memory and diminishing its ability to treat heart disease because of concurrent  effects on the vasculature[245]. Likewise, atenolol’s selectivity for  is rather pedestrian (15-29 fold preference for vs. )[248, 249]. In contrast, betaxolol displays a 2.2 log differential at opposed to , supporting its superiority as a

 antagonist when directly compared to atenolol[248]. Moreover, within-study comparison of atenolol with other  antagonists illustrates that atenolol is much less selective than several  blockers (acebutolol, practolol, I-metoprolol)[248, 249].

Betaxolol is amongst the most-specific, supporting its utility to selectively block  receptors in order to discriminate between xamoterol’s effects at and -ARs.

[Notably, ICI 118,551 displays a 100 fold selectivity for  and is amongst the most selective  antagonists available][248].

The goal of this study, then, was to resolve the seemingly conflicting observations that 1 signaling is required for hippocampus-dependent memory retrieval and is also responsible for the impairment of retrieval. Of note, somewhat higher doses of xamoterol were needed to impair retrieval in wild-type rats as compared to those needed to rescue retrieval in mutant mice. This could be due to differences in genetics, species or mechanisms, for example. To address this, we first confirmed that xamoterol impairs 75 memory retrieval in wild-type mice, making a difference between species unlikely. As a result, we hypothesized that the pharmacologic characterization of xamoterol-induced retrieval impairment was incomplete, and that another adrenergic receptor might mediate this effect. In this study, animals were tested in both Pavlovian and instrumental fear conditioning paradigms. Xamoterol and other pharmacologic reagents previously untested in these paradigms were administered to wild-type mice, as well as to mice genetically modified to lack either 1- or 2-adrenergic receptors. Our results provide substantial and unexpected evidence that signaling by Gi/o-coupled 2-adrenergic receptors in the hippocampus mediates xamoterol-induced impairment of memory retrieval.

3.3 MATERIALS AND METHODS

3.3.1 Subjects

Wild-type, β1 KO and β2 KO mice were on a hybrid 129/Sv x C57BL/6 background, and were derived by mating either heterozygotes or homozygotes[226, 227].

Gender and parental genotype did not affect the results, so data were combined.

Genotype was determined by PCR. Mice were maintained on ad lib food and water and a

12 h light:dark cycle, with lights on beginning at 07:00. Animals were housed in relatively small, quiet rooms for 3-4 weeks before studies began to minimize the stress associated with caretaking and colony management during the light phase. Mice were 3-

6 months old when tested. Studies were performed during the light phase, with most experiments taking place between 08:00 - 15:00. Studies were in accordance with NIH guidelines and had the approval of the IACUC at the University of Pennsylvania. 76

3.3.2 Pavlovian Fear Conditioning

Adjacent to the training room, animals were placed in pairs into opaque holding buckets (~12 cm diameter) with bedding and covers for 30-60 min before being manipulated further. Animals were given two 3-min handling sessions over 2 days in the training room. Saline was injected at the end of handling each day. Training consisted of placing the animal in the conditioning apparatus (ENV-010MC with ENV-414S, Med

Associates, St. Albans, VT) for 2 min, after which an 84 dB, 4.5 kHz tone was activated for 30 s. Two seconds before the end of the tone, a 2 s, 1 mA footshock was delivered.

The animal was returned to its home cage 30 s after shock. Contextual fear was tested for

5 min in the conditioning apparatus in the absence of the tone. Cued fear was tested in a context containing distinct visual, tactile and olfactory cues. After 2 min, the training tone was turned on for 3 min. Percent freezing was estimated by scoring the presence or absence of non-respiratory movement every 5 s. Tests were conducted 1 day after training except where indicated.

3.3.3 Instrumental Fear Conditioning

Animals were handled as described above. Training consisted of placing a rodent in the lighted chamber of the apparatus used for Pavlovian conditioning and timing its latency to enter (except for the tail) the dark chamber. Once the animal entered the dark chamber, the retractable partition separating the two chambers was lowered and a shock was delivered for 2 s (0.30-0.35 mA). After waiting 15 s, rodents were returned to their home cage. Animals that didn’t enter the dark chamber after 100 s during conditioning were excluded. Testing was identical to training except that no shock was delivered and 77 the partition remained up. Latencies to enter the dark chamber were recorded. If an animal did not enter the dark chamber within 10 min, it was returned to its cage and assigned a latency of 10 min. Tests were conducted 1 day after training except where indicated.

3.3.4 Drugs

Betaxolol HCl, ICI 118,551 HCl and/or xamoterol hemifumarate (Tocris,

Ellisville, MO) were dissolved in 0.9% saline and administered subcutaneously 30 min before testing. For DH infusion, xamoterol and pertussis toxin (Tocris) were dissolved in

0.9% saline.

3.3.5 Dorsal Hippocampus Infusion

A double guide cannula (C235 system, Plastics One, Roanoke, VA) was implanted under pentobarbital anesthesia (72.5 mg/kg ip) using a stereotax

(SAS75/EM40M, Cartesian Research, Sandy, OR). The guide was placed −1.7 mm AP and 1.5 mm bilateral for DH infusions. The guide projected 1.5 mm below the base and the dummy cannula extended 0.5 mm below the guide. One week after surgery, bilateral infusions were made into conscious mice while gently holding the nape of the neck. The dual injection cannula extended 0.9 mm below the guide. Infusions were 1 μl/side at 0.4

μl/min, and the injection cannula was left in place for 30 s before the mouse was returned to its home cage. Because studies indicate that the effects of PTx are best evaluated 3 days after infusion[213, 214], PTx was infused 3 days before testing (i.e. 1 day before training, which was 2 days before testing).

3.3.6 Statistics

3.4 RESULTS

3.4.1 Xamoterol Impairs Hippocampus-dependent Memory Retrieval

To confirm the impairing effects of xamoterol on hippocampus-dependent memory retrieval, mice were trained using either Pavlovian or instrumental fear conditioning. When 1 or 6 mg/kg of xamoterol was administered 30 min prior to testing, wild-type (WT) mice showed reduced contextual freezing and shorter avoidance latencies one day after training (Fig. 3.1A & 3.3A).

To determine if xamoterol acts directly in the hippocampus to impair retrieval, cannulae were implanted in mice so that bilateral DH infusions could be performed. One week after surgery, cannulated mice were fear conditioned. One day later either saline or xamoterol was infused 15 min before testing. Xamoterol decreased contextual freezing with a maximal effect at 10 g (Fig. 3.1B).

To assess whether xamoterol-induced impairment of contextual fear was specific to memory retrieval, xamoterol (6 mg/kg) was administered 30 min prior to testing cued fear 1 day after training. Xamoterol did not affect cued fear expression, indicating that it does not alter fear or freezing per se (Fig. 3.1C).

Context Context A 80 B 80 60 * 60 ^ * * 40 # 40 *

20 20

Freezing (%) Freezing (%) Freezing 0 0 0 0.6 1 3 6 20 0 5 10 15 Xamoterol (mg/kg) DH Xamoterol (g)

Cue C 80

20 Freezing (%) Freezing 0 0 6 Xam (mg/kg)

FIGURE 3.1 Xamoterol Impairs Hippocampus-dependent Memory Retrieval

(A) Mice were fear conditioned and one day later vehicle or xamoterol was administered subcutaneously 30 min before testing contextual fear. Systemic xamoterol caused of dose- dependent impairment (4-10/group, P < 0.001 for the main effect of dose). (B) Bilateral cannulae targeting the DH were implanted in mice. One week later the mice were fear conditioned, and on the following day they were tested for contextual fear. DH infusions were performed 15 min before testing. DH xamoterol caused of dose-dependent impairment (4-

6/group, P < 0.05 for the main effect of dose). (C) Mice were fear conditioned and one day later vehicle or xamoterol was administered subcutaneously 30 min before testing auditory fear. Xamoterol did not affect cued fear expression, indicating that it does not alter fear or freezing per se (5-8/group, P > 0.9). For all figures: *, P < 0.05; ^, P < 0.01; #, P < 0.001 80 for post-hoc comparisons, and systemic injections occurred 30 min prior to testing one day after training unless stated otherwise.

3.4.2 Pharmacologic Evidence that Xamoterol Disrupts Retrieval through 2- and not 1-adrenergic Signaling

The above data indicate that 0.6 mg/kg xamoterol does not impair contextual memory retrieval. In comparison, 0.3 mg/kg xamoterol induces significant rescue of contextual memory retrieval in NE-deficient Dbh-/- mice[92]. This difference could be

-/- the result of a potentially greater sensitivity to 1-adrenergic agonists in Dbh mice.

Alternatively, the impairment of retrieval could be due to effects of xamoterol that are independent of 1 signaling. In this regard, a closely related receptor to 1 is the 2- adrenergic receptor. To test the idea that higher doses of xamoterol might stimulate 2 receptors, a selective antagonist of 2 receptors (ICI 118,551) was employed. In agreement with previous results[92], 1 mg/kg ICI 118,551 had no effect on contextual memory retrieval 1 day after training (Fig. 3.2A). However, co-injection of the 2 antagonist with xamoterol (6 mg/kg) blocked the impairment of contextual fear in a dose- dependent manner, with 1 mg/kg ICI producing full rescue of retrieval (Fig. 3.2A). To determine whether ICI non-specifically elevates freezing when freezing is relatively low, mice were conditioned using a less intense shock (0.35 mA) and saline or ICI was administered 30 min prior to testing. Contextual fear did not differ between the groups, indicating that ICI does not enhance fear or freezing per se (Fig. 3.2B).

In the above experiments, the use of contextual fear was limited by the fact that 1 antagonists impair contextual memory retrieval on their own. To directly compare the effects of 1 and 2 antagonists on the retrieval impairment induced by xamoterol, mice 81 were subjected to instrumental fear conditioning. Similar to the results for contextual fear, 1 mg/kg ICI did not reduce retention latencies; however, it completely prevented the impairment in retrieval induced by xamoterol (Fig. 3.2C). Further, the selective 1 antagonist betaxolol (1 mg/kg) did not affect retention latencies when administered alone.

In addition, betaxolol did not prevent xamoterol-induced disruption of inhibitory avoidance (Fig. 3.2C), even though 1 mg/kg betaxolol is a maximally effective dose for blocking the retrieval of contextual fear memory[92].

Context B Context A 80 80

60 60

40 # 40

20 20

Freezing (%) Freezing Freezing (%) Freezing 0 0 ICI (mg/kg): 0 0 0.3 1 2 1 0 1 Xam: - + + + + - ICI (mg/kg)

C Avoidance Veh 8 * Xam 6 Xam + ICI 4 ^ * Xam + Bet

2 ICI Latency (min) Latency 0 Bet 0 6 6+1 6+1 1 1 Drug (mg/kg)

FIGURE 3.2 Impairment of Retrieval by Xamoterol is Prevented by a 2 but not a 1 Antagonist

(A) Co-injection of xamoterol (Xam, 6 mg/kg) with the 2 antagonist ICI 118,551 (ICI, mg/kg) blocked xamoterol-induced impairment of contextual fear, with 1 mg/kg ICI completely preventing disruption of retrieval. ICI (1 mg/kg) alone had no effect on retrieval (4-6/group, P

< 0.01 for the main effect of treatment). (B) ICI does not enhance low levels of contextual fear in the absence of xamoterol (4-5/group, P > 0.9). Animals were trained with lower shock intensity (0.35 mA) to approximate xamoterol-impaired freezing. (C) Xamoterol (6 mg/kg) reduced retention latencies for inhibitory avoidance relative to saline (Sal). Co-administration of xamoterol with ICI (1 mg/kg) prevented xamoterol-induced disruption of avoidance, while co-injection of xamoterol with the 1 antagonist betaxolol (Bet, 1 mg/kg) failed to prevent disruption of retrieval. Neither Bet nor ICI altered avoidance when administered alone (6-

22/group, P < 0.01 for the main effect of treatment).

3.4.3 Genetic Evidence that Xamoterol Disrupts Retrieval through 2- and not 1- adrenergic Signaling

We sought to confirm the pharmacologic evidence suggesting that xamoterol is operating through 2 and not 1 receptors to impair retrieval. Toward this goal, we examined the effects of xamoterol on retrieval in mice lacking either 1- or 2-adrenergic receptors. Instrumental fear conditioning was used because 1 KOs exhibit impaired contextual memory retrieval[92]. 1 KO and 2 KO mice displayed normal retention latencies when treated with saline, which allowed for the examination of drug affects

(Fig. 3.3A-C). In WT mice, xamoterol significantly reduced avoidance latencies at 1 and

6 mg/kg. Importantly, xamoterol did not reduce retention latencies in 2 KOs at any of these doses, indicating that 2 signaling is necessary for the effects of xamoterol on retrieval. In contrast, xamoterol at 0.3 and 1 mg/kg impaired avoidance in 1KOs, indicating that 1 signaling is not required for the disruption of retrieval by xamoterol.

Interestingly, there was about a three-fold leftward shift in the U-shaped dose-response curve for xamoterol in the absence of 1 receptors.

A Avoidance WT B Avoidance 1 KO 8 8

6 6 4 * * 4 * *

2 2 Latency (min) Latency 0 0 0 0.1 0.3 1 6 20 0 0.1 0.3 1 6 Xamoterol (mg/kg) Xamoterol (mg/kg)

C Avoidance  2 KO 8

0 0 0.1 0.3 1 6 Xamoterol (mg/kg)

FIGURE 3.3 Genetic Evidence that 2 Signaling Mediates the Impairment of Memory Retrieval by Xamoterol

(A) Xamoterol at 1 or 6 mg/kg maximally disrupted avoidance in WT mice (7-10/group, P <

0.05 for the main effect of dose). (B) Xamoterol at 0.3 and 1 mg/kg maximally disrupted avoidance in 1 KO mice (9-24/group, P < 0.05 for the main effect of dose). (C) Xamoterol did not impair avoidance at any dose when tested in 2 KO mice, indicating that 2 activation is required for xamoterol-induced disruption of memory retrieval (7-9/group, P > 0.7 for the main effect of dose).

3.4.4 Pertussis Toxin Blocks the Impairing Effects of Xamoterol on Retrieval

The traditional view is that -adrenergic receptors couple to and signal via the adenylyl cyclase-stimulatory G protein Gs. However, there is evidence that 2 receptors can also signal through Gi/o[210-212]. Because cAMP signaling is necessary and sufficient for contextual memory retrieval mediated by NE and 1 receptors[153], we asked whether 2 signaling might oppose 1 and Gs signaling in retrieval by activating

Gi/o.

To test this idea, pertussis toxin (PTx) was employed because it uncouples Gi/o proteins from their receptors by ADP-ribosylation. Because PTx can take several days to be fully effective in the hippocampus in vivo[213, 214], PTx was infused into the DH one day before fear conditioning to minimize potential effects on acquisition and consolidation. Mice were tested for retrieval two days after conditioning. Pretreatment with PTx did not affect retrieval in mice infused with vehicle 15 min before testing (Fig.

3.4). Strikingly, however, PTx pretreatment completely prevented the impairment of retrieval induced by xamoterol.

