PSD-95 REGULATES SEROTONIN RECEPTOR FUNCTION in VIVO By

Home , Frizzled, Quipazine, Sertindole, Smoothened

PSD-95 REGULATES SEROTONIN RECEPTOR FUNCTION IN VIVO

ATHEIR IBRAHIM ABBAS

Submitted in partial fulfillment of the requirements

for the degree of Doctor of Philosophy

Dissertation Adviser: Dr. Bryan Roth, M.D., Ph.D.

Department of Biochemistry

CASE WESTERN RESERVE UNIVERSITY

May, 2009

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

Atheir Ibrahim Abbas ______candidate for the Ph.D. degree*.

(signed) Martin D. Snider, Ph.D. ______(chair of the committee)

Bryan L. Roth, M.D., Ph.D. ______

William C. Merrick, Ph.D. ______

Paul R. Ernsberger, Ph.D. ______

Vernon Anderson, Ph.D. ______

George Dubyak, Ph.D. ______

th October 24 , 2008 (date) ______

* We also certify that written approval has been obtained for any

proprietary material contained therein.

TABLE OF CONTENTS

LIST OF TABLES vii

LIST OF FIGURES viii

LIST OF ABBREVIATIONS xi

ACKNOWLEDGEMENTS xv

ABSTRACT xvii

CHAPTER 1: Introduction

1.1 - G-protein-Coupled Receptors……………………………………………………1

1.2 - GPCR Signal Transduction and Regulation…………………………………....2

1.2.1 – Canonical GPCR Signaling………………………………………….2

1.3. - Multiplicity of GPCR Signaling and Regulation……………………………..…3

1.3.1 – Mechanisms of GPCR Desensitization………………………………3

1.3.2 – Regulation of GPCR Internalization, Trafficking, and

Resensitization……………………………………………..…………..4

1.3.3 – Non-canonical GPCR Signaling………………………….…………5

1.4 – Models of GPCR Behavior………………………………………………………6

1.4.1 – Classic GPCR Models…………………………………………………6

1.4.2 – Functional Selectivity…………………………………………………..7

1.5 – 5-HT2A Receptors……………………………………………………………….10

1.5.1 – 5-HT2A Receptor Neuroanatomy………………………………….…10

1.5.2 – Hallucinogens…………………………………………………………11

1.5.2.1 – Hallucinogens are 5-HT2A agonists……………………….11

1.5.2.2 – Signaling via the 5-HT2A Receptor……………………..…12

1.5.2.3 – The Neuronal Correlates of Consciousness………..……19

1.5.2.4 – The Presynaptic, Thalamocortical Hypothesis of

Hallucinogen Action….………………………………….….20

1.5.2.5 – The Postsynaptic, Corticocortical Hypothesis of

Hallucinogen Action…………..………………………….…23

1.5.2.6 – Hallucinogen Action – From Neurochemistry to Altered

States of Consciousness (ASCs) …………………………25

1.5.3 – The Neurochemical Basis for Schizophrenia and Psychosis…….33

1.5.4 – Animal Models of Psychosis…………………….…………………..37

1.5.5 – Atypical Antipsychotics………………………….……………………40

1.5.5.1 - Atypical Antipsychotics are Potent 5-HT2A Antagonists…40

1.6 – 5-HT2C Receptors………………………………….……………………………42

1.6.1 – 5-HT2C Overview and Neuroanatomy………………………………42

1.6.2 – 5-HT2C Receptor Function………………………………….………..43

1.6.2.1 – 5-HT2C Receptor Signaling…………..…………………...43

1.6.2.2 – RNA Editing of the 5-HT2C Receptor……………………..45

1.6.2.3 – 5-HT2C Receptor Modulation of Synaptic Activity and

Associated Behaviors………..………………...……….…47

CHAPTER 2: Materials and Methods

2.1 – Materials……………………………...……….……..…………………………..50

2.1.1 – Chemicals…………………….………………………………………50

2.1.2 – Mice…………………………..….…………………………………….50

2.1.3 – cDNA Constructs……………………………..…………...……….…51

2.1.4 – Antibodies…………………………….……….……………………..51

2.2 – Methods……………………………...……….……………………….…………51

2.2.1 – Immunochemistry………………………..…………………………...51

2.2.2 – Saturation Radioligand Binding……………………………...……...52

2.2.3 – EEDQ Time Course………………….………………………………54

2.2.4 – Quantitative RT PCR……………………………...……….…………54

2.2.5 – Microarray Experiment……………………………...……….…….…55

2.2.6 – 5-HT2C mRNA Editing……………………………...……….…….….55

2.2.7 – Cortical Neuronal Cultures……………………………...……….…..56

2.2.8 – Lentiviral Preparation……………………………..……….………...56

2.2.9 – MK-212-induced c-fos in Hippocampus…………………………….57

2.2.10 – DOI-induced Head Twitch……………………………...……….….57

2.2.11 – PPI……………………………………………………………...…….58

CHAPTER 3: PSD-95 Regulates 5-HT2A Receptor Function in vivo

3.1 – Introduction and Rationale……………………………...……….……………..60

3.2 – Results……………………………………...…………………………………...62

3.2.1 – Genetic Deletion of PSD-95 Results in a Selective Loss of 5-HT2A

Receptors …………………………………….……………………...62

3.2.1.1 – 5-HT2A Immunochemistry……………………………...….62

3.2.1.2 – Measuring 5-HT2A Receptor Density ………………...….63

iii

3.2.2 – PSD-95 Does Not Modulate Expression of the

5-HT1A Receptor……………………………………………………....64

3.2.3 – PSD-95 Does Not Play a Prominent Role in Modulating Gene

Expression……………………………………..…………………...….65

3.2.3.1 – PSD-95 Does Not Alter 5-HT2A Receptor

mRNA Levels……………………………………….……….65

3.2.3.2 – Global Gene Expression Does Not Change Substantially

in the Absence of PSD-95……………………………...….66

3.2.4 – 5-HT2A Receptor Turnover Rate is Accelerated in the Absence of

PSD-95……………………………...….………………………………69

3.2.5 – PSD-95 is Required for Normal Expression and Polarized Sorting

of 5-HT2A Receptors to Pyramidal Neuron Apical Dendrites….….72

3.2.5.1 – 5-HT2A Receptor Expression and Dendritic Targeting is

Attenuated in Neurons Prepared From PSD 95 Knockout

Mice………………………………………………………….72

3.2.5.2 – Lentiviral Addback of PSD-95 to Knockout Cortical

Neurons Rescues Targeting and Expression of 5-HT2A

Receptors……………………………………………………75

3.2.6 – PSD-95 Helps Mediate Hallucinogen Actions in vivo…..………..79

3.2.7 – Deletion of PSD-95 Renders Clozapine “Propsychotic” …………83

3.3 – Discussion……………………………………...….…………………………….90

3.3.1 – Major Findings………………………………....……………………...90

CHAPTER 4: PSD-95 Regulates 5-HT2C Receptor Function in vivo

4.1 – Introduction and Rationale………………………..…...….…………………..97

4.2 – Results…………………….……………...…………………………………...... 99

4.2.1 – Deletion of PSD-95 Results in a Selective Loss of 5-HT2C

Receptors …………………………………………………………..…99

4.2.1.1 – 5-HT2C Immunochemistry……………………………...…..99

4.2.1.2 – 5-HT2C Saturation Binding…………………………….....100

4.2.2 – PSD-95 Does Not Modulate Expression of the

5-HT1A Receptor……………………………………………………..101

4.2.3 – PSD-95 Does Not Play a Prominent Role in Modulating Gene

Expression……………………………...….…………………………102

4.2.3.1 – PSD-95 Does Not Alter 5-HT2C Receptor Levels Via

Transcriptional or Post-transcriptional Mechanisms…..102

4.2.3.2 – PSD-95 Does Not Modulate RNA Editing of the 5-HT2C

Receptor…………………………...….……………………103

4.2.4 – PSD-95 is Required for 5-HT2C Signaling in vivo…..………...….105

4.3 – Discussion……………………………………...….…………………………...110

4.3.1 – Major Findings……………………………………………………….110

CHAPTER 5: Future Directions

5.1 – Regulation of Cortical 5-HT2A Receptor Function, Mechanisms of

Hallucinogen Action, and the Basis for Psychosis-Related Signaling

Events………………………………..………………………………………....114

5.1.1 – Exploring the Relative Importance of PDZ Domain-Mediated

Interactions With Respect to Neuronal

5-HT2A Receptor Function……………………………………….…114

5.1.2 – 5-HT2A Receptor Functional Selectivity as it Relates to

Hallucinogen Action……………...….…..………………………….117

5.1.3 – Hallucinogenic Signaling Events Downstream of

5-HT2A Activation……….……..………………………….………...118

5.1.4 – Exploring the Extent of Macromolecular Disruption and How

Different Proteins in the Extended PSD-95 Scaffolded Network

Modulate 5-HT2A Receptor Function……………………………….119

LIST OF TABLES

Table 1.1 – G-protein Subunits and Their Effectors…………………………………3

Table 1.2 – Rank Order Efficacies of Five 5-HT2A Agonists in the IP and AA

Signaling Pathways……………………………………………………14

Table 1.3 – C-terminal sequences of the 5-HT2 receptors………………………..27

Table 1.4 – Antipsychotic Animal Models………………………………….………..38

Table 1.5 – Rank Order Efficacies of Five 5-HT2C Agonists in the IP and AA

Signaling Pathways……………………….…..………………………44

Table 3.1 – All Genes Affected in PSD-95 Knockout Mice…………..……………67

Table 3.2 – Genes of Interest Affected in PSD-95 Knockout Mice………………68

Table 5.1 – Regulation of 5-HT2A-mediated Hallucinogen-Related Signaling…115

Table 5.2 – Possible Outcomes of Group I mGluR Experiment in the Presence

and Absence of PSD-95…………………………..………….……….121

vii

LIST OF FIGURES

Figure 1.1 – 5-HT2A Signaling Pathways……………………………………………13

Figure 1.2 – The Presynaptic and Postsynaptic Hypotheses of Hallucinogen

Action………………………………….……………………………..…24

Figure 1.3 – PSD-95-interacting Proteins…………………………………………...28

Figure 1.4 – The PSD-95-scaffolded Postsynaptic Glutamatergic Signaling

Complex………………………………….……………………………..30

Figure 1.5 – Model for 5-HT2A-mediated Hallucinogenic Signaling at the

Postsynaptic Density………………………………….………………32

Figure 1.6 – 5-HT2C Signaling Pathways……………………………………………43

Figure 1.7 – RNA Editing of the 5-HT2C Receptor pre-mRNA………………….…46

Figure 3.1 – 5-HT2A Immunochemistry……………………………...……….……...63

Figure 3.2 – 5-HT2A Bmax Measurements in PSD-95 Mice………………………...64

Figure 3.3 – 5-HT1A Bmax Measurements in PSD-95 Mice………………………...65

Figure 3.4 – 5-HT2A Receptor mRNA Levels in Cortex……………………………66

Figure 3.5 – Modeling in vivo Receptor Turnover Using EEDQ………………….70

Figure 3.6 – Modeling 5-HT2A Turnover Kinetics……………………………...……71

Figure 3.7 – 5-HT2A Receptor Turnover Rate Constant Comparison…………….71

Figure 3.8 – 5-HT2A and MAP2 Immunochemistry in Cortical Neurons………….73

Figure 3.9 – Quantitative Comparison of 5-HT2A Expression in Cultured Cortical

Neurons……………………………………...….……………………...74

Figure 3.10 – Quantitative Comparison of 5-HT2A Dendritic Targeting in Cultured

Cortical Neurons……………………………………………………….74

viii

Figure 3.11 – 5-HT2A Immunochemistry in PSD-95 Knockout Cortical Neurons

Infected with GFP or PSD-95-GFP Lentivirus.. ……………………76

Figure 3.12 – Quantitative Comparison of 5-HT2A Expression in GFP- and PSD-

95-GFP Infected Cortical Neurons……………………………………….………….78

Figure 3.13 – Quantitative Comparison of 5-HT2A Dendritic Targeting in GFP- and

PSD-95-GFP Infected Cortical Neurons…………………..…………………...…...79

Figure 3.14 – DOI-induced Head Twitch in PSD-95 Mice…………………………81

Figure 3.15 – .3 mg/kg DOI-induced Head Twitch Time Course…………………81

Figure 3.16 – 1 mg/kg DOI-induced Head Twitch Time Course …………………82

Figure 3.17 – 5 mg/kg DOI-induced Head Twitch Time Course………………….82

Figure 3.18 – Dose Dependent Increase in Head Twitch is Greater in PSD-95

Wildtype Mice …………………………………….……………………83

Figure 3.19 – Comparison of Raw Acoustic Startle Responses………………….85

Figure 3.20 – Measuring the Effect of Clozapine on PCP-induced Disruption of

PPI in PSD-95 Mice…………………………………………………...86

Figure 3.21 – Clozapine Effect on PCP-induced Disruption of PPI at 16 dB

Prepulse in PSD-95 Mice………………………...…………………...87

Figure 3.22 – Effect of Increasing Doses of Clozapine on PCP-induced

Disruption of PPI…………………………….…………………………88

Figure 4.1 – 5-HT2C Immunochemistry in PSD-95 Mice…………………………..99

Figure 4.2 – 5-HT2C Bmax Measurements in PSD-95 Mice…………………...….100

Figure 4.3 – 5-HT2C Bmax Measurements in PSD-95 Mice……………………….101

Figure 4.4 – 5-HT2C Receptor mRNA Levels in Hippocampus………………….102

Figure 4.5 – 5-HT2C mRNA Editing Frequencies in PSD-95 Mice by Site……..104

Figure 4.6 – Frequencies of Edited Isoforms in PSD-95…………………………104

Figure 4.7 – MK-212 Induction of c-fos in the Hippocampus of PSD-95 Mice..106

Figure 4.8 – c-fos Induction in 5-HT2C-Expressing Cells in Hippocampus……..108

Figure 4.9 – c-fos Quantitation in PSD-95 Wildtype and Knockout

Hippocampus……………………………………………………….…109

LIST OF ABBREVIATIONS

5-HT2A – 5-hydroxytryptamine2A

AA – arachidonic acid

AC – adenylate cyclase

AD - Alzheimer’s Disease

ADAR - adenosine deaminase that acts on RNA

ADP - adenosine 5’-diphosphate

AKAP79/150 – A-kinase anchoring protein

Akt - AKT8 virus oncogene cellular homolog

ARF1 – ADP ribosylation factor 1

ASC – altered state of consciousness

β1AR – Beta-1 adrenergic receptor

BAI1 – brain angiogenesis factor 1 cAMP – cyclic adenosine monophosphate

CAR – conditioned avoidance response cGMP – cyclic guanosine 5’-monophosphate

CNV – copy number variant

COMT - catechol-O-methyl transferase

CRIPT – cysteine rich interactor of PDZ3

CTC – cubic ternary complex

δ2 GluR – delta2 ionotropic glutamate receptor

D2 – dopamine 2 receptor

DAG – diacyl glycerol

DISC1 - disrupted-in-schizophrenia-1

DOI - 2,5-dimethoxy-4-iodoamphetamine

EGFR – epidermal growth factor receptor egr-2 – early growth response 2

EPS – extrapyramidal symptoms

EPSC - excitatory postsynaptic current

EPSP - excitatory postsynaptic potential

ERK1/2 – extracellular signal-related kinase

ETC - extended ternary complex

FRT - forelimb retraction time

GDP – guanosine 5’-diphosphate

GK – guanylate kinase

GKAP – guanylate kinase associated protein

GluR6 – kainite receptor subunit 6

GPCR – G protein-coupled receptor

GRK – G protein-couple receptor kinase

GTP – guanosine 5’-triphosphate

HRT - hindlimb retraction time

1 4 5 ICI 154129 - [N,N’-diallyl-Tyr , Ψ (CH2S)Phe ,Leu ]enkephalin

ICI 174864 - N,N’-diallyl-Tyr1,Aib2,3,Leu5]enkephalin

IP – inositol phosphate

IP3 – inositol-1,4,5-triphosphate

JNK - Jun N-terminal kinase

xii

KA2 GluR – kainite receptor subunit

KIF1Bα - kinesin family member 1 B alpha

LSD- lysergic acid diethylamide

LTD – long-term depression

LTP - long-term potentiation

MAGUK - membrane-associatied guanylate kinase

mCPP - 2-(4-chloro-2-methylphenoxy)propanoic acid

NLS – neuroleptic malignant syndrome

NMDAR – N-methyl-D-aspartate receptor

nNOS – neuronal nitric oxide synthase

NR2A/2B/2C/2D – NMDA receptor subunit 2A/2B/2C/2D

NRG – neuregulin

NRPTK – non-receptor protein tyrosine kinase

p38 - phosphoprotein of 38 kDa

PACAP - pituitary adenylyl-cyclase activating peptide

PACAP-R - pituitary adenylyl-cyclase activating peptide receptor

PDD - Parkinson’s Disease Dementia

PDZ - PSD95/dlg/zonular occludens-1

PI – phosphatidyl inositol

PI3K - phosphatidylinositol 3-kinase

PIP2 – phosphatidylinositol 4’,5’-bisphosphate

PKA – protein kinase A

PKC – protein kinase C

xiii

PKCα – protein kinase C alpha

PLA2 - phospholipase A2

PLCβ - phospholipase Cβ

PLD – phospholipase D

PMCA – plasma membrane Ca2+ ATPase

PFC – prefrontal cortex

PP2 - 3-(4-chlorophenyl) 1-(1,1-dimethylethyl)-1H-pyrazolo[3,4-d]pyrimidin-

4amine

PPI – prepulse inhibition

PSD-95 - postsynaptic density protein of 95kDa

RGS – regulator of G protein signaling

RhoA - ras homolog gene family, member A

SAPAP – SAP90/PSD-95-associated protein

Sema 4b, 4c – semaphorin 4b, 4c

SH3 - Src homology 3

SPAR – spine-associate RapGAP

Src - Rous sarcoma oncogene cellular homolog

SynGAP – synaptic GTPase-activating protein

TFMPP - m-trifluoromethylphenylpiperazine