Context Sal DH Pretreatment: PTx 80 * 60

40 *

20 Freezing(%) 0 Sal Xam DH Test Treatment

FIGURE 3.4 Gi/o Signaling is Required for the Impairment of Retrieval by Xamoterol

Bilateral cannulae targeting the DH were implanted in mice, and one week later the mice were infused with either saline (Sal) or pertussis toxin (PTx, 10 ng/side) into the DH. The next day the mice were fear conditioned, and two days later the mice were infused with either saline or xamoterol (Xam, 10 g/side) into the DH. Contextual fear was tested 15 min later. Pertussis toxin pretreatment blocked the impairing effect of xamoterol on retrieval, indicating that this effect is mediated by the Gi/o class of G proteins (5/group, P < 0.05 for the main effect of treatment, and P < 0.05 for the interaction between pretreatment and treatment).

3.4.5 Xamoterol Impairs Hippocampus-dependent Memory Retrieval in a Time- limited Manner

NE, 1, cAMP and PKA signaling all play a time-limited role in hippocampus- dependent emotional memory retrieval[92, 153]. NE-deficient mice display deficits in contextual fear 0.1 - 4 days but not 5 or more days after training, and mice and rats treated with a antagonist show compromised contextual fear and spatial navigation over

86 a similar time course. Therefore, we determined if 2 signaling mediated by xamoterol might also have time-limited effects on retrieval. Contextual fear did not differ between days 1, 3 and 5 in vehicle-injected animals (Fig. 3.5). However, the effects of xamoterol were time-sensitive, with animals displaying the greatest impairment 1 day after training, less impairment 3 days after training and no impairment 5 days after training. The results indicate that xamoterol affects an intermediate phase of memory retrieval but not contextual information processing per se.

Context Sal Xam (6 mg/kg) 80

60 # 40 #

20 Freezing(%) 0 1 3 5 Days After Training

FIGURE 3.5 Xamoterol Impairs Hippocampus-dependent Memory Retrieval in a Time- limited Manner

Xamoterol was administered 30 min prior to testing at 1, 3 or 5 days after training.

Xamoterol impaired contextual fear at 1 and 3 but not 5 days after training (5-10/group, P <

0.001 for the main effect of treatment, P < 0.001 for the main effect of day, and P < 0.001 for the interaction between treatment and day).

3.5 DISCUSSION

A role for NE and 1-adrenergic signaling in hippocampus-dependent emotional memory retrieval is supported by multiple experimental outcomes. NE-deficient Dbh-/- mice exhibit impaired retrieval of contextual and spatial reference memory, and contextual memory retrieval is rescued in these mice by the administration of the 1- selective agonist xamoterol[92, 93]. Further, 1 KO mice exhibit impaired contextual memory retrieval, as do wild-type mice treated with one of several 1 antagonists. In addition, the locus coeruleus has been found to degenerate in people with Down syndrome[239, 240]. In mouse models of Down syndrome such as Ts65Dn, there is less brain NE, fewer locus coeruleus adrenergic neurons and more  receptor-positive neurons in the hippocampus, suggesting the presence of functional NE deficiency[241, 242].

Similar to Dbh-/- mice, Ts65Dn mice display deficits in contextual fear memory retrieval but not in cued fear memory, and administration of a NE precursor or the 1-selective agonist xamoterol reverses the impairment of memory retrieval in Ts65Dn mice[242].

The goal of this study was to resolve the conflicting ideas that 1 signaling is required for the retrieval of hippocampus-dependent emotional memory as outlined above, and is also responsible for the impaired retrieval of such memories by xamoterol[141]. In Roozendaal et al., doses of 3 and 10 mg/kg were used to demonstrate retrieval impairment. In contrast, xamoterol rescues contextual memory retrieval in NE-

-/- deficient Dbh mice at 0.3 and 1 mg/kg. The potency of xamoterol for 1 versus 2 receptors is ~10-fold in rat tissue ex vivo and at mouse adrenergic receptors in vitro[92,

243]. As a result, we hypothesized that the higher doses of xamoterol activate 2 as well

88 as 1 receptors, and that the activation of 2 receptors impairs retrieval. In the current study, we found that xamoterol impairs memory retrieval in the presence of a 1 antagonist and in 1 KO mice (Figs. 3.2C and 3.3B), indicating that 1 signaling is not required for the impairment of retrieval by xamoterol. In contrast, we found that the impairing effects of xamoterol were greatly reduced both in the presence of a 2 antagonist and in 2 KO mice (Figs. 3.2 and 3.3C). These results demonstrate that xamoterol impairs retrieval through the stimulation of 2 receptors. In support of a critical role for 2 receptors, we have also found that low doses of the 2 receptor agonists procaterol and zinterol impair memory retrieval in mice and rats (Chapter 2).

Classically, -adrenergic receptors couple to the adenylyl cyclase-stimulating G protein Gs, and their activation increases cAMP levels. Given that 1 receptors and cAMP signaling in the hippocampus are required for hippocampus-dependent emotional memory retrieval[92, 153], it was not apparent how activation of 2 receptors in the hippocampus would impair retrieval. One possibility was that these two receptor subtypes are expressed on different cells or in distinct subcellular compartments, where increases in cAMP would have opposing effects on processes important for retrieval.

Based on the above, the simplest explanation for the impairing effects of xamoterol is that they oppose the increase in cAMP that is necessary for retrieval. This idea is consistent with the observation that xamoterol was more potent at impairing memory retrieval in 1 KO mice as compared to wild-type mice (Fig. 3.3). In other words, when both receptors are present, higher levels of 2 stimulation are required to offset the stimulation of 1 receptors. With regard to the mechanism of retrieval impairment, 2 signaling through Gi/o has been linked to effects that may be independent of or in addition to a constraint on cAMP levels[217], and it is possible that some of these effects are also relevant to the impairment of retrieval. Additional studies will be needed to address this. Highlighting the contrasting roles of these two receptors, 2 signaling in the heart acts to protect cardiomyocytes from excessive 1 stimulation that can lead to apoptosis and heart failure if unchecked[206, 207].

Interestingly, our data demonstrate that the duration between acquisition and retrieval is an important factor for determining the extent of retrieval impairment (Fig.

3.5). The time course for the effects of xamoterol on retrieval was essentially identical to that for the requirement of NE, 1, cAMP and PKA signaling in retrieval[92, 153], i.e. an intermediate phase of memory was susceptible. These findings are consistent with the idea that 2 signaling acts to oppose the role of 1 signaling in retrieval.

A potentially important aspect of -adrenergic receptor physiology is that NE, the adrenergic neurotransmitter in the DH, is an order of magnitude more potent in activating

1 receptors than 2 receptors[222]. As a result, it seems likely that moderately arousing conditions would facilitate hippocampus-dependent emotional memory retrieval through

90 the release of NE and activation of 1 receptors. Extending this idea, strongly arousing conditions typical of ―stress‖ would impair memory retrieval through the release of additional NE and also glucocorticoids that together would activate 2 receptors. Other studies have demonstrated that glucocorticoids impair hippocampus-dependent emotional memory retrieval[123, 124]. Studies from our lab indicate that this effect of glucocorticoids is mediated by activation of the 2 receptor, and that NE and glucocorticoids synergize to activate this receptor at doses that would be ineffective when present individually (Chapter 2). Because strongly arousing conditions enhance memory consolidation, this could facilitate a temporary shift away from processes favoring retrieval toward those favoring consolidation.

3.7 ACKNOWLEDGEMENTS

We thank B. Kobilka for providing stock for the  receptor KO mouse lines. This work was supported by NIH grant 5R01MH063352 and DOD grant PT075099 to S.A.T, and NIH grant 5T32GM008076 to R. Pittman.

Chapter 4

Roles of 2- and 3-adrenergic Signaling in Aversive, Hippocampus- dependent Memory Retrieval

Keith Schutsky & Steven A. Thomas

Department of Pharmacology, University of Pennsylvania, Philadelphia, PA 19104

4.1 ABSTRACT

Similar to 1 receptors, additional receptors can influence fear memory.

Pharmacologic and genetic evidence support a transient and novel role for 2 signaling in intermediate-term hippocampus-dependent memory retrieval, identical to the time-course of retrieval impairment exhibited by NE, 1, cAMP, PKA and Epac signaling[92, 153].

Moreover, 2 signaling may function in the maintenance of short-term aversive memory.

Conversely, multiple experimental outcomes do not support a general requirement for 2 signaling in the long-term acquisition, consolidation or retrieval of emotional memory.

The finding that 2 agonists and antagonists impair retrieval to a similar extent suggests that specific confirmation(s) of the 2 receptor might be imperative in affecting retrieval processes. This is supported by evidence that glucocorticoids may impair recall through either direct interaction with the 2 receptor, inducing conformational changes in 2 that prevent signaling necessary for retrieval, or through coincident signaling between 2 and

GR. Moreover, 3 receptors can influence retrieval processes and receptor interactions may influence the extent of retrieval deficits displayed. Finally, the net effect of

receptor stimulation (or antagonism) on hippocampal cAMP signaling may determine the degree of retrieval impairment expressed, as individual receptors can have similar or opposing effects on cAMP levels.

4.2 INTRODUCTION

Over the last decade, it was demonstrated that 1-adrenergic signaling is necessary for hippocampus-dependent memory retrieval mediated by NE/E[92, 153]. In

93 addition, I recently discovered that additional  receptors, including 2 and 3, influence retrieval processes both individually and in conjunction with 1. Results by our lab demonstrate that systemic administration or DH infusion of several 2 agonists (and a partial 1 agonist with 2 activity) impair the retrieval of both IA and contextual fear

(Chapter 2-3). In part, this observation served as the impetus for a more detailed and thorough examination of the roles of 2- and 3-adrenergic signaling in aversive memory.

Experiments accessing the function of 2 and 3 in learning are assisted by availability of subtype selective receptor agonists and antagonists. Further, 2KO and 3KO animals have been generated, enabling the utilization of both genetic and pharmacologic approaches to study potential roles of 2 and 3 signaling in fear memory.

4.3 MATERIALS AND METHODS (see section 2.5, 3.3)

4.4 RESULTS & DISCUSSION

4.4.1 2 Signaling and Fear Memory

Accumulating evidence suggests that 2-adrenergic signaling plays a time-limited role in the retrieval of hippocampus-dependent memory. 2KO mice express deficits in contextual fear beginning as early as 30 min that last approximately 3-4 d, similar to the time-course of impairment for NE, 1, cAMP, PKA and Epac signaling[92, 153] (Figure

4.1A). However, 2 signaling may be initially required at earlier times (~30 min after acquisition) than when NE or 1 signaling is necessary (~2 hrs)[92] (Figure 4.1A). This could be related to a deficit in the maintenance of short-term memory. Consistent with this is the observation that systemic administration of a 2 antagonist (ICI) 30 min before

94 training impairs contextual fear at 1 hr, but not 2 hrs post-conditioning (Figure 4.1B)., indicating acquisition is not affected by blockade of 2 signaling. Interestingly, administration of ICI 30 min post-training does not affect the expression of context fear

30 min later (Figure 4.1C), whereas ICI administered immediately after training impairs freezing 1 hr later (Figure 4.1D). Thus, there appears to be a brief time window after acquisition through which 2 signaling influences the expression of short-term memory

(STM). Additional studies are needed not only to confirm this result but also to address whether administration of ICI pre- or post-training affects the acquisition or expression of hippocampus-independent cued fear, a paradigm that controls for potential non-specific effects of ICI on fear and motor performance.

Interestingly, retrieval deficits expressed by 2KOs dissipate completely at 5 days and beyond, supporting a time-limited role for 2 signaling in intermediate-term retrieval rather than in long-term acquisition or consolidation of aversive fear memory (Figure

4.1A). A ―spontaneous recovery‖ of contextual fear at later times is not consistent with impaired consolidation, which would not be expected to recover, but is consistent with a temporary impairment of memory retrieval. Further, 2KO and 12 DKO animals, compared to their WT counterparts, express equivalent cued fear 1 d after training, arguing against a general requirement for 2 signaling in fear memory consolidation

(Figure 4.1E). 2KO mice also express normal short and long-term retention of inhibitory avoidance (IA), independent of training intensity (Chapter 2-3). IA would be impaired if consolidation processes were affected. (Notably, however, 2 signaling can enhance the consolidation of fear memory, analogous to increasing endogenous NE systemically or

95 centrally within in the BLA, hippocampus or NTS[10, 11, 73, 78, 250][communications with M. Ouyang and S.A. Thomas]).

Consistent with genetic results, pharmacologic evidence suggests that 2 signaling plays a specific role in intermediate-term, aversive memory retrieval. Systemic administration of ICI 118551, a selective 2 antagonist, 1 d after training dose- dependently reduces contextual fear in control mice in a u-shaped fashion, with 10 g/kg producing maximal impairment (Figure 4.1F). The extent and nature of the impairment may be related to retrieval deficits induced by stress, GCs or 2 agonists (Chapter 2), altering common effectors or signaling pathways influenced by 2 signaling. Future studies are needed to address this. A u-shaped dose-response phenomenon may be observed because low-moderate concentrations of ICI selectively bind to and change the confirmation of the receptor, causing retrieval-mediated deficits whereas higher concentrations of ICI also antagonize the 1 receptor, negating the compound’s effectiveness in impairing retrieval. The discovery that CNS-based 1 and 2 signaling can oppose one another is supported by multiple experimental outcomes, summarized as follows: 1) 12 DKO mice, compared to control animals, equivalent levels of contextual fear (Chapter 2); 2) Treatment with high-dose ICI 1d after training produces no affect on the retrieval of IA (Chapter 2); 3) 2 agonist mediated retrieval deficits are likely u- shaped because higher concentrations stimulate 1 receptors which support retrieval

(Chapter 2); and 4) selective stimulation of either 1 or 2 in the DH produce opposite effects on cAMP signaling and combined stimulation of both receptors does not significantly alter levels of cAMP (Chapter 2). In addition, direct infusion of ICI into the

96 dorsal hippocampus 1d after training dose-dependently mirrors the affects of systemic administration, with 50-200 ng producing disruption of contextual fear (communications with M. Ouyang and S.A. Thomas). Consistent with genetic results, pharmacologic evidence also supports a transient role for 2 signaling in CF retrieval. The time-course of

ICI impairment mirrors that of 2KOs, NE, 1, cAMP, PKA and Epac signaling (~2 hr –

4 days)[92, 153](Figure 4.1E). Future studies will determine if ICI administered 1 or 3 d after training affects cued fear in WT animals. If it does not, this would support a selective role for 2 signaling in hippocampus-dependent, retrieval-based processes. In sum, genetic and pharmacologic data converge to convey a time-limited role for 2 signaling in the retrieval of hippocampus-dependent fear memory.