xiv

ACKNOWLEDGEMENTS

I would first like to thank Dr. Bryan Roth for his enthusiasm, guidance, and patience during my training. In his laboratory I went from novice to scientist

(though still with much to learn), and I will always have the greatest respect for

Dr. Roth’s breadth of knowledge and, most of all, his approach to science. I have

learned to focus on asking tough questions to test my hypotheses, and to always

welcome an unexpected result as an opportunity, not a roadblock. He practices

science in an honorable, conscientious, and collaborative manner, and I would

be hard pressed to find a mentor with whom I could have learned more.

I am also thankful to Case Western Reserve University and especially the MSTP

program for the opportunity to learn and research at a world class university.

I am grateful to my committee members, Drs. William C. Merrick, Martin D.

Snider, Paul R. Ernsberger, Vernon E. Anderson, and George R. Dubyak for

taking the time out of their busy schedules to guide and oversee my

development.

I would also like to thank the many Roth lab members from whom I have learned and with whom I have had countless fruitful discussions. In particular, I thank

Blaine Armbruster, Tim Vortherms, Vincent Setola, Ryan Strachan, and Feng

Yan for their help and friendship. I also thank the multitude of other Roth lab members with whom I have worked in the last 4 years.

Finally, thanks to my parents and two younger brothers for their loving support throughout my studies. I would not be who I am and could not have accomplished what I have without their boundless love and support.

xvi

PSD-95 Regulates Serotonin Receptor Function in vivo

Abstract

ATHEIR IBRAHIM ABBAS

The 5-hydroxytryptamine 2A (5-HT2A) receptor, a target for hallucinogens and

some antipsychotics, is thought to play a prominent role in regulating mood,

perception, and cognition. The closely related 5-hydroxytryptamine 2C (5-HT2C)

receptor is also thought to be involved in a number of central nervous system

processes including mood and temperature regulation. Due to the behavioral

effects that result from activation and blockade of 5-HT2A and 5-HT2C receptors, it

has been suggested that these receptors can modulate glutamatergic

neurotransmission, though the biochemical links between the metabotropic

serotonin and ionotropic glutamate systems have remained a mystery. In the

studies presented herein we show that the postsynaptic PDZ domain-containing

scaffolding protein postsynaptic density protein of 95kDa (PSD-95), a 5-HT2A/2C- interacting protein, is an important biochemical link between the serotonin and glutamate systems. We show that, in the absence of PSD-95 in vivo, 5-HT2A and

5-HT2C receptor expression is reduced due to an increase in the rate of receptor

turnover. We also provide evidence that targeting to the appropriate apical

dendritic compartment is impaired in neurons cultured from PSD-95 knockout mice, and that lentiviral addback of PSD-95 to knockout neurons rescues targeting. We also examine signaling at both the biochemical and behavioral

xvii

level. With respect to the 5-HT2C receptor, we show that the ability of a 5-HT2C agonist to induce c-fos, a marker of neuronal activation, is greatly reduced in the absence of PSD-95. We also present data showing that 5-HT2A-mediated

hallucinogen-induced head twitch is also reduced in the absence of PSD-95.

Finally, we provide evidence that clozapine, which is thought to correct the

abnormalities in glutamatergic neurontransmission seen in some animal models

of psychosis via a 5-HT2A-dependent mechanism, is unable to exert its

therapeutic efficacy in PSD-95 knockout mice. Together the data presented

herein provide the first biochemical link between the metabotropic serotonin and

ionotropic glutamate systems. Our studies also suggest that this link is relevant

not only with respect to the regulation of 5-HT2A and 5-HT2C receptor function, but also with respect to hallucinogen action and the neurochemistry underlying psychosis.

xviii

CHAPTER 1: Introduction

1.1 – G protein-Coupled Receptors

G protein-coupled receptors (GPCRs) represent a large and diverse class of proteins in mammalian genomes, and their presence in organisms ranging from bacteria to higher vertebrates suggests an early evolutionary origin (1). GPCR membrane topology is characterized by an extracellular N-terminus, seven transmembrane α-helices connected by alternating intracellular and extracellular loops, and an intracellular C-terminus (2), (3). A number of rhodopsin and β2- adrenergic GPCR crystal structures over the past decade have been published, all very similar in their basic topology, and all confirming a prototypical α-helical seven transmembrane structure (4-8).

GPCRs have been classified into seven families in mammals: A, B, C,

Frizzled/Smoothened, large N-terminal family B-7 transmembrane helix, taste 2, and vomeronasal 1 receptors (3). It has been estimated that roughly 80% of known hormones and neurotransmitters activate GPCRs. Furthermore, it is thought that GPCRs represent some 30-50% of commercially available drug targets (1), (3). It has been demonstrated in a wide variety of cases that ligand binding to GPCRs typically involves key residues in the pocket formed by the seven transmembrane helices, with varying participation of extracellular N- terminus or loop residues depending on the GPCR class and ligand (9).

1.2 - GPCR Signal Transduction and Regulation

1.2.1 - Canonical GPCR Signaling

In the canonical view of GPCR signaling, ligand binding to the receptor either

stabilizes or induces a conformational change that enables the receptor to

associate with heterotrimeric G-proteins composed of an α subunit and a βγ dimer (9), (10). Binding of the heterotrimeric G protein to the receptor promotes

exchange of an α subunit-associated guanosine 5’-diphosphate (GDP) for

guanosine 5’-triphosphate (GTP), which in turn promotes dissociation of the α

subunit from the βγ dimer, both of which are signaling intermediates which initiate

a range of actions depending on which α, β, and γ subunits are involved (9).

Hydrolysis of the α subunit bound GTP to GDP, which can happen due to its intrinsic GTPase activity (9) or with the aid of GTPase-activating proteins (GAPs)

(3), leads to inactivation of the α subunit and promotes its reassociation with the

βγ dimer to form the inactive heterotrimeric G-protein.

The existence of 23 different α subunits, six β subunits, and 12 distinct γ subunits

imparts significant complexity to the possible downstream signaling (3), (9). α

subunit-catalyzed signaling has been the most extensively studied, and receptors

are often classed according to the α subunit to which they most commonly

couple (10). The 23 different α subunits are grouped into four major families: αs,

αi/o, αq/11, α12 (3), (9). The majority of receptor subtypes comprising the major

2 neurotransmitter systems in the brain (serotonergic, dopaminergic, cholinergic, and noradrenergic) couple prototypically to α , α , or α (1). Table 1.1 (3) lists the s i q canonical downstream effectors of these three major G proteins.

Table 1.1 – G-protein Subunits and Their Effectors

G-protein subunit Effector and Manner of Regulation

αs Stimulation of adenylate cyclase (AC1-9)

αi Inhibition of adenylate cyclase (AC5, AC6)

αq Stimulation of phospholipase Cβ (PLCβ1-4)

1.3. - Multiplicity of GPCR Signaling and Regulation

1.3.1 - Mechanisms of GPCR Desensitization

One mechanism by which GPCR signaling is regulated is receptor desensitization, which can happen in a number of different ways. Protein kinase

A (PKA) or protein kinase C (PKC) signaling downstream of GPCR activation can negatively feedback on the activated receptor by phosphorylating it, which is termed homologous desensitization (10). PKA and PKC activation can also phosphorylate non-signaling receptors, leading to heterologous desensitization

(10), and thus general cellular hypo-responsiveness in many instances (3). A family of proteins called G protein-coupled receptor kinases (GRKs) is responsible for another form of homologous desensitization which targets active

(typically agonist-occupied) receptors for phosphorylation, recruiting specific

arrestins and inhibiting further interactions with G-proteins (10), (3). Another

protein family, the RGS proteins, functions as GAPs, regulating GPCR signal

transduction by increasing the rate of GTP hydrolysis (3). The particular

mechanism or mechanisms by which a particular GPCR can be desensitized are

receptor-, cell type/tissue type-, and ligand- dependent (i.e., dependent on the

local conformational distribution and environment of the receptor) (3), (11).

1.3.2 - Regulation of GPCR Internalization, Trafficking, and Resensitization

Some of the same proteins (e.g., GRKs and arrestins) that mediate homologous

agonist-induced desensitization are also involved in regulating GPCR

internalization (11). It should be noted that though internalization was initially

conceptualized as part of the process of a longer term form of desensitization

that is often associated with receptor down-regulation (internalization being

necessary for targeting to lysosomal degradative compartments), it has become

increasingly clear in recent years that internalization is likely also important in

regulating alternative, non-degradative intracellular trafficking pathways. These

alternative pathways include resensitization, and non-G protein-mediated

signaling pathways initiated by GPCR-arrestin interactions (10), (11). The

possibility of disconnecting desensitization and internalization from down-

regulation has been shown for a number of receptors, including 5-

hydroxytryptamine2A (5-HT2A) receptors (12), (13). Furthermore, 5-HT2A receptors are relatively unique in that agonists and antagonists induce

downregulation, which is contrary to the predictions of classical pharmacology

(antagonists should induce upregulation) (14). As with short-term desensitization, a GPCR’s trafficking and resensitization “signature” is often receptor-, cell type/tissue type-, and ligand-dependent (3), (11), (10), (15).

1.3.3 - Non-canonical GPCR Signaling

The prototypical mediators of GPCR internalization are the arrestins, a family of adaptor proteins of which there are four members, arrestins 1-4 (16). Arrestins 1 and 4 are expressed in eye, while 2 and 3 (β-arrestins 1 and 2) are expressed elsewhere (16). The long established role for arrestins is in mediating internalization of GPCRs after receptor activation and/or GRK-mediated desensitization (17). Internalization takes place because arrestins have higher affinity for both the active state of the receptor and the GRK-phosphorylated form of the receptor (17). What has only recently been made clear is that arrestins can mediate G-protein-independent signaling pathways. This was first shown for the β2-adrenergic receptor, where β-arrestin 1 is critical for Src-mediated

activation of ERK1/2 signaling (18). Arrestin-dependent (i.e., G-protein-

independent) signaling has now been shown in numerous other instances,

including additional instances for extracellular signal-related kinase (ERK1/2),

Jun N-terminal kinase (JNK), phosphoprotein of 38 kDa (p38), AKT8 virus

oncogene cellular homolog (Akt), and phosphatidylinositol 3-kinase (PI3K) (17).

Arrestin-dependent signaling in vivo has also recently been shown for 5-

hydroxytryptamine (5-HT), but not 2,5-dimethoxy-4-iodoamphetamine (DOI;

IUPAC name: 1-(4-iodo-2,5-dimethoxyphenyl)propan-2-amine), at the 5-HT2A receptor (19).

1.4 – Models of GPCR Behavior

1.4.1 – Classic GPCR Models

The dogma in GPCR pharmacology for many decades has revolved around the concept of efficacy first elucidated in 1956 by Stephenson (20) and then modified by Furchgott in 1966 (21), (223). As stated by Stephenson, efficacy refers to the property that “different drugs may have varying capacities to initiate a response and consequently occupy different proportions of the receptors when producing equal responses.” Stephenson’s then revolutionary conceptualization allowed for the incorporation of partial agonists into receptor theory, which he recognized and illustrated (20). Furchgott modified Stephenson’s definition of efficacy to be tissue-independent, calling it intrinsic efficacy, as opposed to Stephenson’s constant efficacy (21), (223).

GPCR theory has evolved considerably since. Many of the subsequent developments were precipitated by unexpected or anomalous experimental data which could not be accounted for by existing paradigms. For example, to accommodate the experimental observations that agonists recognize two

6 receptor states with different affinities (the high and low affinity sites), and that

GTP analogs abolish the high affinity site, the ternary complex model was developed and shown to fit GPCR competition binding data better than other theoretical alternatives (22). Another important conceptual leap took place when

Costa and Herz (23) showed experimentally that drugs such as [N,N’-diallyl-Tyr1,

4 5 Ψ (CH2S)Phe ,Leu ]enkephalin (ICI 154129) and [N,N’-diallyl-

Tyr1,Aib2,3,Leu5]enkephalin (ICI 174864) inhibited constitutive opioid receptor activity, as would be predicted for a compound that preferred the low affinity, inactive state of the receptor. In the same study, MR 2266 had no inhibitory effect in the same system, indicating the presence of two classes of

“antagonizing” drugs, inverse agonists (or antagonists with negative intrinsic activity) and neutral antagonists, the former preferring the low affinity, inactive state, and the latter having no preference for either receptor state (23). To accommodate the phenomenon of constitutive activity, Samama et al. proposed the extended ternary complex (ETC) model (24). Finally, the cubic ternary complex (CTC) model was recently proposed as being more technically complete than the ETC model since it accounts for the interaction of receptors in the inactive state with G proteins (21), (25).

1.4.2 - Functional Selectivity

The implicit assumption in all of the aforementioned GPCR models is that a ligand induces or selects only one receptor active state (though different ligands

correspond to unique active states). A corollary of that assumption is that

“intrinsic efficacy,” or the “stimulus per receptor molecule produced by a ligand,”

is indeed a system-independent parameter, just as Furchgott originally conceived

(26). Thus, different drugs were thought to vary only in the quantity of stimulation that they produce, and not the “texture” of downstream signaling, and the critical role of the cellular context in determining GPCR signaling texture was either ignored or underappreciated (27). More specifically, the notion of intrinsic efficacy predicts that two drugs acting at the same receptor will never be different in their relative order of potency with respect to stimulating downstream signaling

pathways, only in the strength with which they are able to stimulate those

downstream pathways (28). Another related prediction would be that a drug

cannot be an agonist for one pathway and an antagonist for another.

It has also become increasingly clear in the last few decades that the cellular and

even subcellular environment can affect receptors in unexpected ways. This was

recognized for serotonergic signal transduction some time ago (29). As an

example, 5-HT1A agonists could stimulate or inhibit adenylyl cyclase, depending

on the tissue or cell preparation used (29). Such “environmental” effects have

also been shown with respect to the 5-HT2A receptor. Expression of a constitutively active arrestin, Arr2-R169E, leads to a decrease in 5-HT-mediated maximal inositol phosphate (IP) accumulation, an increase in agonist-dependent

internalization, and a shift in the receptor equilibrium favoring the high affinity

state for DOI (30). Another early illustration that the predictions of the classic

GPCR models were suspect was at the pituitary adenylyl-cyclase activating

peptide receptor (PACAP-R), where one form of the peptide ligand for this

receptor, PACAP-27, is more potent at stimulating cyclic adenosine

monophosphate (cAMP) production than PACAP-38, whereas vice versa is true

when inositol phosphate production is examined (31). Similar findings were made at the 5-HT2C receptor using bufotenin, lysergic acid diethylamide (LSD),

quipazine, m-trifluoromethylphenylpiperazine (TFMPP), and 2-(4-chloro-2-

methylphenoxy)propanoic acid (mCPP) while measuring arachidonic acid (AA)

release, IP accumulation, AA desensitization, and IP desensitization (26), (32).