Train Test

0 Varies (min - days)

A Context WT 2KO 80

60 # # 40 * # #

20 Freezing(%)

1 Hr 1 Hr 1 Day 1 Day 1WK 1WK 30 Min30 Min 3 Days3 Days 5 Days5 Days

1 Hr (No1 Hr Tone) (No Tone) Time After Training

Inj Train Test Train Inj Test

-30 min 0 min 1 or 2 hr 0 min 30 min 1 hr

Sal Sal (WT) (WT) B Context ICI (10 g/kg) C Context ICI (10 g/kg) 80 80 ICI (1 mg/kg)

60 60 40 ^ 40

20 20

Freezing(%) Freezing(%)

0 0

1 Hr 1 Hr 2 Hr 2 Hr 1 Hr 1 Hr 1 Hr Time After Training Time After Training

Train Test

Train Inj Test 0 24 hr

0 min 1 min 1 hr WT 2KO Cued   DKO D Context (WT) E  2 80 80

60 60 Sal 40 ^ ICI (10 g/kg) 40

20 20

Freezing(%) Freezing(%)

0 0

1 Hr 1 Hr Time After Training

Train Inj Test

0 24 hr 24.5 hr

F Context (WT) Sal ICI 80

60 * ^ 40 ^ ^

Freezing(%) 20

0 0 5 10 20 30 100 1000 Treatment (ug/kg)

Train Inj Test

0 Varies (min - days) Inj + 30 min

Sal ICI (10 g/kg) G Context (WT) ICI (1 mg/kg) 80 60 * 40 ^ ^

20 Freezing(%)

1 Hr 1 Hr 2 Hr 2 Hr 1 Day 1 Day 3 Days3 Days 5 Days5 Days

Time After Training

Figure 4.1 2 Signaling & Fear Memory

(A) 2KO mice exhibit impaired contextual fear (CF) 30 min to 3 d post-conditioning, but not at

5 d and beyond. P < 0.001 for the main effect of genotype on freezing, P < 0.001 for the main

99 effect of time on freezing and P < 0.001 for the interaction of genotype and time (4-

10/group). * P < 0.05, ^ P < 0.01, # P < 0.001 for post-hoc comparisons. (B) ICI administration 30 min prior to conditioning impairs CF at 1 hr, but not 2 hrs post-injection. P <

0.05 for the main effect of treatment on freezing, P < 0.01 for the main effect of time on freezing and P < 0.01 for the interaction of treatment and time (4-5/group). ICI was systemic administered in all figures, unless otherwise noted. (C) ICI administration 30 min after conditioning does not impair CF 30 min later (4-5/group). (D) ICI administration immediately after training inhibits CF 1 hr post-injection. P < 0.01 for the main effect of treatment on freezing (5/group). (E) 2KO and 12 DKO mice do not exhibit deficits in cued fear 1 d after training, arguing against a general role for 2 signaling in freezing, fear or aversive memory consolidation(4-5/group). (F) ICI, a 2 antagonist, dose-dependently inhibits CF 1 d after training in WT mice. P < 0.01 for the main effect of dose (5/group). (G) WT mice receiving ICI display deficits in CF 1 hr to 3 d post-conditioning, but not at 5 d. P < 0.001 for the main effect of treatment on freezing, P < 0.09 for the main effect of time on freezing and P < 0.05 for the interaction of treatment and time (4-5/group).

Somewhat paradoxical is the result that 2 agonists and antagonists both transiently impair the recall of fear memory (Chapters 2, 4). One could speculate this occurs for a number of reasons. Signaling pathways and effectors affected by 2 signaling may also influence retrieval processes in a U-shaped fashion. For example, inactivating the receptor may change cAMP signaling or reduce signaling by other pathways (i.e.

ERK and/or PI3K) that are necessary for retrieval[251]. Conversely, overstimulation of the receptor can decrease cAMP signaling, thus potentially decreasing signaling by other effectors, including PKA, Epac and Rap2, which are required for retrieval[153](Chapter

2)(communications with M. Ouyang & S.A. Thomas). Alternatively, excessive activation of 2 could cause hypersensitization and rapid internalization of the receptor[252] whose

100 proper expression at the surface of the membrane may influence the ability to retrieve information. Finally, altering 2 signaling could throw off a delicate balance necessary between 2 and other  receptors in regulating retrieval processes.

4.4.2 Evidence for Coincident Signaling or Direct Molecular Interaction between GR and 2-adrenergic Signaling in Aversive, Hippocampus-dependent Memory Retrieval

Several lines of evidence suggest that the conformation of 2 receptor itself is important to the retrieval process. For example, systemic administration or DH infusion of a GR-antagonist 1d after training prevents 2 agonist-mediated impairment of contextual fear (Figures 4.2A & 4.2B). Preliminarily, this would suggest that GR signaling lies downstream of 2 signaling. However, co-administration of a 2 antagonist completely mitigates GR agonist-induced impairment of both IA and contextual fear, suggesting 2 signaling may be downstream of GR-signaling (Chapter 2). Taken together, these results suggest either 1) coincident signaling by 2 and GR; 2) a direct molecular interaction whereas conformational changes in GR can induce simultaneous changes in the 2 receptor; or 3) that GR-like compounds can bind directly and activate or deactivate the 2 receptor. The latter two possibilities seem more likely than the former because GCs can impair retrieval in the absence of NE-signaling (Chapter 2). A direct molecular interaction model is also supported by recent findings that insect ecdysteroids can directly bind to and activate a dopamine GCPR that is most similar in sequence homology to  adrenergic receptors[253]. However, a coincident signaling model cannot be excluded if constitutive signaling by 2 receptor is the absence of ligand (NE) influences retrieval

101 process. The observation that contextual retrieval deficits exhibited by NE-deficient mice are reversed by administration of ICI is consistent with a role for constitutive signaling at the 2 receptor (Chapter 2). Future studies will attempt to discriminate between coincident signaling and direct molecular interaction models (5.2.3).

Proc Proc Org + Proc RU486 + Proc

Context (WT) A Context (WT) B 80 # 80 * * ^ 60 60

40 40

20 (%) Freezing 20

Freezing (%) Freezing 0 0

0.01 2000 5 + 2000 15 + 0.01 30 + 0.01 10 + 200025 + 2000 Treatment (mg/kg) Treatment (ng)(DH)

Figure 4.2 GR Antagonists Prevent 2 Agonist-Mediated Disruption of Hippocampus- Dependent Memory Retrieval

(A) Procaterol (2 Agonist) impairment of CF is reversed by co-administration of a GR antagonist, Org 34850. P < 0.05 for the main effect of treatment on freezing (4-6/group).

Compounds were administered systemically. (B) DH infusion of procaterol inhibits CF, and procaterol-induced deficits are prevented by co-infusion of a GR antagonist, RU486. P < 0.001 for the main effect of treatment on freezing (2-5/group). Infusions were administered 15 min before testing 1 d after training.

102

4.4.3 Putative Role of 3-adrenergic Signaling in Aversive, Hippocampus-dependent Memory Retrieval

Initial evidence suggests that 3-adrenergic signaling can influence memory retrieval. Systemic administration of a 3 antagonist (SR59230A) 1d after training dose- dependently reduces contextual freezing in 12 DKO mice and a similar concentration of

SR likewise impairs retrieval in control animals (Figure 4.3A). Furthermore, subcutaneous injection of the non-selective  blocker, propranolol, 1d after training inhibits contextual retrieval in 12 DKO mice in a dose-dependent fashion, suggesting propranolol’s affects are due (at least in part) to the antagonism of 3 receptors (Figure

4.3B). Notably, propranolol transiently impairs the retrieval of contextual fear and spatial navigation but not cued fear in rats and control mice, suggesting it selectively influences retrieval rather than consolidation processes[92]. Moreover, doses of SR59230A that impair contextual fear do not significantly impair cued fear in control animals, suggesting its effects on 3 are likely retrieval-based (Figure 4.3C). Additional evidence supports the hypothesis that 3 signaling mediates retrieval processes. Co-administration of a 2 antagonist 1d after training in control mice readily reverses 3 antagonist-induced impairment of contextual fear (Figure 4.3D). In addition, contextual memory deficits exhibited 1d after training by 2KO mice are fully rescued by co-administration of a 3 antagonist (Figure 4.3D). Taken together, these results suggest a general role for 3 in retrieval, because deficits created by inactivating 3 signaling can be prevented by similarly reducing 2 signaling at a time after acquisition and consolidation have already occurred. Molecular analysis of changes in hippocampal cAMP signaling induced by

103 activation or inactivation of 3 signaling supports behavioral analysis that the receptor functions in ways similar to 1[unpublished]. For instance, selective stimulation of either

1 or 3 receptors produces nearly equivalent increases in levels of cAMP[unpublished].

Chapter 5 discusses future experiments involving 3 signaling.

A Context Sal SR 59230A B Context 12 DKO 80 2 DKO WT 60 Sal 60 Propranolol ^ 40 40 * # ^ #

20 20

Freezing (%) Freezing Freezing (%) Freezing

0 0 0 0.5 1 5 0 2 0 0.3 1 3 Treatment (mg/kg) Treatment (mg/kg)

Sal Context ICI (1 mg/kg) + SR 59230A (2 mg/kg) C Cued WT D SR 59230A (2 mg/kg) 80 Sal 60 SR 59230A 60 ^ 40 40

20 20

Freezing (%) Freezing Freezing (%) Freezing

0 0 0 2 WT 2KO Treatment (mg/kg) Genotype

Figure 4.3 3 Signaling Influences Hippocampus-dependent Memory Retrieval

(A) Administration of a 3 antagonist (SR59230A) impairs CF in 12 DKO and WT mice. P <

0.01 for the main effect of treatment on freezing (4-17/group). Experiments in figure 4.3 were performed 1 d after training and systemic injections were administered 30 min before testing. (B) Administration of the non-selective  antagonist propranolol reduces freezing in

12 DKO animals. P < 0.01 for the main effect of treatment on freezing (4-17/group). (C)

SR59230A does not impair hippocampus-independent cued fear in control mice (4-5/group).

(D) Co-administration of a 2 antagonist (ICI) prevents 3 antagonist-mediated inhibition of CF

104 in WT mice. Likewise, freezing deficits exhibited by 2KO mice are reversed by administration of a 3 antagonist. WT: P < 0.01 for the main effect of treatment on freezing. 2KO: WT: P <

0.01 for the main effect of treatment on freezing (3-24/group).

4.4.4 Combined Inactivation of 1, 2 and/or 3-adrenergic Signaling Influences Aversive, Hippocampus-dependent Memory Retrieval

Initial evidence that inactivating all 3 -adrenergic receptors simultaneously impairs hippocampus-dependent memory retrieval comes from studies using propranolol to inhibit both contextual and spatial navigation[92]. Combined pharmacologic and genetic approaches are being employed to access the respective contributions of multiple receptor combinations to the retrieval process. For example, simultaneous co- administration of 1, 2 and 3 antagonists in control animals 1d after training dramatically reduces contextual freezing (Figure 4.4). Likewise, co-injection of a 1 and

3 antagonist in 2KO mice produces similar levels of impairment whereas co- administration of a 2 and 3 antagonist in 1KO mice produces only moderate deficits

(Figure 4.4). Why the latter occurs is unclear, but likely depends on the net effect of combined  antagonism on cAMP levels or other signaling pathways that influence retrieval.

Interestingly, administration of a 2 antagonist in 1KO mice 1d after training fully reverses contextual retrieval deficits whereas administration of a 3 antagonist does not (Figure 4.4). This supports the hypothesis that 1 and 3 signal in a related fashion to facilitate retrieval, and differently than that of 2. Of note, co-administration of a 1 and

3 antagonist in control mice does not impair contextual retrieval to a greater extent that

105 individual antagonism of 1 or 3[92] (Figure 4.4), suggesting antagonism of either receptor is sufficient to inhibit pathways needed for retrieval. On a different note, subcutaneous injection of a 1 antagonist in 2KO mice partially restores contextual freezing (Figure 4.4), supporting the idea that 1 and 2 signaling oppose each other in retrieval. Why partial and not full rescue occurs is unclear but may be related to basal signaling by the 2 receptor (or expression of 2 at the surface of the membrane) that is important for retrieval. Future studies will continue to explore functions of  receptors in fear memory and the intracellular signaling pathways they influence. In addition, regional and subregion-specific floxed -AR mice will eventually assist in determining where in the hippocampus (and in which cell types) individual -ARs are necessary for retrieval

(5.2.4).

106

E Context 80

60 ^

40 #

Freezing (%) Freezing 20 *

0 WT 1KO 2KO Genotype Sal Betax (1) + ICI (1) + SR (2) (mg/kg) = 1+2+3 Betax (1) + SR (2) (mg/kg) = 1+3 ICI (1) + SR (2) (mg/kg) = 2+3 SR (2) (mg/kg) = 3 Betax (1) + SR (2) (mg/kg) = 1+3 Betax (1) (mg/kg) = 1

Figure 4.4 Signaling by Multiple  Receptors Influence Hippocampus-dependent Memory Retrieval

Genetics and pharmacology demonstrate that concurrently eliminating or reducing signaling at different receptor subtypes effect CF in complex ways. WT: P < 0.001 for the main effect of treatment on freezing (4-5/group). 2KO: P < 0.05 for the main effect of treatment on freezing

(2-24/group). Experiments were performed 1 d after training. Injections were administered 30 min before testing.

107

Chapter 5

Summary & Future Directions

Keith Schutsky & Steven A. Thomas

Department of Pharmacology, University of Pennsylvania, Philadelphia, PA 19104

108

5.1 Summary

Evidence from the literature suggests that corticosteroid impairments in hippocampus-dependent memory retrieval are mediated by GRs in a rapid, non-genomic fashion. Consistent with this are our findings that 1) a selective GR agonist impairs retrieval and 2) hippocampal infusion of a GC impairs retrieval within minutes after administration. Interestingly, retrieval deficits induced by stress or administration of GCs

(or xamoterol) mimic effects of  agonists. Strikingly, effects mediated by stress or any one of these agents are fully reversed by concurrent administration of a  antagonist (or absent in KOs), suggesting that signaling by GR and are intimately linked. The discovery that co-administration of a GR antagonist fully mitigates retrieval deficits caused by a  agonist suggests that direct molecular interaction or coincident signaling between GR and  is likely, and aggregate results demonstrate that GR or  signaling is neither upstream nor downstream of one another. Pharmacogenetic experiments will assist in distinguishing between these models (5.2.3).

Remarkably, all aforementioned agents appear to selectively affect hippocampus- dependent memory retrieval, as these compounds produce no effect on the acquisition, consolidation or expression of hippocampus-independent auditory fear memory.

Moreover, the actions of these compounds also appear to be mnemonic in nature because these agents no longer affect memory expression 5 days after conditioning. In addition,

KOs express normal cued fear, a result consistent with the examined compounds that effect signaling.

109

Interestingly, GCs impair hippocampus-dependent memory retrieval in the absence of NE, demonstrated by Dbh-/- mice displaying IA impairment after administration of a GC shortly before testing. Our laboratory has shown that NE is required for retrieval of contextual and spatial memory (as opposed to IA)[92, 93], so confirming these results in additional DH-dependent tasks is difficult (as mice lacking

NE display deficits without GC treatment). It is possible that GC-mediated retrieval impairment in NE/E-deficient mice results from developmental and/or compensatory alterations in HPA function, adrenergic or GR signaling that contribute to the lack of a phenotype. Adrenergic changes are unlikely, because memory deficits displayed by Dbh-

/- mice (in the absence of GCs) are conditionally (and fully) reversed by acute administration of L-DOPS[92].

Although GCs can impair retrieval in the absence of NE, stress-mediated retrieval

-/- deficits require both NE and  signaling, supported by the finding that stressed Dbh and KO mice show significantly less IA impairment when compared to controls. The observation that NE is required for stress, but not for GC-mediated effects is consistent with the idea that stress enhances the release of several compounds in addition to GCs

(e.g. NE) than may synergize with GCs to produce retrieval deficits. Moreover, restraint stress, compared to GC administration, results in a 5- to 7- fold lower rise in plasma levels of cort, supporting the notion that other factors besides GCs contribute to the stress response. Conversely, a lack of stress-induced retrieval inhibition displayed by Dbh-/- mice could be due to developmental or compensatory alterations in HPA function, adrenergic or GR signaling, and a future experiment will test this hypothesis. If stress-

110 mediated deficits displayed by Dbh-/- are retrieval-based, it can be predicted that administering L-DOPS prior to stress should fully reverse the phenotype (e.g. KO retrieval will be impaired at levels similar to WT). Such a result would argue against developmental effects of the mutation accounting for GC impairment is the absence of

NE.