Finally, the Mailman laboratory found that dihydrexidine and a dopamine D2 receptor-selective analog are 100% efficacious at inhibiting forskolin-stimulated cAMP production via adenylyl cyclase (AC), but cannot inhibit nigral neuron firing, which is also D2-mediated (33), (34). Furthermore, dihydrexidine and the

aforementioned dopamine D2 receptor-selective analog antagonize apomorphine

inhibition of nigral neuron firing (33). Thus, it has become evident that the established framework for conceptualizing how drugs activate receptors and signal downstream through multiple pathways is inadequate (34).

A number of researchers have attempted to modify the framework underlying receptor theory to accommodate the aforementioned phenomena, and many of the ideas generated fall under the rubric of “functional selectivity” (26), (34).

There are three key aspects of the concept of functional selectivity that should be noted. First, the idea that there are two states in which a receptor may exist,

9 active and inactive, though a useful approximation in modeling GPCR behavior

under many circumstances (for example, when measuring only one functional

readout), is incomplete. It is now thought that there are multiple active state

conformations induced by one ligand, and receptor populations exist in what are

essentially ‘conformational ensembles’ (27). Second, different ligands can

induce or stabilize different ensembles of active conformations (26). In a similar

manner, the network of interacting proteins that form unique local environments

depending on the cellular and even subcellular context likely play an important

role in determining signal texture by influencing the repertoire of available

conformational ensembles and/or the localization and regulation of important

effector proteins. The result can be different levels of constitutive activity,

differential downstream signaling, and/or altered responses to ligand binding, all

depending on the identity of the receptor and the particulars of that local

environment. Third, these different active conformations correspond to the

different signaling pathways downstream of a receptor. Thus, it would be

possible that a ligand selectively activates only a subset of downstream signaling

pathways by inducing or stabilizing the appropriate active states.

1.5 – 5-HT2A Receptors

1.5.1 – 5-HT2A Receptor Neuroanatomy

5-HT2A receptors have been reported in a number of neuroanatomical locations

in the brain. 5-HT2A receptors are most heavily expressed in the apical dendrites

10 and soma of pyramidal neurons in layers II, III, V, and VI (35), (36) and in GABA-

ergic interneurons. Furthermore, there is evidence that they are expressed in a

number of other regions, including nucleus accumbens, claustrum, caudate, and

olfactory tubercle (37), (38). This work is primarily focused on cortical 5-HT2A receptors, which likely mediate the more interesting effects of 5-HT2A-receptor activation and blockade, as will become clear in the coming chapters. It should be noted that the regulation and precise subcellular location of this subpopulation of 5-HT2A receptors, along with the key relevant signaling mechanisms, are

poorly characterized despite over two decades of intensive research.

1.5.2 – Hallucinogens

1.5.2.1 – Classical Hallucinogens are 5-HT2A agonists

Known hallucinogens include LSD-like hallucinogens – for example, mescaline,

LSD, psilocybin, and N, N’-dimethyltryptamine (39) - and non-LSD-like hallucinogens such as salvinorin A (40). Hypotheses concerning the mechanism of action of classical LSD-like hallucinogens have evolved considerably over the years, from antagonism at 5-HT sites (41), (42); to agonism at 5-HT sites (43),

(44); to antagonism of raphe neuron firing (45), (46); to the present consensus, which is that the primary action of hallucinogens takes place via activation of 5-

HT2A receptors (47-50). LSD-like hallucinogens produce extremely powerful

somatic, perceptual, and psychic symptoms at doses that are not toxic in

mammals. Furthermore, LSD-like drugs have minimal abuse liability, with their

11 main danger being the potential for hallucinogens to precipitate psychosis in

certain individuals. In fact, the effects of hallucinogens have previously been

described as leading to altered states of consciousness (ASCs) (39). As a result

of their unique ability to produce ASCs, understanding the signaling mechanisms

that underlie hallucinogen action may provide significant insight into the

biochemical mediators of consciousness, about which very little is known. What is known about hallucinogenic signaling, antipsychotic action at 5-HT2A receptors,

and the neural correlates of consciousness will be discussed in the following

sections.

1.5.2.2 – Signaling via the 5-HT2A Receptor (Figure 1.1)

The canonical signaling pathway of the 5-HT2A receptor is via Gαq, which leads

to PLCβ activation (51). Subsequently, PLCβ hydrolyzes phosphatidylinositol

4’,5’-bisphosphate (PIP2), which is present in the membrane, at the sn-3 position, resulting in the production of inositol-1,4,5-triphosphate (IP3) and diacyl glycerol

2+ (DAG). IP3 in turn binds to its cognate receptors, resulting in Ca release from

intracellular stores. Meanwhile, DAG can also activate protein PKC (39), which provides negative feedback on phosphatidyl inositol (PI) hydrolysis (52), (53).

Early research studies were undertaken under the assumption that the aforementioned canonical signaling pathway was responsible for the effects of hallucinogens, but there are a number of issues with such a hypothesis, and they will be discussed in the coming sections.

Figure 1.1 – 5-HT2A Signaling Pathways*

* (all references used to make this figure are contained in 1.5.2.2) 13

First of all, despite its remarkable potency as an in vivo hallucinogen, LSD is unremarkable in terms of its efficacy in stimulating PI hydrolysis, which amounts to less than a third of the maximal stimulation that is possible with 5-HT treatment

(54). Furthermore, the non-hallucinogenic agonist lisuride exhibits very similar activity through the 5-HT2A receptor when PI hydrolysis is measured (54).

Other studies showed that there was no correlation between agonist efficacy and

the ability of an agonist to promote formation of the high affinity state of the

receptor, which is inconsistent with a simple ternary complex model (55). 5-HT2A agonists, both hallucinogenic and non-hallucinogenic, have also been shown to stimulate phospholipase A2 (PLA2), leading to the production of the second

messenger AA (32), (56). Furthermore, the rank orders of efficacy of the

different compounds tested varied between the IP and AA pathways (Table 1.2),

which conflicts with the predictions of the classical GPCR paradigms described

earlier (32).

Table 1.2 – Rank Order Efficacies of Five 5-HT2A Agonists in the IP and AA

Signaling Pathways (32)

Signaling Pathway Rank Order of Efficacies IP LSD < DOI < bufotenin < quipazine < TFMPP

AA LSD < quipazine = TFMPP < DOI < bufotenin

There is evidence that the 5-HT2A receptor couples to multiple G proteins,

including to Gαi/o, through which it activates Rous sarcoma oncogene cellular

homolog (Src) via liberation of Gβγ subunits (57). Src inhibitors also inhibit what

may be an immediate early gene marker of hallucinogen action, early growth

response 2 (egr-2) (discussed in more detail shortly), suggesting that Gαi/o coupling may be important for mediating hallucinogenic effects (57). Inhibitor studies have implicated a number of pathways in activating PLA2, including Gαi/o,

Gα12/13, ERK, Rho, and p38 (58). Whether PLA2 activation has anything to do

with hallucinogenic signaling is unknown, though it appears unlikely since LSD’s potency and efficacy in stimulating AA release are unremarkable, and the non- hallucinogenic compound lisuride has similar potency and efficacy to LSD (39).

5-HT2A signaling also activates Akt (59), and other pathways leading to ERK1/2 phosphorylation are Gαq-PKC-Raf-dependent (60) or PKC-independent and

Ca2+/calmodulin/tyrosine kinase-dependent (61), depending on the cellular system. There are other putative cascades that activate ERK1/2, one involving

epidermal growth factor receptor (EGFR) transactivation (62) and the other

involving PKC and free radicals (63). Recent evidence suggests that 5-HT2A activation can also couple to phospholipase D (PLD) via NPxxY motif-mediated direct interactions with adenosine 5’-diphosphate(ADP)-ribosylation factor 1

(ARF1, a GTPase) and ras homolog gene family, member A (RhoA, a monomeric G protein), but the details and significance of this pathway with respect to hallucinogen action have not been studied (39), (64), (65). Finally, 5-

HT2A receptors have been shown to increase cGMP in an indirect manner that is

15 dependent on glutamatergic neurotransmission (66), suggesting an interplay between the serotonergic and glutamatergic systems (about which more will be said later).

It has been shown that the head twitch response (HTR) in mice is a reliable behavioral correlate of hallucinogenic activity, as non-hallucinogenic 5-HT2A agonists such as lisuride do not induce HTR (57). In the same study, it was shown unequivocally by genetic deletion that the 5-HT2A receptor is the only

receptor involved in mediating head twitch behavior, and thus, hallucinogenic

signaling; furthermore, some evidence suggests that the signaling is via non-

canonical pathways (57). Finally, genetic rescue of the 5-HT2A receptor

expression confined to cortical regions is sufficient to rescue HTR (57). Indeed,

earlier studies had shown that in Gαq knockout mice, DOI-induced head twitch is

only reduced by approximately 30%, providing further evidence that largely non-

canonical G-protein pathways underlie hallucinogenic signaling (67). In contrast,

the 5-HT2A-mediated anxiolytic effects of DOI are completely abolished in Gαq knockout mice, suggesting that canonical Gαq signaling is absolutely critical for

5-HT2A-mediated anxiolysis (67). The aforementioned data is difficult to reconcile

with classical models of GPCR function, but highly consistent with the theory of

functional selectivity, which predicts that different pathways (some non-

canonical) may underlie different behaviors mediated through the same receptor.

A critical litmus test for any putative hallucinogenic signaling cascade is that it

should be activated preferentially by hallucinogens. Towards this end, some

efforts have been made to identify markers of hallucinogen action. One early

study identified transcripts induced 90 minutes after LSD treatment in rats: ania3,

arc, c-fos, Iκβ a, egr-2, NOR1, and sgk (68). A critical shortcoming of this study

is that LSD was only compared to saline, and not to a non-hallucinogenic 5-HT2A agonist such as quipazine or lisuride. A later, more comprehensive study by a different laboratory addressed this shortcoming, comparing genes induced by

LSD, DOI, and lisuride in 5-HT2A wildtype and knockout mice (69). The

researchers identified a number of transcripts modulated by both hallucinogenic

and non-hallucinogenic agonists, including c-fos and Iκβ a (which were identified

in the earlier study). They also identified three transcripts that were induced by both hallucinogens but not lisuride: egr-1, egr-2, and per1. egr-2 was also identified as being induced by the earlier study. It should be noted that egr-2, as a transcription factor, is unlikely to be directly responsible for the effects of hallucinogens. It is possible, however, that egr-2 is a marker for upstream signaling events that are important for mediating the effects of hallucinogens on consciousness.

Efforts have recently been made to identify some of these upstream events (57).

These studies confirmed that egr-2 was indeed induced selectively by hallucinogens. Furthermore, the non-hallucinogenic 5-HT2A agonist lisuride did

not induce egr-2. Pertussis toxin and PP2 (a src inhibitor) inhibited the egr-2 17 response, suggesting Gαi/o signaling via src as a candidate hallucinogenic

signaling pathway. Oddly, the PLCβ inhibitor U73122 was also able to inhibit

induction of egr-2, which is difficult to reconcile with previous data. First of all, as

mentioned previously, experiments in Gαq knockout mice show only a small

reduction in hallucinogen-induced head twitch in the absence of Gαq-mediated

signaling. This could be potentially be explained by the fact that Gβγ can also

activate PLCβ (70), (71). Still problematic, however, is the fact that

hallucinogenic and non-hallucinogenic 5-HT2A agonists have both been shown to activate PLC at similar potencies (39). Thus, given the PLCβ inhibitor data, which clearly suggests that egr-2 expression is dependent on PLCβ, it is not clear why lisuride, with its high potency in that pathway, does not induce egr-2 expression. One possibility might be that Gβγ-mediated PLCβ activation and

Gαq-mediated PLCβ activation are somehow not equivalent, but it is difficult to

conceive of how this might occur.

Taken together, all the evidence suggests strongly that non-canonical GPCR

signaling pathways are likely to underlie hallucinogen action. Candidate

signaling pathways and markers have been identified (i.e., egr-2), but the data

provide a murky story and very few findings have been independently replicated.

Nonetheless, it has become quite clear that hallucinogenic 5-HT2A agonists, in contrast to non-hallucinogenic 5-HT2A agonists, are likely to be functionally

selective at the 5-HT2A receptor. In other words, hallucinogenic agonists are

18 stabilizing or inducing some set of conformations that does not completely overlap with that induced or stabilized by non-hallucinogenic agonists. Thus,

there is some subset of conformations that all hallucinogenic agonists have in

common that is not seen upon non-hallucinogenic agonist exposure. This conformational subset corresponds to some signaling pathway or pathways that are unique to hallucinogens and responsible for their effects on consciousness.

Finally, hallucinogenic signaling events may also be dependent on the local environment of the receptor, the details of which will be discussed in the coming

sections.

1.5.2.3 – The Neuronal Correlates of Consciousness

The previous section summarized the important data regarding which 5-HT2A signaling pathways underlie hallucinogen action. Implicit in much of the hallucinogen signal transduction research is the widely held assumption that whatever hallucinogenic signaling is taking place via the 5-HT2A receptor must, in

the end, modify synaptic activity (and therefore general electrical activity

patterns) in the brain. This assumption is plausible, given the predominant

theories concerning the neural correlates underlying consciousness. It is thought

that consciousnesses is characterized by a “dynamic core” of neurons that

operate as a distributed, highly integrated, and highly differentiated functional

cluster (72). The rapid integration of widely distributed neuronal groups, which is thought to be a critical component of conscious experience, is achieved via the

19 process of reentry (72). Reentry is the continuous, recurrent, and highly parallel communication between distributed groups of neurons that “binds” them together into a dynamic core, sharing minimal information with neurons that are not part of

the functional cluster (72). Reentry is thought to be accomplished at least in part

by corticocortical connections at apical dendrites (73). Furthermore, layer V pyramidal neurons, in whose apical dendrites 5-HT2A receptors are most heavily expressed, might play a central role in the re-entrant apical dendritic activity that underlies consciousness (73). It has been suggested that disorders of consciousness such as schizophrenia may be characterized by dynamic core abnormalities or even the formation of multiple dynamic cores (72). Regardless of the details of the neuronal basis for consciousness, there is strong evidence that apical dendritic activity plays a major role. Thus, 5-HT2A receptors located in the apical dendrites of layer V pyramidal neurons are uniquely positioned to modulate conscious experience.

1.5.2.4 – The Presynaptic, Thalamocortical Hypothesis of Hallucinogen

Action

Significant effort has been dedicated to discovering how 5-HT2A receptors

modulate cortical neuronal activity. Much of this data is derived from isolated

prefrontal cortex (PFC) slices, which eliminate many afferent inputs to the region,

including connections between cortex and thalamus (39). 5-HT has been shown

to increase the amplitude and frequency of excitatory postsynaptic currents

(EPSCs) in PFC (74). Since increases in the frequency of synaptic currents or potentials are traditionally considered as evidence for presynaptic modulation, it has been proposed that activation of 5-HT2A increases glutamate release onto

pyramidal neurons via a presynaptic mechanism (74). It has also been shown

that LSD and other hallucinogenic drugs promote a late, asynchronous phase of

neurotransmitter release during electrically evoked excitatory postsynaptic potentials (EPSPs) (74). In contrast, 5-HT does not promote this asynchronous component of neurotransmitter release (74).

The aforementioned data led to the hypothesis that hallucinogens modulate glutamate release from thalamic afferents synapsing onto apical dendrites of cortical pyramidal neurons, particularly those in layer V. In support of this hypothesis, fiber-sparing thalamic lesions decreased the frequency of 5-HT induced EPSCs (75). Another study showed that lesions of the ventrobasal thalamus inhibited DOI-induced Fos expression in cortex, implicating a presynaptic, thalamocortical mechanism for DOI’s effects on Fos, presumably by increasing thalamic afferent glutamate release at apical dendritic synapses (76).

Finally, thalamic lesions are known in some instances to lead to complex visual hallucinations in humans, further implicating thalamocortical afferents in at least some types of hallucinations (77), (78). There are two hypotheses concerning the mechanism whereby hallucinogens increase thalamic glutamate release.

The first is that presynaptic 5-HT2A receptors, which may form a small percentage

of 5-HT2A receptors in cortical regions (36), are mediating glutamate release

21 directly (76). The second is that an as yet unidentified retrograde messenger relays the signal from activated postsynaptic 5-HT2A receptors to presynaptic

thalamic terminals (79).