Selective stimulation of signaling reduces levels of DH cAMP ex-vivo and retrieval deficits induced by either stress, GCs, agonists or xamoterol are blocked by expression of pertussis toxin in the DH, suggesting these agents also operate by decreasing cAMP signaling. To confirm and extend these results, DH cAMP will be measured in control and KO slices after application of GCs (or xamoterol) (5.2.1). The prediction would be that these agents would reduce levels of cAMP in controls, and cause little or no reduction in KOs. In addition, a cAMP activating analog will be infused into the DH of WT mice to determine if enhancing cAMP signaling prevents retrieval deficits induced by stress, GCs, agonists and xamoterol.

Initial evidence suggests that  and  receptors can influence retrieval, independent of GCs. The finding that KOs and ICI-treated controls display contextual fear impairments at 1-3, but not 5 or more days after conditioning suggest a specific role for signaling in intermediate-term, hippocampus-dependent retrieval. To further support this hypothesis, ICI will be administered to controls prior to testing cued fear 1 or

3 days after conditioning. A requirement for  in either the acquisition or long-term consolidation of fear memory would be unlikely if ICI-treated animals exhibit unimpaired auditory fear at these times. -ARs may also play a fleeting role in short-

111 term memory, which is supported by the findings that administration of ICI 30 min prior to or immediately after training impairs contextual fear expression at 1 h, while injection of ICI 30 min before training does not affect CF expression at 2 h (Figure 4.1). The later result suggests the deficit is in the maintenance rather than the acquisition of memory and is generally consistent with KOs displaying impaired CF at early time points. In addition, the observation that administration of ICI 30 min after training does not impair

CF at 1 h (in WT mice) argues against a deficit in short-term memory retrieval.

Furthermore, it can be predicted that administering ICI immediately after training will not impair CF expression at 2 h (in WT animals), which would provide additional evidence that impairments in STM are not due to deficits in acquisition.

Preliminary data from pharmacologic and genetic experiments suggest that  signaling may be required for retrieval. Additional studies will access the time course of impairment mediated by 3 signaling in WT mice. If deficits created by blocking 3 signaling are time-dependent in nature (as they are for other receptors)[1](Chapter 2), this would support a role for the receptor in intermediate-term retrieval rather than in acquisition or long-term consolidation. Future studies will examine 3KO animals (and control littermates) in order to assign a more definitive role for 3 in fear learning.

Moreover, testing cued fear in KOs will be important in determining whether freezing deficits caused by the lack of  signaling are hippocampus/retrieval-based or non- specifically effect levels of fear or freezing.

Despite the findings that both non-selective and subtype-selective  antagonists impair retrieval, several outcomes support the importance of distinguishing between

112 receptor subtypes. For example, behavioral, pharmacological and molecular evidence suggests that hippocampal - and -ARs fundamentally oppose one another in retrieval, and initial data suggests a similar dichotomy might exist between - and -ARs.

Finally, the failure of non-selective antagonists to significantly reduce the retrieval of traumatic memories in PTSD may be due to negating effects caused by combined

receptor antagonism.

Results in vivo and ex-vivo demonstrate that -coupled Gi/o signaling suppresses levels of cAMP signaling needed for retrieval of aversive, hippocampus-dependent memories (Chapter 2).  and GR agonist effects on retrieval mirror those of other agents that similarly reduce levels of cAMP, PKA and/or PKA-Epac signaling[153].

Importantly, Gi/o  and  subunits signal through proteins other than adenylate cyclase[254](Figure 5.1), and some these effectors may be involved in pathways necessary for episodic memory recall. For example, signaling by Gi/o  alters the activity of calcium/potassium ion channels and phospholipases[254]. Likewise, GTP binding of

Gi/o  (induced by  agonist or possibly GC activation of -ARs) results in the dissociation of  subunits, which exert effects on ACs, ion channels, PLC PI3K and other molecules[254]. Accumulating evidence demonstrates that many of these effectors can influence the recall of hippocampus-dependent memory. For example, calcium channel blockers (e.g. verapamil) dose-dependently rescue contextual retrieval deficits in mice lacking NE/E while calcium channel agonists (e.g. Bay K) inhibit retrieval in WT animals (communications with S.A. Thomas). Moreover, PLCmice display deficits in contextual fear 1d after training, consistent with either impaired consolidation or

113 retrieval[255]. In addition, CA1 infusion of either of two PI3K inhibitors 48 hr after training diminish contextual freezing[251]. These inhibitors also reduce the activity of

ERK, which also may be necessary for retrieval[251]. As a whole, such results support the probability that context-spatial retrieval is regulated by multiple signaling pathways that may act in concert, parallel or separately to effect the recall of fear-based information necessary for survival. Thus, redundancy in DH signaling processes influencing retrieval is of likely evolutionary benefit. The finding that multiple neurotransmitter and adrenergic receptors couple either to Gs, Gi/o and/or both support the importance of G- protein signaling in retrieval[254], with NE/E serving as a critical modulator[92]. Adding to the complexity are the many isoforms of G, and expressed in the hippocampus[256], which may ultimately combine in similar or different ways to regulate retrieval.

The observations that GCs rapidly impair contextual retrieval, that protein synthesis inhibitors do not prevent GC induced spatial deficits and that protein- conjugated cort impairs hippocampus-dependent memory all support non-genomic effects of GCs in influencing retrieval[131, 204](Chapter 2), some of which may be mediated by intracellular signaling through membrane-bound GR receptors[200, 201]. Interestingly, moderate to high levels of stress/GCs increases mGR-induced calcium channel, PLC and

GABAA activity and reduces NMDA receptor signaling[200, 201]. Notably, GABA agonists administered ICV or into the DH impair retrieval[237], mimicking the effects of stress and glucocorticoids, while NMDA antagonists produce no effect of hippocampus- dependent memory recall[257]. It is possible that similar isoforms of and  subunits

114 localized with mGRs and 2-ARs converge on common effectors, such as phospholipase

C to influence processes relating to fear memory. For example, PLC agonists dose- dependently enhance the consolidation of auditory fear memory (communications with

S.A. Thomas) and may impair the retrieval of fear memory in WT animals, analogous to how GCs display opposing effects on consolidation and retrieval that are concentration- dependent. Future experiments will test this hypothesis. There also exists the possibility that NE binds to mGRs, concurrently regulating GR activity along with the activity of 2-

ARs. Additional experiments are needed to test this possibility.

Alternatively, NE could directly interact with mGRs, changing their confirmation, increasing the affinity of mGRs for GCs (relative to MRs) and/or lowering the concentration of GCs needed to impair retrieval. Importantly, peripheral administration of the adrenal steroid synthesis inhibitor, metyrapone completely blocks the deleterious effects of stress and GCs on spatial memory retrieval[123]. Such results suggest

(indirectly) that GCs are not readily located in pools within the CNS, but are rapidly recruited during stress from the periphery to impair retrieval and promote consolidation of a newer, more salient stressful event. Of note, plasma cort levels peak 30 min following a stressful event, at a time associated with maximal retrieval impairment[123].

In contrast, cort is not significantly elevated 2 min or 4 hrs after stress, correlating to times when animals do not display cognitive deficits[123]. In contrast, levels of extracellular NE in the CNS reach a zenith 10-15 minutes after a stressor, faster than that of GCs[123, 258]. This raises the prospect that NE bound to either 2-ARs or mGRs helps directs the effects of GCs on retrieval. The finding that GCs released during low-

115 moderate restraint stress do not impair retrieval in mice lacking NE (but in WT) support the observation that NE priming of adrenergic and/or GR signaling might be critical in observing GCs effects, at least in some cases. The failure GCs to significantly reduce cAMP signaling both in vitro and ex vivo suggest that, under some physiological conditions, basal NE activity might be critical to observe GCs/stress effects. Mice lacking hippocampal GR will likely play an important role in distinguishing between some of

NE’s potential effects on GRs and 2-ARs. Fortunately, there are a growing number of methods to regionally, conditionally and/or temporarily reduce, eliminate or upregulate

GR and/or 2-AR expression and the combined use of pharmacological and genetic approaches will ultimately play an important role understanding the complex interactions between stress, GCs and 2 signaling.

Unfortunately, IHC, ISH and radioligand studies have been inconsistent concerning the extent of GR and -AR localization within specific subfields of the hippocampus and amygdala[96-98, 100, 101, 197]. Recent reports suggest that hippocampal and amygdalar GRs and 2-ARs are localized within the membrane, cytosol and nucleus, while 1-ARs are limited to the membrane and cytosol[96, 194].

Understanding such information might have function significance for both the consolidation and retrieval of hippocampus-dependent fear memories. For example, GR,

1-AR and or 2-ARs subcellular localization could change rapidly during times of stress, favoring processes that support consolidation while opposing those needed for retrieval.

For example, 2-ARs (or GRs) may be recruited to hippocampal membranes while 1-

ARs may be sequestered away from the membrane, both of which may negatively impact

116 retrieval. It would also help explain the supersensitivity of 2 agonists in the presence of

GCs. Alternatively, factors influenced by stress or GCs could increase nuclear levels of

2-ARs (or GRs) in the BLA which could bind GREs (or 2-AR response elements?), serving to regulate the expression of genes critical for learning (e.g. PKM-Arc, c-fos).

For example, administration of cort immediately after IA training increases hippocampal

Arc expression 45-60 min later, which can be blocked by co-administration of a non- selective -antagonist[258]. Interestingly, mice lacking NE show impaired Arc in CA1 the 30 min following retrieval[259], suggesting that GC-adrenergic interactions might be critical for the expression of IEGs at particular times during various stages of fear learning. Similarly, the possibility of nuclear 2 localization raises the possibility that 2 and GR might form heterodimers in the cytosol or nucleus that regulate consolidation through the recruitment or repression of various transcription factors.

117

Figure 5.1 Effectors of 2-coupled, Gi/o Signaling (Adapted from Wettschureck et al.,

Physiol Rev, 2005. 1159-204).

5.2 Future Directions This section describes experiments and alternative approaches that can be used to further explore the interactions between GCs, GRs, 2-ARs and cAMP signaling and between -AR subtypes.

5.2.1 Ex-Vivo Application of  and GR Agents in Control and AR KO Mice to Determine Effects on DH cAMP Signaling

Chapter 2 assesses contributions of individual receptors to DH cAMP signaling using ex-vivo slice preparation. Future studies will use this technique to explore how stress and GCs affect levels of DH cAMP in control and KO animals. Because selective stimulation of  decreases levels of cAMP, it is likely that stress and GCs will

118 reduce levels of cAMP in WT animals and have less effect in mice lacking -ARs. It can also be predicted that co-administration of a antagonist will prevent the decline in cAMP signaling mediated by GCs, analogous to initial results demonstrating that ICI blocks the reduction in cAMP signaling caused by stimulation of [unpublished].

Moreover, if  receptors are required for retrieval and mediate their effects by enhancing cAMP signaling, administration of a either a or antagonist should reduce DH cAMP levels in WT mice and have less effect in mice lacking - or -ARs. Furthermore, pharmacologic and genetic manipulation of multiple  receptors will access combinatory effects on DH cAMP signaling.

Advantages of this method are that cAMP manipulation occurs in a highly- controlled environment, not impacted by the activity of the animal or cAMP signaling from adjacent brain regions. Disadvantages of this technique are the artificial environment in which the tissue is maintained and that acute cAMP levels can be affected by decapitation or damaged tissue that accompanies slice preparation. Such effects are minimized by perfusing DH slices (of unanaesthetized animals) for 1 h before pharmacologic manipulation. Ex-vivo results can be confirmed by in vivo microdialysis, a technique that possesses some advantages.

5.2.2 In-Vivo Microdialysis to Measure Extracellular cAMP in Naïve and Conditioned Animals Following DH Infusion of AR & GR Compounds

Microdialysis has been used to measure extracellular cAMP in the CNS after administration of stress and agents affecting both  and GR signaling[260, 261].

Although most cAMP resides intracellularly, a measurable fraction is released

119 extracellularly following an aversive event or after systemic/central manipulation of adrenergic or GR signaling[260, 261]. Unlike ex-vivo experiments, levels of cAMP can be measured in real time in both naïve and fear conditioned animals (e.g. measuring cAMP before, during and after retrieval of fear memory). Perhaps the most compelling reason to perform microdialysis is that or GC effects on cAMP levels might differ using in-vivo and ex-vivo analysis[3]. For example, corticosterone produces a 40% reduction in extracellular cortical cAMP measured in vivo, but produces much less of an effect in cortical slice preparations[3]. Hence, achieving consistent results across preparations would be the gold-standard in cAMP analysis.

Several outcomes can be predicted using microdialysis. One prediction is that administration of a GC,  agonist, or  antagonist shortly before testing may impair levels of DH cAMP signaling that are required for retrieval. Similar pharmacologic manipulation may also reduce cAMP levels in unconditioned animals. In addition, DH cAMP levels might rise following retrieval in untreated animals, contributing to the reconsolidation or extinction of fear memory. Moreover, it can be hypothesized that co- administration of a  antagonist shortly before testing retrieval will prevent the decline in DH-cAMP mediated by concurrent GC or  agonist administration. Conversely, co- application of a GR antagonist should prevent a  agonist-mediated decrease in cAMP levels.

There are caveats to using in-vivo methods as well. Efforts need to be made to discriminate between alterations in DH-cAMP signaling that occur from behavioral, pharmacologic and non-specific effects (e.g. changes in the environment prior to

120 manipulation that impact levels of anxiety or activity). Also, implantation of probe causes necrosis and long-lasting molecular changes induced by apoptotic or inflammatory processes may affect cAMP levels, producing outcomes that differ from ex-vivo experiments.

5.2.3 Use of GRfl/fl X CaMKII-Cre+ Mice to Distinguish Between Direct Interaction or Coincident Signaling Between GR & -AR

The CRE-lox P system provides a method for excising genes in specific cell types and/or brain regions. For example, floxed GR mice lacking GR expression in forebrain excitatory neurons have been developed (GRfl/fl X CaMKII-Cre+)[262, 263]. Moreover, lines that selectively express Cre in specific subfields within the hippocampus (e.g. CA1,

CA3, DG) are available (CaMKII-Cre+, KA1-Cre+, POMC-Cre+)[263-265]. Using

GRfl/fl X CaMKII-Cre+ mice may assist in distinguishing between coincident signaling and direct interaction models to account for how GR & signaling intersect to impair retrieval. First, GRfl/fl X CaMKII-Cre+ mice will be tested to determine if they display retrieval deficits. It is not predicted that animals would display such impairments, because infusion of a GR antagonist into the hippocampus does not reduce retrieval in

WT mice. Most previous experiments have utilized corticosterone (a non-selective GR agonist) to assess GC involvement in retrieval, so cort will be administered to GRfl/fl X

CaMKII-Cre+ mutants to examine whether cort-induced retrieval deficits require GR.

Because a selective GR antagonist blocks cort-mediated retrieval impairment in WT animals, it is likely that GR is necessary for cort’s effects on retrieval.

121

If GR is required for cort-mediated deficits, hippocampus subfield-specific

GRKOs will be employed to determine which hippocampal regions mediate GR effects.

In addition, hippocampal homogenates from floxed and WT mice will be subjected to co- immunoprecipitation (co-IP) to determine if GR and  might interact (directly or through an intermediary) at the molecular level.

If GR is necessary for cort effects on retrieval, it will also be determined if effects

fl/fl of  agonists also depend on GR. If a agonist does not impair retrieval in GR X

CaMKII-Cre+ mice, this favors a coincident signaling model because concurrent GR signaling would be necessary for  effects. Alternatively, if a agonist impairs retrieval in mice lacking hippocampal GR, this suggests that there might be a direct interaction between  and GR, in order to explain why a GR antagonist blocks the effects of a 2 agonist in WT mice. To test this idea further, the ability of a GR antagonist to block the

fl/fl + effects of a 2 agonist would be examined in GR X CaMKII-Cre mice.

Conversely, GR may not be relevant for cort-induced impairment of retrieval.

Several possibilities could account for such a result. For example, GR may not be eliminated in all hippocampus subfields or cell types that are important for retrieval. For example, GR excision could be incomplete (performing either ISH or IHC on GR mutants and comparing GR expression to WT mice would serve as an important control).

Another possibility is that GR located in glial or inhibitory neurons mediates retrieval deficits. In addition, cort could activate proteins other than GR (e.g. MR) that play a role in retrieval. Alternatively, cort-mediated retrieval deficits might depend on direct interaction between cort and -ARs. One could test this possibility by performing a

122

3 saturated binding assay in which H-cort is administered to -transfected HEK-293,

CHO or SF9 cells, which possess little or no GR expression[266-268]. If results suggest that cort interacts with the  receptor, a competitive binding assay will be used to determine if cort acts specifically at the NE/E binding site.

5.2.4 Use of Floxed R Mice to Examine Hippocampal Subfield-specific Contributions of  Signaling in Retrieval

fl/f Floxed mice for hippocampal  are being developed (β1 ) (S.A. Thomas).

Efforts are planned to create similar floxed  and mice. Such mutants can be used to confirm pharmacologic and conventional mutant data indicating that hippocampal signaling by all 3 Rs may be necessary for retrieval, and more importantly, in what regions of the hippocampus individual Rs might be required. effects could arise from signaling in excitatory or inhibitory neurons or glia, and that all three possibilities will be tested by crossing floxed mice with cell-type specific Cre driver lines such as

CaMKII-Cre+, GAD67-Cre+ and GFAP-Cre+[263, 269, 270]. It will be necessary to determine the specificity and extent of regional and subfield  receptor ablation in floxed

 mice crossed with various Cre-expressing mice using ISH or IHC. Conventional -KOs will provide important controls for the specificity and extent of these techniques.

Evidence suggests that CA1 is critical in contextual and spatial memory retrieval[271, 272]. Interestingly, NE/E-deficient mice exhibit CA1-selective deficits in

Arc and Fos expression following retrieval and CA1 Fos expression is similarly impaired in KOs[259, 273]. It can be predicted that CA1 might also be critical for and 

123 effects, and the use of subfield-specific KOs will assist in determining the regions where individual ARs may converge to influence retrieval.

5.2.5 Intracellular Recording from CA1 Pyramidal Neurons to Examine if -cAMP & GR Signaling Mediate Voltage-gated Calcium Channel Effects on the Slow- afterhypolarization (sAHP)

It is well established that synaptic plasticity in the hippocampus, including CA1 pyramidal neurons, is critical for contextual and spatial memory consolidation[274].

Interestingly, enhanced cAMP-PKA signaling is believed to contribute to the reduction in the sAHP that occurs following conditioning in aversive, hippocampus-dependent tasks[275, 276], and may lead to a similar decline during retrieval. Notably, naive NE/E- deficient mice do not express deficits in either basal synaptic CA1 physiology or in the early or late phases of LTP[1], consistent with evidence indicating that NE is not required for consolidation of hippocampus-dependent memory[1]. On the other hand, -ARs can alter cAMP-PKA signaling in the DH (Chapter 2), which can influence the activity of voltage-gated, Ca++ channels (VGCCs) that are known to mediate the sAHP[277].

Blockade of VGCCs with calcium channel blockers reduce the slow phase of the AHP, decreasing the time between action potentials (APs), decreasing accommodation and promoting LTP (by increasing the rate of neuronal firing), similar to the effects of a NE,

1 agonists and cAMP-activating analogs on the sAHP[278-280]. Interestingly, all these agents facilitate the retrieval of hippocampus-dependent memory[92, 153][unpublished].

Conversely, cAMP antagonists, GCs, stress and Ca++ channel agonists enhance the activity of VGCCs, increasing the sAHP[280-283]. Notably, our lab has shown that all of these treatments impair the retrieval of fear memory [153](Chapter 2)[unpublished].

124

Hence, intracellular recording will be utilized to explore the hypothesis that 2 agonists mediate their effects by enhancing Ca++ signaling and increasing the sAHP, which may prevent the neuronal activation in CA1 required for retrieval. In addition, future experiments will test if 2 antagonists prevent GC and stress mediated enhancement of the sAHP, similar to GR antagonists[282]. A positive outcome would support the hypothesis that 2-AR and GR signaling mediate their effects through similar mechanisms, perhaps through VGCC modulation of the sAHP. If so, a 2 or GR antagonist should prevent enhancement of the sAHP mediated by a VGCC agonist.

125

References

1. Abel, T. and K.M. Lattal, Molecular mechanisms of memory acquisition, consolidation and retrieval. Curr Opin Neurobiol, 2001. 11(2): p. 180-7.

2. Pittenger, C., et al., Impaired bidirectional synaptic plasticity and procedural memory formation in striatum-specific cAMP response element-binding protein- deficient mice. J Neurosci, 2006. 26(10): p. 2808-13.

3. Abel, T. and P.V. Nguyen, Regulation of hippocampus-dependent memory by cyclic AMP-dependent protein kinase. Prog Brain Res, 2008. 169: p. 97-115.

4. Squire, L.R. and P.J. Bayley, The neuroscience of remote memory. Curr Opin Neurobiol, 2007. 17(2): p. 185-96.

5. Frankland, P.W. and B. Bontempi, The organization of recent and remote memories. Nat Rev Neurosci, 2005. 6(2): p. 119-30.

6. Kapur, N. and D.J. Brooks, Temporally-specific retrograde amnesia in two cases of discrete bilateral hippocampal pathology. Hippocampus, 1999. 9(3): p. 247-54.

7. Manns, J.R., R.O. Hopkins, and L.R. Squire, Semantic memory and the human hippocampus. Neuron, 2003. 38(1): p. 127-33.

8. Cahill, L. and J.L. McGaugh, A novel demonstration of enhanced memory associated with emotional arousal. Conscious Cogn, 1995. 4(4): p. 410-21.

9. Kensinger, E.A., Remembering emotional experiences: the contribution of valence and arousal. Rev Neurosci, 2004. 15(4): p. 241-51.

10. McGaugh, J.L., Memory--a century of consolidation. Science, 2000. 287(5451): p. 248-51.

11. Roozendaal, B., et al., Glucocorticoids interact with emotion-induced noradrenergic activation in influencing different memory functions. Neuroscience, 2006. 138(3): p. 901-10.

12. de Quervain, D.J., Glucocorticoid-induced reduction of traumatic memories: implications for the treatment of PTSD. Prog Brain Res, 2008. 167: p. 239-47.

13. van Stegeren, A.H., The role of the noradrenergic system in emotional memory. Acta Psychol (Amst), 2008. 127(3): p. 532-41.

126

14. Wingenfeld, K. and O.T. Wolf, HPA Axis Alterations in Mental Disorders: Impact on Memory and its Relevance for Therapeutic Interventions. CNS Neurosci Ther.

15. Wolkowitz, O.M., et al., Glucocorticoids. Mood, memory, and mechanisms. Ann N Y Acad Sci, 2009. 1179: p. 19-40.

16. Lind, S.E., Memory and the self in autism: A review and theoretical framework. Autism. 14(5): p. 430-56.

17. Meulenbroek, O., et al., Autobiographical memory retrieval in patients with Alzheimer's disease. Neuroimage. 53(1): p. 331-40.

18. Halliday, G., K. Cullen, and A. Harding, Neuropathological correlates of memory dysfunction in the Wernicke-Korsakoff syndrome. Alcohol Alcohol Suppl, 1994. 2: p. 245-51.

19. Michaud, K., H. Forget, and H. Cohen, Chronic glucocorticoid hypersecretion in Cushing's syndrome exacerbates cognitive aging. Brain Cogn, 2009. 71(1): p. 1-8.

20. Mair, R.G. and W.J. McEntee, Korsakoff's psychosis: noradrenergic systems and cognitive impairment. Behav Brain Res, 1983. 9(1): p. 1-32.

21. Weinshenker, D., Functional consequences of locus coeruleus degeneration in Alzheimer's disease. Curr Alzheimer Res, 2008. 5(3): p. 342-5.

22. Kim, J.J. and M.S. Fanselow, Modality-specific retrograde amnesia of fear. Science, 1992. 256(5057): p. 675-7.

23. Phillips, R.G. and J.E. LeDoux, Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci, 1992. 106(2): p. 274-85.

24. Maren, S., Neurobiology of Pavlovian fear conditioning. Annu Rev Neurosci, 2001. 24: p. 897-931.

25. LeDoux, J.E., et al., The lateral amygdaloid nucleus: sensory interface of the amygdala in fear conditioning. J Neurosci, 1990. 10(4): p. 1062-9.

26. Maren, S., G. Aharonov, and M.S. Fanselow, Retrograde abolition of conditional fear after excitotoxic lesions in the basolateral amygdala of rats: absence of a temporal gradient. Behav Neurosci, 1996. 110(4): p. 718-26.

27. Muller, J., et al., Functional inactivation of the lateral and basal nuclei of the amygdala by muscimol infusion prevents fear conditioning to an explicit

127

conditioned stimulus and to contextual stimuli. Behav Neurosci, 1997. 111(4): p. 683-91.

28. Goosens, K.A. and S. Maren, Contextual and auditory fear conditioning are mediated by the lateral, basal, and central amygdaloid nuclei in rats. Learn Mem, 2001. 8(3): p. 148-55.

29. Nader, K., et al., Damage to the lateral and central, but not other, amygdaloid nuclei prevents the acquisition of auditory fear conditioning. Learn Mem, 2001. 8(3): p. 156-63.

30. Wilensky, A.E., et al., Rethinking the fear circuit: the central nucleus of the amygdala is required for the acquisition, consolidation, and expression of Pavlovian fear conditioning. J Neurosci, 2006. 26(48): p. 12387-96.

31. Kim, J.J. and M.W. Jung, Neural circuits and mechanisms involved in Pavlovian fear conditioning: a critical review. Neurosci Biobehav Rev, 2006. 30(2): p. 188- 202.

32. LeDoux, J.E., et al., Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. J Neurosci, 1988. 8(7): p. 2517-29.

33. Cousens, G. and T. Otto, Both pre- and posttraining excitotoxic lesions of the basolateral amygdala abolish the expression of olfactory and contextual fear conditioning. Behav Neurosci, 1998. 112(5): p. 1092-103.

34. Helmstetter, F.J. and P.S. Bellgowan, Effects of muscimol applied to the basolateral amygdala on acquisition and expression of contextual fear conditioning in rats. Behav Neurosci, 1994. 108(5): p. 1005-9.

35. Kim, J.J., R.A. Rison, and M.S. Fanselow, Effects of amygdala, hippocampus, and periaqueductal gray lesions on short- and long-term contextual fear. Behav Neurosci, 1993. 107(6): p. 1093-8.

36. Zimmerman, J.M., et al., The central nucleus of the amygdala is essential for acquiring and expressing conditional fear after overtraining. Learn Mem, 2007. 14(9): p. 634-44.

37. Vianna, D.M., et al., Lesion of the ventral periaqueductal gray reduces conditioned fear but does not change freezing induced by stimulation of the dorsal periaqueductal gray. Learn Mem, 2001. 8(3): p. 164-9.

38. Anagnostaras, S.G., S. Maren, and M.S. Fanselow, Temporally graded retrograde amnesia of contextual fear after hippocampal damage in rats: within-subjects examination. J Neurosci, 1999. 19(3): p. 1106-14. 128

39. Maren, S., G. Aharonov, and M.S. Fanselow, Neurotoxic lesions of the dorsal hippocampus and Pavlovian fear conditioning in rats. Behav Brain Res, 1997. 88(2): p. 261-74.

40. Bast, T., W.N. Zhang, and J. Feldon, The ventral hippocampus and fear conditioning in rats. Different anterograde amnesias of fear after tetrodotoxin inactivation and infusion of the GABA(A) agonist muscimol. Exp Brain Res, 2001. 139(1): p. 39-52.

41. Maren, S., Neurotoxic or electrolytic lesions of the ventral subiculum produce deficits in the acquisition and expression of Pavlovian fear conditioning in rats. Behav Neurosci, 1999. 113(2): p. 283-90.

42. Maren, S. and M.S. Fanselow, Synaptic plasticity in the basolateral amygdala induced by hippocampal formation stimulation in vivo. J Neurosci, 1995. 15(11): p. 7548-64.

43. Richmond, M.A., et al., Dissociating context and space within the hippocampus: effects of complete, dorsal, and ventral excitotoxic hippocampal lesions on conditioned freezing and spatial learning. Behav Neurosci, 1999. 113(6): p. 1189- 203.

44. Ambrogi Lorenzini, C.G., et al., Neural topography and chronology of memory consolidation: a review of functional inactivation findings. Neurobiol Learn Mem, 1999. 71(1): p. 1-18.

45. Amorapanth, P., J.E. LeDoux, and K. Nader, Different lateral amygdala outputs mediate reactions and actions elicited by a fear-arousing stimulus. Nat Neurosci, 2000. 3(1): p. 74-9.

46. Izquierdo, I., et al., Sequential role of hippocampus and amygdala, entorhinal cortex and parietal cortex in formation and retrieval of memory for inhibitory avoidance in rats. Eur J Neurosci, 1997. 9(4): p. 786-93.

47. Ambrogi Lorenzini, C.G., et al., Role of ventral hippocampus in acquisition, consolidation and retrieval of rat's passive avoidance response memory trace. Brain Res, 1997. 768(1-2): p. 242-8.

48. Lorenzini, C.A., et al., Role of dorsal hippocampus in acquisition, consolidation and retrieval of rat's passive avoidance response: a tetrodotoxin functional inactivation study. Brain Res, 1996. 730(1-2): p. 32-9.

49. Moser, E., M.B. Moser, and P. Andersen, Spatial learning impairment parallels the magnitude of dorsal hippocampal lesions, but is hardly present following ventral lesions. J Neurosci, 1993. 13(9): p. 3916-25.

129

50. Bucherelli, C., G. Tassoni, and J. Bures, Time-dependent disruption of passive avoidance acquisition by post-training intra-amygdala injection of tetrodotoxin in rats. Neurosci Lett, 1992. 140(2): p. 231-4.

51. Parent, M.B. and J.L. McGaugh, Posttraining infusion of lidocaine into the amygdala basolateral complex impairs retention of inhibitory avoidance training. Brain Res, 1994. 661(1-2): p. 97-103.

52. Lorenzini, C.A., et al., Time-dependent deficits of rat's memory consolidation induced by tetrodotoxin injections into the caudate-putamen, nucleus accumbens, and globus pallidus. Neurobiol Learn Mem, 1995. 63(1): p. 87-93.