It should be noted that there are a number of major issues with a number of

these studies. First, many of them were performed using 5-HT and a selective

antagonist (usually MDL100907, a 5-HT2A-selective antagonist) to implicate 5-

HT2A receptors. The actions of 5-HT at 5-HT2A receptors, however, are unlikely

to contribute to hallucinogen-related signaling pathways for the simple reasons

that 5-HT is non-hallucinogenic and different classes of hallucinogens have

different effects on 5-HT release depending on their pharmacological profile (74).

As mentioned earlier, any 5-HT2A-mediated effect that is relevant to hallucinogen

action should be induced only by hallucinogens. The selective enhancement of

late, asynchronous neurotransmitter release during electrically evoked EPSPs by

hallucinogens, and not 5-HT, represents a potential candidate. Unfortunately,

those studies were not performed with non-hallucinogenic 5-HT2A agonists. Also,

much of this data was generated in isolated prefrontal cortex (PFC) slices, which,

as mentioned earlier, eliminate many afferent inputs to the region, such as connections between cortex and thalamus (39). Thus, inferences concerning presynaptic action, especially at thalamocortical synapses, are difficult to justify.

Finally, the paucity of in vivo studies that have been published use signaling

readouts (frequency of EPSCs, Fos expression) that may or not be related to

hallucinogen action. The in vivo studies suggest that there may be a presynaptic

22 mechanism for mediating, for example, an increase in EPSC frequency or Fos

expression after hallucinogen treatment, but it cannot be inferred that either of

these changes has anything to do with hallucinogen action since none of these

studies used non-hallucinogenic 5-HT2A agonist controls. In fact, what little

evidence there is suggests otherwise, at least for Fos expression, which is

increased by both hallucinogenic and non-hallucinogenic 5-HT2A agonists (69).

Nonetheless, a number of researchers favor a presynaptic, thalamocortical mechanism of action for hallucinogens.

1.5.2.5 – The Postsynaptic, Corticocortical Hypothesis of Hallucinogen

Action

In part because of the paucity of evidence, there have been a number of recent

efforts to experimentally test the presynaptic, thalamocortical hypothesis. One of

the more important studies, which has already been mentioned, showed that

expressing 5-HT2A receptors solely in cortex is sufficient to achieve full rescue of

hallucinogen-induced head twitch behavior – strong evidence that postsynaptic

5-HT2A receptors, as opposed to 5-HT2A receptors on thalamocortical afferents,

mediate hallucinogenic signaling (57). Nonetheless, these studies do not rule out

the possibility that postsynaptic 5-HT2A receptors are initiating a retrograde signal

to thalamic afferents, inducing glutamate release. Another set of in vivo studies

showed that DOI altered cortical neuronal firing rates, resulting in a reduction in

low frequency oscillations and neuronal synchrony which was not affected by

23 extensive thalamic lesions and was reversed by antipsychotics and 5-HT2A antagonists, suggesting an intracortical origin for DOI’s actions on electrical activity that does not involve the thalamus (80), (81). Furthermore, complex visual hallucinations rarely result from thalamic lesions, even more rarely occur without insight (82), and are more common in diseases such Alzheimer’s

Disease (AD) and Parkinson’s Disease Dementia (PDD), which are characterized by a primarily cortical pathology (83).

Figure 1.2 – The Presynaptic and Postsynaptic Hypotheses of Hallucinogen

Action

The conclusions of the aforementioned studies are strengthened by the fact that they were conducted in vivo using hallucinogenic compounds. Together, the

data suggest that alterations in intra-cortical activity are sufficient to cause some

types of hallucinations, and cortical postsynaptic 5-HT2A receptors are sufficient

to mediate the effects of hallucinogens without the involvement of thalamic

afferents (see Figure 1.2 on previous page for comparison of hypotheses).

1.5.2.6 – Hallucinogen Action – From Neurochemistry to Altered States of

Consciousness (ASCs)

Despite decades of research on hallucinogenic drugs, many questions remain

concerning how hallucinogenic activation of 5-HT2A receptors leads to ASCs.

Nonetheless, a handful of broad conclusions can be drawn from the large body of

data that is available. First, the evidence is strong that hallucinogenic modulation

of synaptic activity (and therefore consciousness) is mediated by non-canonical

GPCR signaling pathways in a functionally selective manner. Second, multiple

lines of inquiry suggest that cortical, apical dendritic, postsynaptic 5-HT2A receptors mediate the more interesting effects of hallucinogens without the involvement of thalamic afferents. Furthermore, work on the electrical basis of consciousness is highly consistent with such a localization for the 5-HT2A subpopulation that mediates hallucinogenic signaling events. Finally, since it is thought that some definable subset of glutamatergic signaling activity underlies consciousness, and 5-HT2A receptor signaling is known to affect cortical

25 glutamatergic activity, it is likely that the relevant distal actions of hallucinogenic

signaling pathways alter consciousness by affecting glutamatergic activity.

As the data thus far described make clear, the biochemical basis for the

hallucinogenic action is still largely unknown, with important questions remaining.

More specifically, what are the putative non-canonical signaling events specific to hallucinogenic agonist activation at 5-HT2A receptors that are responsible for the

effects on consciousness? What is the precise localization of the subpopulation

of 5-HT2A receptors responsible for mediating hallucinogenic effects, and what are the mechanisms whereby this subpopulation of receptors is targeted and regulated at that location? And, finally, how does the 5-HT2A receptor interact

with the glutamatergic system?

Recent work has begun shedding light on these questions. Studies have shown

that the 5-HT2A receptor contains a PDZ ligand motif responsible for mediating interactions with PDZ domain-containing proteins including postsynaptic density

protein of 95kDa (PSD-95) (84), (85). PSD-95/SAP90/DLG4 is perhaps the best

characterized mammalian PDZ domain-containing protein (86).

PSD95/dlg/zonular occludens-1 (PDZ) domains function to mediate protein-

protein interactions (87). The most common manner in which this takes place is

for the C-terminus of some protein to serve as a PDZ-ligand to associate

specifically within a binding cleft of the PDZ domain (87), which is composed of

26 six β-sheets (βA - βF) and two α-helices (αA and αB) (88). Based on sequences in C-terminal PDZ ligands, a scheme has been proposed in which there are two main PDZ domain types: types I and II (89). The type I consensus sequence is

S/T – X - Φ, where X is any amino acid and Φ is a hydrophobic amino, most often

V, I, or L (89). The 5-HT2A, 5-HT2B, and 5-HT2C receptors contain the type I motif

(84), (85), (90), and their C-terminal sequences are shown below:

Table 1.3 – C-terminal Sequences of the 5-HT2 Receptors

Receptor C-terminal sequence 5-HT2A KDNSDGVNEKVSCV

5-HT2B ENEGDKTEEQVSYV

5-HT2C VNPSSVVSERISSV

PSD-95 is a member of a family of four primarily CNS PDZ proteins called

membrane-associated guanylate kinases (MAGUKs) which also includes PSD-

93/chapsyn-110, SAP97, and SAP102 (86). They are characterized by 3 N-

terminal PDZ domains and an Src homology 3 (SH3) and guanylate kinase (GK)

domain C-terminal to the PDZ domains (86). Evidence suggests that PDZ

domain-mediated scaffolding by PSD-95 and related family members plays an

important role in scaffolding the postsynaptic density (PSD), a subcellular

specialization that appears to be involved in regulating glutamatergic (excitatory)

signal transduction (91). As would be expected, PSD-95 is clustered in the PSD,

and PDZ1-2, N-terminal palmitoylation, and a portion of the C-terminus are

27 required for proper clustering (92). PSD-95 interacts with a large number of

proteins, and a nearly comprehensive list of known partners is shown below in

Figure 1.3.

Figure 1.3 – PSD-95-interacting Proteins*

*The information for this figure is largely adapted from (86) and (93), with the rest of the information derived from the following: (94-98)

In most cases a protein either interacts with just the PDZ3 domain, as in the case of CRIPT (a microtubule-associated protein) (99), or with PDZ1 and PDZ2, as in the cases of the NR2A/2B/2C/2D N-methyl-D-aspartate receptor (NMDAR) subunits

and 5-HT2A/2C receptors.

Though much remains to be understood, it is widely agreed that PSD-95 is a

regulator of glutamatergic synaptic transmission and plasticity (Figure 1.4) (86).

More specifically, PSD-95 appears to regulate both long-term potentiation (LTP),

which is the conversion of silent synapses to functional ones; and long-term

depression (LTD), which is the silencing of a previously functional synapse. Data

concerning the precise role of PSD-95 in regulating plasticity has proven difficult

to interpret. Evidence suggests that NMDAR-mediated EPSCs are unaffected by

either gain or loss-of-function of PSD-95 (86). AMPAR-mediated EPSCs

increase with overexpression of PSD-95 and decrease with knockdown (86).