53. Rothman, A.H. and S.D. Glick, Differential effects of unilateral and bilateral caudate lesions on side preference and passive avoidance behavior in rats. Brain Res, 1976. 118(3): p. 361-9.

54. Winocur, G., Functional dissociation within the caudate nucleus of rats. J Comp Physiol Psychol, 1974. 86(3): p. 432-9.

55. Taghzouti, K., et al., Behavioral study after local injection of 6-hydroxydopamine into the nucleus accumbens in the rat. Brain Res, 1985. 344(1): p. 9-20.

56. Flicker, C., et al., Behavioral and neurochemical effects following neurotoxic lesions of a major cholinergic input to the cerebral cortex in the rat. Pharmacol Biochem Behav, 1983. 18(6): p. 973-81.

57. D'Hooge, R. and P.P. De Deyn, Applications of the Morris water maze in the study of learning and memory. Brain Res Brain Res Rev, 2001. 36(1): p. 60-90.

58. Decker, M.W., P. Curzon, and J.D. Brioni, Influence of separate and combined septal and amygdala lesions on memory, acoustic startle, anxiety, and locomotor activity in rats. Neurobiol Learn Mem, 1995. 64(2): p. 156-68.

59. Roozendaal, B. and J.L. McGaugh, Basolateral amygdala lesions block the memory-enhancing effect of glucocorticoid administration in the dorsal hippocampus of rats. Eur J Neurosci, 1997. 9(1): p. 76-83.

60. Spanis, C.W., et al., Excitotoxic basolateral amygdala lesions potentiate the memory impairment effect of muscimol injected into the medial septal area. Brain Res, 1999. 816(2): p. 329-36.

61. Gallagher, M. and P.C. Holland, Preserved configural learning and spatial learning impairment in rats with hippocampal damage. Hippocampus, 1992. 2(1): p. 81-8.

130

62. Riedel, G., et al., Reversible neural inactivation reveals hippocampal participation in several memory processes. Nat Neurosci, 1999. 2(10): p. 898- 905.

63. Moser, M.B., et al., Spatial learning with a minislab in the dorsal hippocampus. Proc Natl Acad Sci U S A, 1995. 92(21): p. 9697-701.

64. Gold, P.E., Coordination of multiple memory systems. Neurobiol Learn Mem, 2004. 82(3): p. 230-42.

65. Gold, P.E. and R.B. Van Buskirk, Facilitation of time-dependent memory processes with posttrial epinephrine injections. Behav Biol, 1975. 13(2): p. 145- 53.

66. Cahill, L. and M.T. Alkire, Epinephrine enhancement of human memory consolidation: interaction with arousal at encoding. Neurobiol Learn Mem, 2003. 79(2): p. 194-8.

67. Gold, P.E., J. Vogt, and J.L. Hall, Glucose effects on memory: behavioral and pharmacological characteristics. Behav Neural Biol, 1986. 46(2): p. 145-55.

68. Talley, C.E., et al., Epinephrine fails to enhance performance of food-deprived rats on a delayed spontaneous alternation task. Neurobiol Learn Mem, 2000. 73(1): p. 79-86.

69. Axelrod, J., H. Weil-Malherbe, and R. Tomchick, The physiological disposition of H3-epinephrine and its metabolite metanephrine. J Pharmacol Exp Ther, 1959. 127: p. 251-6.

70. Hassert, D.L., T. Miyashita, and C.L. Williams, The effects of peripheral vagal nerve stimulation at a memory-modulating intensity on norepinephrine output in the basolateral amygdala. Behav Neurosci, 2004. 118(1): p. 79-88.

71. McGaugh, J.L., L. Cahill, and B. Roozendaal, Involvement of the amygdala in memory storage: interaction with other brain systems. Proc Natl Acad Sci U S A, 1996. 93(24): p. 13508-14.

72. Introini-Collison, I., et al., Memory-enhancing effects of post-training dipivefrin and epinephrine: involvement of peripheral and central adrenergic receptors. Brain Res, 1992. 572(1-2): p. 81-6.

73. McGaugh, J.L., The amygdala modulates the consolidation of memories of emotionally arousing experiences. Annu Rev Neurosci, 2004. 27: p. 1-28.

131

74. Liang, K.C., L.L. Chen, and T.E. Huang, The role of amygdala norepinephrine in memory formation: involvement in the memory enhancing effect of peripheral epinephrine. Chin J Physiol, 1995. 38(2): p. 81-91.

75. Vermetten, E. and J.D. Bremner, Circuits and systems in stress. I. Preclinical studies. Depress Anxiety, 2002. 15(3): p. 126-47.

76. Aston-Jones, G., J. Rajkowski, and J. Cohen, Locus coeruleus and regulation of behavioral flexibility and attention. Prog Brain Res, 2000. 126: p. 165-82.

77. McGaugh, J.L. and B. Roozendaal, Role of adrenal stress hormones in forming lasting memories in the brain. Curr Opin Neurobiol, 2002. 12(2): p. 205-10.

78. Liang, K.C., J.L. McGaugh, and H.Y. Yao, Involvement of amygdala pathways in the influence of post-training intra-amygdala norepinephrine and peripheral epinephrine on memory storage. Brain Res, 1990. 508(2): p. 225-33.

79. Bevilaqua, L., et al., Drugs acting upon the cyclic adenosine monophosphate/protein kinase A signalling pathway modulate memory consolidation when given late after training into rat hippocampus but not amygdala. Behav Pharmacol, 1997. 8(4): p. 331-8.

80. Gallagher, M., et al., Memory formation: evidence for a specific neurochemical system in the amygdala. Science, 1977. 198(4315): p. 423-5.

81. Lee, E.H., et al., Hippocampal CRF, NE, and NMDA system interactions in memory processing in the rat. Synapse, 1993. 14(2): p. 144-53.

82. Allain, P. and N. Krari, Diethyldithiocarbamate, copper and neurological disorders. Life Sci, 1991. 48(3): p. 291-9.

83. Delmaestro, E. and L.D. Trombetta, The effects of disulfiram on the hippocampus and cerebellum of the rat brain: a study on oxidative stress. Toxicol Lett, 1995. 75(1-3): p. 235-43.

84. Grassi Zucconi, G., et al., Sleep impairment by diethyldithiocarbamate in rat. Protective effects of pre-conditioning and antioxidants. Brain Res, 2002. 939(1- 2): p. 87-94.

85. Abercrombie, E.D. and M.J. Zigmond, Partial injury to central noradrenergic neurons: reduction of tissue norepinephrine content is greater than reduction of extracellular norepinephrine measured by microdialysis. J Neurosci, 1989. 9(11): p. 4062-7.

132

86. Acheson, A.L., M.J. Zigmond, and E.M. Stricker, Compensatory increase in tyrosine hydroxylase activity in rat brain after intraventricular injections of 6- hydroxydopamine. Science, 1980. 207(4430): p. 537-40.

87. Chiodo, L.A., et al., Subtotal destruction of central noradrenergic projections increases the firing rate of locus coeruleus cells. Brain Res, 1983. 264(1): p. 123- 6.

88. Gage, F.H., A. Bjorklund, and U. Stenevi, Local regulation of compensatory noradrenergic hyperactivity in the partially denervated hippocampus. Nature, 1983. 303(5920): p. 819-21.

89. Hughes, Z.A. and S.C. Stanford, A partial noradrenergic lesion induced by DSP- 4 increases extracellular noradrenaline concentration in rat frontal cortex: a microdialysis study in vivo. Psychopharmacology (Berl), 1998. 136(3): p. 299- 303.

90. Thomas, S.A., et al., Restoration of norepinephrine and reversal of phenotypes in mice lacking dopamine beta-hydroxylase. J Neurochem, 1998. 70(6): p. 2468-76.

91. Thomas, S.A., A.M. Matsumoto, and R.D. Palmiter, Noradrenaline is essential for mouse fetal development. Nature, 1995. 374(6523): p. 643-6.

92. Murchison, C.F., et al., A distinct role for norepinephrine in memory retrieval. Cell, 2004. 117(1): p. 131-43.

93. Thomas, S.A. and R.D. Palmiter, Disruption of the dopamine beta-hydroxylase gene in mice suggests roles for norepinephrine in motor function, learning, and memory. Behav Neurosci, 1997. 111(3): p. 579-89.

94. Frick, K.M., E.T. Stillner, and J. Berger-Sweeney, Mice are not little rats: species differences in a one-day water maze task. Neuroreport, 2000. 11(16): p. 3461-5.

95. Pupo, A.S. and K.P. Minneman, Adrenergic pharmacology: focus on the central nervous system. CNS Spectr, 2001. 6(8): p. 656-62.

96. Guo, N.N. and B.M. Li, Cellular and subcellular distributions of beta1- and beta2-adrenoceptors in the CA1 and CA3 regions of the rat hippocampus. Neuroscience, 2007. 146(1): p. 298-305.

97. Joyce, J.N., et al., Distribution of beta-adrenergic receptor subtypes in human post-mortem brain: alterations in limbic regions of schizophrenics. Synapse, 1992. 10(3): p. 228-46.

133

98. Nicholas, A.P., V.A. Pieribone, and T. Hokfelt, Cellular localization of messenger RNA for beta-1 and beta-2 adrenergic receptors in rat brain: an in situ hybridization study. Neuroscience, 1993. 56(4): p. 1023-39.

99. Summers, R.J., et al., Expression of beta 3-adrenoceptor mRNA in rat brain. Br J Pharmacol, 1995. 116(6): p. 2547-8.

100. Wanaka, A., et al., Immunocytochemical localization of beta-adrenergic receptors in the rat brain. Brain Res, 1989. 485(1): p. 125-40.

101. Reul, J.M. and E.R. de Kloet, Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology, 1985. 117(6): p. 2505-11.

102. Oitzl, M.S. and E.R. de Kloet, Selective corticosteroid antagonists modulate specific aspects of spatial orientation learning. Behav Neurosci, 1992. 106(1): p. 62-71.

103. Roozendaal, B. and J.L. McGaugh, Glucocorticoid receptor agonist and antagonist administration into the basolateral but not central amygdala modulates memory storage. Neurobiol Learn Mem, 1997. 67(2): p. 176-9.

104. Roozendaal, B., et al., Basolateral amygdala noradrenergic influence enables enhancement of memory consolidation induced by hippocampal glucocorticoid receptor activation. Proc Natl Acad Sci U S A, 1999. 96(20): p. 11642-7.

105. Roozendaal, B., G. Portillo-Marquez, and J.L. McGaugh, Basolateral amygdala lesions block glucocorticoid-induced modulation of memory for spatial learning. Behav Neurosci, 1996. 110(5): p. 1074-83.

106. Ferguson, D. and R. Sapolsky, Overexpression of mineralocorticoid and transdominant glucocorticoid receptor blocks the impairing effects of glucocorticoids on memory. Hippocampus, 2008. 18(11): p. 1103-11.

107. Roozendaal, B., et al., The hippocampus mediates glucocorticoid-induced impairment of spatial memory retrieval: dependence on the basolateral amygdala. Proc Natl Acad Sci U S A, 2003. 100(3): p. 1328-33.

108. Khaksari, M., A. Rashidy-Pour, and A.A. Vafaei, Central mineralocorticoid receptors are indispensable for corticosterone-induced impairment of memory retrieval in rats. Neuroscience, 2007. 149(4): p. 729-38.

109. Buchanan, T.W. and W.R. Lovallo, Enhanced memory for emotional material following stress-level cortisol treatment in humans. Psychoneuroendocrinology, 2001. 26(3): p. 307-17.

134

110. Flood, J.F., et al., Memory facilitating and anti-amnesic effects of corticosteroids. Pharmacol Biochem Behav, 1978. 8(1): p. 81-7.

111. Lupien, S.J. and B.S. McEwen, The acute effects of corticosteroids on cognition: integration of animal and human model studies. Brain Res Brain Res Rev, 1997. 24(1): p. 1-27.

112. Roozendaal, B., 1999 Curt P. Richter award. Glucocorticoids and the regulation of memory consolidation. Psychoneuroendocrinology, 2000. 25(3): p. 213-38.

113. Conrad, C.D., et al., Influence of chronic corticosterone and glucocorticoid receptor antagonism in the amygdala on fear conditioning. Neurobiol Learn Mem, 2004. 81(3): p. 185-99.

114. Donley, M.P., J. Schulkin, and J.B. Rosen, Glucocorticoid receptor antagonism in the basolateral amygdala and ventral hippocampus interferes with long-term memory of contextual fear. Behav Brain Res, 2005. 164(2): p. 197-205.

115. Roozendaal, B., C.L. Williams, and J.L. McGaugh, Glucocorticoid receptor activation in the rat nucleus of the solitary tract facilitates memory consolidation: involvement of the basolateral amygdala. Eur J Neurosci, 1999. 11(4): p. 1317- 23.

116. Beato, M., P. Herrlich, and G. Schutz, Steroid hormone receptors: many actors in search of a plot. Cell, 1995. 83(6): p. 851-7.

117. Datson, N.A., et al., Identification of corticosteroid-responsive genes in rat hippocampus using serial analysis of gene expression. Eur J Neurosci, 2001. 14(4): p. 675-89.

118. Oitzl, M.S., et al., Point mutation in the mouse glucocorticoid receptor preventing DNA binding impairs spatial memory. Proc Natl Acad Sci U S A, 2001. 98(22): p. 12790-5.

119. Karst, H., et al., Mineralocorticoid receptors are indispensable for nongenomic modulation of hippocampal glutamate transmission by corticosterone. Proc Natl Acad Sci U S A, 2005. 102(52): p. 19204-7.

120. Orchinik, M., T.F. Murray, and F.L. Moore, A corticosteroid receptor in neuronal membranes. Science, 1991. 252(5014): p. 1848-51.

121. Tasker, J.G., S. Di, and R. Malcher-Lopes, Minireview: rapid glucocorticoid signaling via membrane-associated receptors. Endocrinology, 2006. 147(12): p. 5549-56.

135

122. Roozendaal, B., G.L. Quirarte, and J.L. McGaugh, Glucocorticoids interact with the basolateral amygdala beta-adrenoceptor--cAMP/cAMP/PKA system in influencing memory consolidation. Eur J Neurosci, 2002. 15(3): p. 553-60.

123. de Quervain, D.J., B. Roozendaal, and J.L. McGaugh, Stress and glucocorticoids impair retrieval of long-term spatial memory. Nature, 1998. 394(6695): p. 787- 90.

124. de Quervain, D.J., et al., Acute cortisone administration impairs retrieval of long- term declarative memory in humans. Nat Neurosci, 2000. 3(4): p. 313-4.

125. Het, S., G. Ramlow, and O.T. Wolf, A meta-analytic review of the effects of acute cortisol administration on human memory. Psychoneuroendocrinology, 2005. 30(8): p. 771-84.

126. Rashidy-Pour, A., et al., The effects of acute restraint stress and dexamethasone on retrieval of long-term memory in rats: an interaction with opiate system. Behav Brain Res, 2004. 154(1): p. 193-8.

127. Wolf, O.T., Stress and memory in humans: twelve years of progress? Brain Res, 2009. 1293: p. 142-54.

128. Roozendaal, B., A. Barsegyan, and S. Lee, Adrenal stress hormones, amygdala activation, and memory for emotionally arousing experiences. Prog Brain Res, 2008. 167: p. 79-97.

129. Wolf, O.T., HPA axis and memory. Best Pract Res Clin Endocrinol Metab, 2003. 17(2): p. 287-99.

130. Chauveau, F., et al., Rapid stress-induced corticosterone rise in the hippocampus reverses serial memory retrieval pattern. Hippocampus. 20(1): p. 196-207.

131. Sajadi, A.A., S.A. Samaei, and A. Rashidy-Pour, Intra-hippocampal microinjections of anisomycin did not block glucocorticoid-induced impairment of memory retrieval in rats: an evidence for non-genomic effects of glucocorticoids. Behav Brain Res, 2006. 173(1): p. 158-62.

132. Liu, L., et al., Adrenocortical suppression blocks the enhancement of memory storage produced by exposure to psychological stress in rats. Brain Res, 1999. 821(1): p. 134-40.

133. Pugh, C.R., et al., A selective role for corticosterone in contextual-fear conditioning. Behav Neurosci, 1997. 111(3): p. 503-11.

136

134. Roozendaal, B., O. Carmi, and J.L. McGaugh, Adrenocortical suppression blocks the memory-enhancing effects of amphetamine and epinephrine. Proc Natl Acad Sci U S A, 1996. 93(4): p. 1429-33.

135. Roozendaal, B., et al., Basolateral amygdala noradrenergic activity mediates corticosterone-induced enhancement of auditory fear conditioning. Neurobiol Learn Mem, 2006. 86(3): p. 249-55.

136. Quirarte, G.L., B. Roozendaal, and J.L. McGaugh, Glucocorticoid enhancement of memory storage involves noradrenergic activation in the basolateral amygdala. Proc Natl Acad Sci U S A, 1997. 94(25): p. 14048-53.

137. McEwen, B.S., Glucocorticoid-biogenic amine interactions in relation to mood and behavior. Biochem Pharmacol, 1987. 36(11): p. 1755-63.

138. McIntrye CK, R.B., McGaugh JL, Glucocorticoid treatment enhances training- induced norepinephrine release in the amygdala. Soc Neurosci Abstr 772.12, 2004: p. http://sfn.scholarone.com/itin2004/.

139. Grundemann, D., et al., Molecular identification of the corticosterone-sensitive extraneuronal catecholamine transporter. Nat Neurosci, 1998. 1(5): p. 349-51.

140. Roozendaal, B., et al., A systemically administered beta-adrenoceptor antagonist blocks corticosterone-induced impairment of contextual memory retrieval in rats. Neurobiol Learn Mem, 2004. 81(2): p. 150-4.

141. Roozendaal, B., et al., Glucocorticoid effects on memory retrieval require concurrent noradrenergic activity in the hippocampus and basolateral amygdala. J Neurosci, 2004. 24(37): p. 8161-9.

142. Barad, M., et al., Rolipram, a type IV-specific phosphodiesterase inhibitor, facilitates the establishment of long-lasting long-term potentiation and improves memory. Proc Natl Acad Sci U S A, 1998. 95(25): p. 15020-5.

143. Bernabeu, R., et al., Involvement of hippocampal cAMP/cAMP-dependent protein kinase signaling pathways in a late memory consolidation phase of aversively motivated learning in rats. Proc Natl Acad Sci U S A, 1997. 94(13): p. 7041-6.

144. Bourtchouladze, R., et al., Different training procedures recruit either one or two critical periods for contextual memory consolidation, each of which requires protein synthesis and PKA. Learn Mem, 1998. 5(4-5): p. 365-74.

145. Randt, C.T., et al., Brain cyclic AMP and memory in mice. Pharmacol Biochem Behav, 1982. 17(4): p. 677-80.

137

146. Schafe, G.E., et al., Memory consolidation for contextual and auditory fear conditioning is dependent on protein synthesis, PKA, and MAP kinase. Learn Mem, 1999. 6(2): p. 97-110.

147. Abel, T., et al., Genetic demonstration of a role for PKA in the late phase of LTP and in hippocampus-based long-term memory. Cell, 1997. 88(5): p. 615-26.

148. Ahi, J., J. Radulovic, and J. Spiess, The role of hippocampal signaling cascades in consolidation of fear memory. Behav Brain Res, 2004. 149(1): p. 17-31.

149. Wallenstein, G.V., D.R. Vago, and A.M. Walberer, Time-dependent involvement of PKA/PKC in contextual memory consolidation. Behav Brain Res, 2002. 133(2): p. 159-64.

150. Josselyn, S.A. and P.V. Nguyen, CREB, synapses and memory disorders: past progress and future challenges. Curr Drug Targets CNS Neurol Disord, 2005. 4(5): p. 481-97.

151. Wood, M.A., et al., A transcription factor-binding domain of the coactivator CBP is essential for long-term memory and the expression of specific target genes. Learn Mem, 2006. 13(5): p. 609-17.

152. Wood, M.A., et al., Transgenic mice expressing a truncated form of CREB- binding protein (CBP) exhibit deficits in hippocampal synaptic plasticity and memory storage. Learn Mem, 2005. 12(2): p. 111-9.

153. Ouyang, M., et al., Epac signaling is required for hippocampus-dependent memory retrieval. Proc Natl Acad Sci U S A, 2008. 105(33): p. 11993-7.

154. Barros, D.M., et al., Molecular signalling pathways in the cerebral cortex are required for retrieval of one-trial avoidance learning in rats. Behav Brain Res, 2000. 114(1-2): p. 183-92.

155. Boynton, L., et al., Preliminary findings concerning the use of prazosin for the treatment of posttraumatic nightmares in a refugee population. J Psychiatr Pract, 2009. 15(6): p. 454-9.

156. Raskind, M.A., et al., Reduction of nightmares and other PTSD symptoms in combat veterans by prazosin: a placebo-controlled study. Am J Psychiatry, 2003. 160(2): p. 371-3.

157. van Liempt, S., et al., Pharmacotherapeutic treatment of nightmares and insomnia in posttraumatic stress disorder: an overview of the literature. Ann N Y Acad Sci, 2006. 1071: p. 502-7.

138

158. Boehnlein, J.K. and J.D. Kinzie, Pharmacologic reduction of CNS noradrenergic activity in PTSD: the case for clonidine and prazosin. J Psychiatr Pract, 2007. 13(2): p. 72-8.

159. Pitman, R.K. and D.L. Delahanty, Conceptually driven pharmacologic approaches to acute trauma. CNS Spectr, 2005. 10(2): p. 99-106.

160. Pitman, R.K., et al., Pilot study of secondary prevention of posttraumatic stress disorder with propranolol. Biol Psychiatry, 2002. 51(2): p. 189-92.

161. Vaiva, G., et al., Immediate treatment with propranolol decreases posttraumatic stress disorder two months after trauma. Biol Psychiatry, 2003. 54(9): p. 947-9.

162. McGhee, L.L., et al., The effect of propranolol on posttraumatic stress disorder in burned service members. J Burn Care Res, 2009. 30(1): p. 92-7.

163. Sharp, S., et al., Propranolol does not reduce risk for acute stress disorder in pediatric burn trauma. J Trauma. 68(1): p. 193-7.

164. Stein, M.B., et al., Pharmacotherapy to prevent PTSD: Results from a randomized controlled proof-of-concept trial in physically injured patients. J Trauma Stress, 2007. 20(6): p. 923-32.

165. Debiec, J. and J.E. Ledoux, Disruption of reconsolidation but not consolidation of auditory fear conditioning by noradrenergic blockade in the amygdala. Neuroscience, 2004. 129(2): p. 267-72.

166. Alexander, J.K., et al., Beta-adrenergic modulation of cognitive flexibility during stress. J Cogn Neurosci, 2007. 19(3): p. 468-78.

167. Brewer, C., Beneficial effect of beta-adrenergic blockade on "exam. nerves". Lancet, 1972. 2(7774): p. 435.

168. Faigel, H.C., The effect of beta blockade on stress-induced cognitive dysfunction in adolescents. Clin Pediatr (Phila), 1991. 30(7): p. 441-5.

169. Diergaarde, L., A.N. Schoffelmeer, and T.J. De Vries, Pharmacological manipulation of memory reconsolidation: towards a novel treatment of pathogenic memories. Eur J Pharmacol, 2008. 585(2-3): p. 453-7.

170. Debiec, J., D.E. Bush, and J.E. Ledoux, Noradrenergic enhancement of reconsolidation in the amygdala impairs extinction of conditioned fear in rats-a possible mechanism for the persistence of traumatic memories in PTSD. Depress Anxiety. 28(3): p. 186-93.

139

171. Przybyslawski, J., P. Roullet, and S.J. Sara, Attenuation of emotional and nonemotional memories after their reactivation: role of beta adrenergic receptors. J Neurosci, 1999. 19(15): p. 6623-8.

172. Kroes, M.C., B.A. Strange, and R.J. Dolan, Beta-adrenergic blockade during memory retrieval in humans evokes a sustained reduction of declarative emotional memory enhancement. J Neurosci. 30(11): p. 3959-63.

173. Muravieva, E.V. and C.M. Alberini, Limited efficacy of propranolol on the reconsolidation of fear memories. Learn Mem. 17(6): p. 306-13.

174. Soeter, M. and M. Kindt, Dissociating response systems: erasing fear from memory. Neurobiol Learn Mem. 94(1): p. 30-41.

175. Schelling, G., et al., Stress doses of hydrocortisone, traumatic memories, and symptoms of posttraumatic stress disorder in patients after cardiac surgery: a randomized study. Biol Psychiatry, 2004. 55(6): p. 627-33.

176. Schelling, G., et al., Efficacy of hydrocortisone in preventing posttraumatic stress disorder following critical illness and major surgery. Ann N Y Acad Sci, 2006. 1071: p. 46-53.

177. Weis, F., et al., Stress doses of hydrocortisone reduce chronic stress symptoms and improve health-related quality of life in high-risk patients after cardiac surgery: a randomized study. J Thorac Cardiovasc Surg, 2006. 131(2): p. 277-82.

178. Ehring, T., et al., Do acute psychological and psychobiological responses to trauma predict subsequent symptom severities of PTSD and depression? Psychiatry Res, 2008. 161(1): p. 67-75.

179. Fletcher, S., M. Creamer, and D. Forbes, Preventing post traumatic stress disorder: are drugs the answer? Aust N Z J Psychiatry. 44(12): p. 1064-71.

180. Simon, A. and J. Gorman, Psychopharmacological possibilities in the acute disaster setting. Psychiatr Clin North Am, 2004. 27(3): p. 425-58.

181. Yehuda, R., Risk and resilience in posttraumatic stress disorder. J Clin Psychiatry, 2004. 65 Suppl 1: p. 29-36.

182. McEwen, B.S., Stress and hippocampal plasticity. Annu Rev Neurosci, 1999. 22: p. 105-22.

183. McKittrick, C.R., et al., Chronic social stress reduces dendritic arbors in CA3 of hippocampus and decreases binding to serotonin transporter sites. Synapse, 2000. 36(2): p. 85-94.

140

184. Radley, J.J., et al., Chronic behavioral stress induces apical dendritic reorganization in pyramidal neurons of the medial prefrontal cortex. Neuroscience, 2004. 125(1): p. 1-6.

185. Vyas, A., et al., Chronic stress induces contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. J Neurosci, 2002. 22(15): p. 6810-8.

186. Pariante, C.M., Risk factors for development of depression and psychosis. Glucocorticoid receptors and pituitary implications for treatment with antidepressant and glucocorticoids. Ann N Y Acad Sci, 2009. 1179: p. 144-52.

187. Aerni, A., et al., Low-dose cortisol for symptoms of posttraumatic stress disorder. Am J Psychiatry, 2004. 161(8): p. 1488-90.

188. Abrari, K., et al., Administration of corticosterone after memory reactivation disrupts subsequent retrieval of a contextual conditioned fear memory: dependence upon training intensity. Neurobiol Learn Mem, 2008. 89(2): p. 178- 84.

189. Cai, W.H., et al., Postreactivation glucocorticoids impair recall of established fear memory. J Neurosci, 2006. 26(37): p. 9560-6.

190. Suris, A., et al., Effects of exogenous glucocorticoid on combat-related PTSD symptoms. Ann Clin Psychiatry. 22(4): p. 274-9.

191. Jin, X.C., et al., Glucocorticoid receptors in the basolateral nucleus of amygdala are required for postreactivation reconsolidation of auditory fear memory. Eur J Neurosci, 2007. 25(12): p. 3702-12.

192. Tronel, S. and C.M. Alberini, Persistent disruption of a traumatic memory by postretrieval inactivation of glucocorticoid receptors in the amygdala. Biol Psychiatry, 2007. 62(1): p. 33-9.

193. Yang, Y.L., P.K. Chao, and K.T. Lu, Systemic and intra-amygdala administration of glucocorticoid agonist and antagonist modulate extinction of conditioned fear. Neuropsychopharmacology, 2006. 31(5): p. 912-24.

194. Qu, L.L., N.N. Guo, and B.M. Li, Beta1- and beta2-adrenoceptors in basolateral nucleus of amygdala and their roles in consolidation of fear memory in rats. Hippocampus, 2008. 18(11): p. 1131-9.

195. Roozendaal, B., B.S. McEwen, and S. Chattarji, Stress, memory and the amygdala. Nat Rev Neurosci, 2009. 10(6): p. 423-33.

141

196. de Quervain, D.J., et al., Glucocorticoids and the regulation of memory in health and disease. Front Neuroendocrinol, 2009. 30(3): p. 358-70.

197. Morimoto, M., et al., Distribution of glucocorticoid receptor immunoreactivity and mRNA in the rat brain: an immunohistochemical and in situ hybridization study. Neurosci Res, 1996. 26(3): p. 235-69.

198. Pugh, C.R., M. Fleshner, and J.W. Rudy, Type II glucocorticoid receptor antagonists impair contextual but not auditory-cue fear conditioning in juvenile rats. Neurobiol Learn Mem, 1997. 67(1): p. 75-9.

199. Kuhlmann, S., M. Piel, and O.T. Wolf, Impaired memory retrieval after psychosocial stress in healthy young men. J Neurosci, 2005. 25(11): p. 2977-82.

200. de Kloet, E.R., H. Karst, and M. Joels, Corticosteroid hormones in the central stress response: quick-and-slow. Front Neuroendocrinol, 2008. 29(2): p. 268-72.

201. Haller, J., E. Mikics, and G.B. Makara, The effects of non-genomic glucocorticoid mechanisms on bodily functions and the central neural system. A critical evaluation of findings. Front Neuroendocrinol, 2008. 29(2): p. 273-91.

202. Makara, G.B. and J. Haller, Non-genomic effects of glucocorticoids in the neural system. Evidence, mechanisms and implications. Prog Neurobiol, 2001. 65(4): p. 367-90.

203. Zheng, J., A. Ali, and V.D. Ramirez, Steroids conjugated to bovine serum albumin as tools to demonstrate specific steroid neuronal membrane binding sites. J Psychiatry Neurosci, 1996. 21(3): p. 187-97.

204. Chauveau, F., et al., Rapid stress-induced corticosterone rise in the hippocampus reverses serial memory retrieval pattern. Hippocampus, 2010. 20(1): p. 196-207.

205. de Quervain, D.J., A. Aerni, and B. Roozendaal, Preventive effect of beta- adrenoceptor blockade on glucocorticoid-induced memory retrieval deficits. Am J Psychiatry, 2007. 164(6): p. 967-9.

206. Rockman, H.A., W.J. Koch, and R.J. Lefkowitz, Seven-transmembrane-spanning receptors and heart function. Nature, 2002. 415(6868): p. 206-12.

207. Zheng, M., et al., Emerging concepts and therapeutic implications of beta- adrenergic receptor subtype signaling. Pharmacol Ther, 2005. 108(3): p. 257-68.

208. Johnson, M., The anti-inflammatory profile of fluticasone propionate. Allergy, 1995. 50(23 Suppl): p. 11-4.

142

209. Bachmann, C.G., et al., Effect of chronic administration of selective glucocorticoid receptor antagonists on the rat hypothalamic-pituitary- adrenocortical axis. Neuropsychopharmacology, 2003. 28(6): p. 1056-67.

210. Daaka, Y., L.M. Luttrell, and R.J. Lefkowitz, Switching of the coupling of the beta2-adrenergic receptor to different G proteins by protein kinase A. Nature, 1997. 390(6655): p. 88-91.

211. Devic, E., et al., Beta-adrenergic receptor subtype-specific signaling in cardiac myocytes from beta(1) and beta(2) adrenoceptor knockout mice. Mol Pharmacol, 2001. 60(3): p. 577-83.

212. Xiao, R.P., X. Ji, and E.G. Lakatta, Functional coupling of the beta 2- adrenoceptor to a pertussis toxin-sensitive G protein in cardiac myocytes. Mol Pharmacol, 1995. 47(2): p. 322-9.

213. Goh, J.W. and P.S. Pennefather, A pertussis toxin-sensitive G protein in hippocampal long-term potentiation. Science, 1989. 244(4907): p. 980-3.

214. Stratton, K.R., et al., Intrahippocampal injection of pertussis toxin blocks adenosine suppression of synaptic responses. Brain Res, 1989. 494(2): p. 359-64.

215. Abercrombie, E.D., R.W. Keller, Jr., and M.J. Zigmond, Characterization of hippocampal norepinephrine release as measured by microdialysis perfusion: pharmacological and behavioral studies. Neuroscience, 1988. 27(3): p. 897-904.

216. Vahabzadeh, A. and M. Fillenz, Comparison of stress-induced changes in noradrenergic and serotonergic neurons in the rat hippocampus using microdialysis. Eur J Neurosci, 1994. 6(7): p. 1205-12.

217. Chesley, A., et al., The beta(2)-adrenergic receptor delivers an antiapoptotic signal to cardiac myocytes through G(i)-dependent coupling to phosphatidylinositol 3'-kinase. Circ Res, 2000. 87(12): p. 1172-9.

218. Hurlemann, R., Noradrenergic-glucocorticoid mechanisms in emotion-induced amnesia: from adaptation to disease. Psychopharmacology (Berl), 2008. 197(1): p. 13-23.

219. Joels, M. and T.Z. Baram, The neuro-symphony of stress. Nat Rev Neurosci, 2009. 10(6): p. 459-66.

220. Bond, R.A., et al., Physiological effects of inverse agonists in transgenic mice with myocardial overexpression of the beta 2-adrenoceptor. Nature, 1995. 374(6519): p. 272-6.

143

221. Samama, P., et al., Negative antagonists promote an inactive conformation of the beta 2-adrenergic receptor. Mol Pharmacol, 1994. 45(3): p. 390-4.

222. Zhang, W.P., M. Ouyang, and S.A. Thomas, Potency of catecholamines and other L-tyrosine derivatives at the cloned mouse adrenergic receptors. Neuropharmacology, 2004. 47(3): p. 438-49.

223. Tronson, N.C. and J.R. Taylor, Molecular mechanisms of memory reconsolidation. Nat Rev Neurosci, 2007. 8(4): p. 262-75.

224. Kindt, M., M. Soeter, and B. Vervliet, Beyond extinction: erasing human fear responses and preventing the return of fear. Nat Neurosci, 2009. 12(3): p. 256-8.

225. Kroes, M.C., B.A. Strange, and R.J. Dolan, Beta-adrenergic blockade during memory retrieval in humans evokes a sustained reduction of declarative emotional memory enhancement. J Neurosci, 2010. 30(11): p. 3959-63.

226. Rohrer, D.K., et al., Targeted disruption of the mouse beta1-adrenergic receptor gene: developmental and cardiovascular effects. Proc Natl Acad Sci U S A, 1996. 93(14): p. 7375-80.

227. Chruscinski, A.J., et al., Targeted disruption of the beta2 adrenergic receptor gene. J Biol Chem, 1999. 274(24): p. 16694-700.

228. Bachman, E.S., et al., betaAR signaling required for diet-induced thermogenesis and obesity resistance. Science, 2002. 297(5582): p. 843-5.

229. Ouyang, M., et al., Adrenergic signaling plays a critical role in the maintenance of waking and in the regulation of REM sleep. J Neurophysiol, 2004. 92: p. 2071- 2082.

230. McClelland, J.L., B.L. McNaughton, and R.C. O'Reilly, Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. Psychol Rev, 1995. 102(3): p. 419-57.

231. Squire, L.R., R.E. Clark, and B.J. Knowlton, Retrograde amnesia. Hippocampus, 2001. 11(1): p. 50-5.

232. Rubin, D.C., D. Berntsen, and M.K. Bohni, A memory-based model of posttraumatic stress disorder: evaluating basic assumptions underlying the PTSD diagnosis. Psychol Rev, 2008. 115(4): p. 985-1011.

233. Tonegawa, S., K. Nakazawa, and M.A. Wilson, Genetic neuroscience of mammalian learning and memory. Philos Trans R Soc Lond B Biol Sci, 2003. 358(1432): p. 787-95.

144

234. Kandel, E.R., The molecular biology of memory storage: a dialogue between genes and synapses. Science, 2001. 294(5544): p. 1030-8.

235. Morris, R.G., et al., Elements of a neurobiological theory of the hippocampus: the role of activity-dependent synaptic plasticity in memory. Philos Trans R Soc Lond B Biol Sci, 2003. 358(1432): p. 773-86.

236. Martin, S.J. and R.E. Clark, The rodent hippocampus and spatial memory: from synapses to systems. Cell Mol Life Sci, 2007. 64(4): p. 401-31.

237. Szapiro, G., et al., Molecular mechanisms of memory retrieval. Neurochem Res, 2002. 27(11): p. 1491-8.

238. Isiegas, C., et al., A novel conditional genetic system reveals that increasing neuronal cAMP enhances memory and retrieval. J Neurosci, 2008. 28(24): p. 6220-30.

239. Mann, D.M., et al., Pathological evidence for neurotransmitter deficits in Down's syndrome of middle age. J Ment Defic Res, 1985. 29 ( Pt 2): p. 125-35.

240. Marcyniuk, B., et al., Topography of nerve cell loss from the locus caeruleus in middle aged persons with Down's syndrome. J Neurol Sci, 1988. 83(1): p. 15-24.

241. Dierssen, M., et al., Alterations of central noradrenergic transmission in Ts65Dn mouse, a model for Down syndrome. Brain Res, 1997. 749(2): p. 238-44.

242. Salehi, A., et al., Restoration of norepinephrine-modulated contextual memory in a mouse model of Down syndrome. Sci Transl Med, 2009. 1(7): p. 7ra17.

243. Malta, E., M.A. Mian, and C. Raper, The in vitro pharmacology of xamoterol (ICI 118,587). Br J Pharmacol, 1985. 85(1): p. 179-87.

244. Lardizabal, J.A. and P.C. Deedwania, The current state of beta blockers in hypertension therapy. Indian Heart J. 62(2): p. 111-7.

245. Cruickshank, J.M., Beta-blockers and heart failure. Indian Heart J. 62(2): p. 101- 10.

246. Marlow, H.F., Xamoterol, a beta 1-adrenoceptor partial agonist: review of the clinical efficacy in heart failure. Br J Clin Pharmacol, 1989. 28 Suppl 1: p. 23S- 30S.

247. Sriwatanakul, K. and S.R. Nahorski, Disposition and activity of beta- adrenoceptor antagonists in the rat using an ex vivo receptor binding assay. Eur J Pharmacol, 1980. 66(2-3): p. 169-78.

145

248. Tsuchihashi, H., et al., Characteristics of 125I-iodocyanopindolol binding to beta- adrenergic and serotonin-1B receptors of rat brain: selectivity of beta-adrenergic agents. Jpn J Pharmacol, 1990. 52(2): p. 195-200.

249. Beer, M., et al., In vitro selectivity of agonists and antagonists for beta 1- and beta 2-adrenoceptor subtypes in rat brain. Biochem Pharmacol, 1988. 37(6): p. 1145-51.

250. Liang, K.C., C. Bennett, and J.L. McGaugh, Peripheral epinephrine modulates the effects of post-training amygdala stimulation on memory. Behav Brain Res, 1985. 15(2): p. 93-100.

251. Chen, X., et al., PI3 kinase signaling is required for retrieval and extinction of contextual memory. Nat Neurosci, 2005. 8(7): p. 925-31.

252. Williams, B.R., R. Barber, and R.B. Clark, Kinetic analysis of agonist-induced down-regulation of the beta(2)-adrenergic receptor in BEAS-2B cells reveals high- and low-affinity components. Mol Pharmacol, 2000. 58(2): p. 421-30.

253. Srivastava, D.P., et al., Rapid, nongenomic responses to ecdysteroids and catecholamines mediated by a novel Drosophila G-protein-coupled receptor. J Neurosci, 2005. 25(26): p. 6145-55.

254. Wettschureck, N. and S. Offermanns, Mammalian G proteins and their cell type specific functions. Physiol Rev, 2005. 85(4): p. 1159-204.

255. McOmish, C.E., et al., PLC-beta1 knockout mice as a model of disrupted cortical development and plasticity: behavioral endophenotypes and dysregulation of RGS4 gene expression. Hippocampus, 2008. 18(8): p. 824-34.

256. Straiker, A.J., C.R. Borden, and J.M. Sullivan, G-protein alpha subunit isoforms couple differentially to receptors that mediate presynaptic inhibition at rat hippocampal synapses. J Neurosci, 2002. 22(7): p. 2460-8.

257. Matus-Amat, P., et al., The role of dorsal hippocampus and basolateral amygdala NMDA receptors in the acquisition and retrieval of context and contextual fear memories. Behav Neurosci, 2007. 121(4): p. 721-31.

258. McReynolds, J.R., et al., Memory-enhancing corticosterone treatment increases amygdala norepinephrine and Arc protein expression in hippocampal synaptic fractions. Neurobiol Learn Mem. 93(3): p. 312-21.

259. Zhang, W.P., J.F. Guzowski, and S.A. Thomas, Mapping neuronal activation and the influence of adrenergic signaling during contextual memory retrieval. Learn Mem, 2005. 12(3): p. 239-47.

146

260. Egawa, M., B.G. Hoebel, and E.A. Stone, Use of microdialysis to measure brain noradrenergic receptor function in vivo. Brain Res, 1988. 458(2): p. 303-8.

261. Stone, E.A., M. Egawa, and C.M. Colbjornsen, Catecholamine-induced desensitization of brain beta adrenoceptors in vivo and reversal by corticosterone. Life Sci, 1989. 44(3): p. 209-13.

262. Boyle, M.P., et al., Acquired deficit of forebrain glucocorticoid receptor produces depression-like changes in adrenal axis regulation and behavior. Proc Natl Acad Sci U S A, 2005. 102(2): p. 473-8.

263. Tsien, J.Z., et al., Subregion- and cell type-restricted gene knockout in mouse brain. Cell, 1996. 87(7): p. 1317-26.

264. Hill, J.W., et al., Direct insulin and leptin action on pro-opiomelanocortin neurons is required for normal glucose homeostasis and fertility. Cell Metab. 11(4): p. 286-97.

265. Kask, K., et al., Developmental profile of kainate receptor subunit KA1 revealed by Cre expression in YAC transgenic mice. Brain Res, 2000. 876(1-2): p. 55-61.

266. Gumy, C., et al., Dibutyltin disrupts glucocorticoid receptor function and impairs glucocorticoid-induced suppression of cytokine production. PLoS One, 2008. 3(10): p. e3545.

267. Sanchez, E.R., et al., Hormone-free mouse glucocorticoid receptors overexpressed in Chinese hamster ovary cells are localized to the nucleus and are associated with both hsp70 and hsp90. J Biol Chem, 1990. 265(33): p. 20123-30.

268. Srinivasan, G. and E.B. Thompson, Overexpression of full-length human glucocorticoid receptor in Spodoptera frugiperda cells using the baculovirus expression vector system. Mol Endocrinol, 1990. 4(2): p. 209-16.

269. Higo, S., et al., Subtypes of GABAergic neurons project axons in the neocortex. Front Neuroanat, 2009. 3: p. 25.

270. Soda, Y., et al., Feature Article: Transdifferentiation of glioblastoma cells into vascular endothelial cells. Proc Natl Acad Sci U S A.

271. Lisman, J.E., Relating hippocampal circuitry to function: recall of memory sequences by reciprocal dentate-CA3 interactions. Neuron, 1999. 22(2): p. 233- 42.

272. Moser, E.I. and O. Paulsen, New excitement in cognitive space: between place cells and spatial memory. Curr Opin Neurobiol, 2001. 11(6): p. 745-51.

147

273. Murchison, C.F., et al., Norepinephrine and beta1-Adrenergic Signaling Facilitate Activation of Hippocampal CA1 Pyramidal Neurons During Contextual Memory Retrieval. Neuroscience.

274. Shimizu, E., et al., NMDA receptor-dependent synaptic reinforcement as a crucial process for memory consolidation. Science, 2000. 290(5494): p. 1170-4.

275. Coulter, D.A., et al., Classical conditioning reduces amplitude and duration of calcium-dependent afterhyperpolarization in rabbit hippocampal pyramidal cells. J Neurophysiol, 1989. 61(5): p. 971-81.

276. Oh, M.M., et al., Watermaze learning enhances excitability of CA1 pyramidal neurons. J Neurophysiol, 2003. 90(4): p. 2171-9.

277. Haug, T. and J.F. Storm, Protein kinase A mediates the modulation of the slow Ca(2+)-dependent K(+) current, I(sAHP), by the neuropeptides CRF, VIP, and CGRP in hippocampal pyramidal neurons. J Neurophysiol, 2000. 83(4): p. 2071- 9.

278. Madison, D.V. and R.A. Nicoll, Actions of noradrenaline recorded intracellularly in rat hippocampal CA1 pyramidal neurones, in vitro. J Physiol, 1986. 372: p. 221-44.

279. Madison, D.V. and R.A. Nicoll, Cyclic adenosine 3',5'-monophosphate mediates beta-receptor actions of noradrenaline in rat hippocampal pyramidal cells. J Physiol, 1986. 372: p. 245-59.

280. Rascol, O., et al., Effects of calcium channel agonist and antagonists on calcium- dependent events in CA1 hippocampal neurons. Fundam Clin Pharmacol, 1991. 5(4): p. 299-317.

281. Joels, M. and E.R. de Kloet, Effects of glucocorticoids and norepinephrine on the excitability in the hippocampus. Science, 1989. 245(4925): p. 1502-5.

282. Karst, H. and M. Joels, Brief RU 38486 treatment normalizes the effects of chronic stress on calcium currents in rat hippocampal CA1 neurons. Neuropsychopharmacology, 2007. 32(8): p. 1830-9.

283. Kerr, D.S., et al., Corticosteroid modulation of hippocampal potentials: increased effect with aging. Science, 1989. 245(4925): p. 1505-9.

148