Synaptic potentiation induced by PSD-95 overexpression mimics LTP (silent Æ

functional synapse; occluded LTP; enhanced LTD) (86). The opposite is seen

when PSD-95 is examined in vivo, with PSD-95 knockout mice exhibiting

enhanced LTP of the fronto-cortico-accumbal glutamatergic synapses (100).

There is also an augmentation of the acute locomotor effects of cocaine and a

total lack of behavioral sensitization after chronic cocaine treatments (100).

Figure 1.4 – The PSD-95-scaffolded Postsynaptic Glutamatergic Signaling

Complex*, #

*(84), (101-115)

The interaction of 5-HT2A receptors with PSD-95 suggests that they would be

located at the PSD-95-scaffolded PSD. 5-HT2A receptors at the PSD would be

located at apical dendrites, exactly where immunohistochemical visualization has

shown them to be (35), (36). Furthermore, the data is clear in suggesting that

PSD-95 scaffolds a postsynaptic glutamatergic signaling complex. 5-HT2A receptors located at the PSD would be well placed to modulate neural glutamatergic systems, which is the mechanism by which it is widely thought that hallucinogens exert their consciousness-altering effects. Furthermore, important hallucinogenic signaling effectors downstream of the 5-HT2A receptor may be

scaffolded directly or indirectly by PSD-95. As expected, the in vitro data so far suggest that the interaction of 5-HT2A receptors with PSD-95 is critical for their

targeting to the PSD in the apical dendrites of rat cortical neurons (116).

Furthermore, PSD-95 inhibits agonist-mediated internalization of 5-HT2A receptors in HEK293 cells, suggesting that PSD-95 tethers the receptor at the plasma membrane and/or regulates trafficking patterns (84). Together, the data suggest that PSD-95 is likely to be a critical regulator of 5-HT2A function and

downstream hallucinogenic signaling in vivo (Figure 1.5).

Figure 1.5 – Model for 5-HT2A-mediated Hallucinogenic Signaling at the

Postsynaptic Density

1.5.3 – The Neurochemical Basis for Schizophrenia and Psychosis

Schizophrenia is a common mental illness which leads to significant disability in sufferers and incurs enormous societal costs. Schizophrenia is characterized by symptoms in three major domains: positive symptoms like hallucinations and delusions; negative symptoms such as blunted affect; and cognitive symptoms including deficits in attention and working memory (117). The genetic basis remains a mystery despite intensive efforts, though a handful of genes (catechol-

O-methyl transferase (COMT); neuregulin (NRG); dysbindin; disrupted-in- schizophrenia-1 (DISC1)) have been reported to confer susceptibility in multiple studies (118). It has long been thought that multiple genes of modest effect combine to cause schizophrenia in the majority of cases (119). There is some very recent evidence, however, that larger, rare structural variants of recent origin (often de novo mutations) called copy number variants, or CNVs, (i.e., microdeletions and microduplications) may be responsible for much of the disease susceptibility (119-121).

Regardless of the genetic basis of schizophrenia, it is clear from both the genetic studies and the clinical data that it is a highly heterogeneous illness.

Furthermore, the heterogeneity with respect to spectrum and severity of symptoms, treatment response, and the underlying genetics does not preclude a convergence of the genetic heterogeneity of schizophrenia to a more restricted and homogenous neurochemical basis. Accordingly, there are two predominant

33 hypotheses concerning the neurochemical basis for schizophrenia – the dopamine hypothesis, which predominated until recently, and the glutamate

hypothesis, which appears more consistent with most of the scientific evidence.

The dopamine hypothesis arose primarily from two observations: first, that high

doses of amphetamines and other psycho-stimulants can mimic paranoid

schizophrenia (122) and aggravate psychotic symptoms in schizophrenic patients

(123), (124); and second, the affinity of antipsychotic drugs for D2 dopamine

receptors correlates almost perfectly with their therapeutic efficacy (125). In fact, all approved antipsychotics are characterized by D2 blockade (122), (126).

Instances of amphetamine psychosis have been mis-diagnosed and mis-treated

as paranoid schizophrenia for years in some instances (127). Amphetamine

psychosis is characterized by delusions of persecution, auditory and visual

hallucinations, and inconsistent effects on mood, ranging from depression to

euphoria (128). Amphetamine psychosis often takes up to ten days to resolve,

even with treatment, and can precipitate months-long florid psychotic episodes in

schizophrenic individuals (128). Due to the numerous similarities, it was

hypothesized that schizophrenia was characterized by a hyper-dopaminergic

state.

It should be noted some evidence suggests that hyperdopaminergia cannot

account for all the symptoms of schizophrenia. The evidence concerning the

34 ability of psychostimulants to mimic the cognitive and negative symptoms of schizophrenia is conflicting (129), and there are multiple reports suggesting that psychostimulants like amphetamine can actually reverse some of the cognitive

(130) and negative (131) symptoms in schizophrenics without significantly worsening their positive symptoms (132). There is some evidence that chronic,

stable schizophrenics are hypo-responsive to the effects of amphetamine (132).

Furthermore, florid visual hallucinations are more commonly seen in

amphetamine psychosis, in contrast to schizophrenia where auditory

hallucinations predominate (128). In addition, formal thought disorder, a frequent

symptom of schizophrenia, appears to be uncommon or non-existent in

amphetamine psychosis (128).

The glutamatergic hypothesis arose from the observation that phencyclidine

(PCP), an NMDA antagonist, can cause prolonged psychotic reactions in

humans. PCP was originally marketed as Sernyl, an effective anesthetic and

analgesic that did not lead to respiratory and circulatory depression (133).

Reports of prolonged post-operative psychosis (134) led to its withdrawal from

the market and relative obscurity until its re-emergence in the 1970’s as a

popular street drug (221). In some metropolitan areas, the arrival of street PCP

was heralded by a huge increase in psychiatric hospital admissions for prolonged

and severe schizophrenic-like psychosis, at times indistinguishable from florid

schizophrenia (221). It was some years later that PCP was shown to be an

NMDA antagonist (135), (136). It has also been shown that numerous other

NMDA antagonists cause similar psychotic reactions, which in turn led to the formulation of a glutamatergic hypothesis to explain schizophrenia (also known as the NMDA hypofunction hypothesis) (136). The glutamate hypothesis posits that glutamatergic abnormalities underlie schizophrenia, and these abnormalities are mimicked by NMDA antagonists, especially PCP (136). The precise nature

of the glutamatergic abnormalities is unclear, and it has been proposed that an

increase in excitatory neurotransmitters (glutamate and acetylcholine) underlies

schizophrenic psychosis (136).

Regardless of the etiology, PCP psychosis has numerous similarities to

schizophrenia, including wide-ranging delusions, severe paranoia, extreme aggressiveness, formal thought disorder, auditory hallucinations, thought blocking, and blunted affect (132), (221). Psychosis can last hours to weeks in

patients with no previous history of mental illness (more typically hours), and

days to weeks in schizophrenic individuals (often weeks), sometimes

precipitating their illness (132), (136), (221). Even chronic, stabilized schizophrenics appear to be hyper-sensitive to the psychotic effects of PCP – one single dose can induce a weeks-long rekindling of florid psychosis (in contrast to amphetamine, which appears to do so in untreated schizophrenic individuals) (132). In support of the glutamatergic hypothesis, it has been proposed that PCP is unique in its ability to reproduce all the symptoms of schizophrenia – positive, negative and cognitive (132), (136).

Though the glutamatergic hypothesis has become predominant in the last decade or so, there is not necessarily a strong scientific basis for the transition.

There are a number of studies that show that PCP does appear to mimic more of

the symptoms of schizophrenia – negative and cognitive symptoms in particular.

The cautionary note is that there is very little in the way of blinded, controlled

clinical comparisons of amphetamine and PCP psychosis to schizophrenia.

Regardless of which is the better psychosis model, it is clear that both mimic

schizophrenia quite well, and both can precipitate schizophrenic episodes in

predisposed individuals. With the evidence available, it is difficult to say which

hypothesis better describes the neurochemical abnormalities underlying

schizophrenia and psychosis. Given the broad spectrum of schizotypal

disorders, it may very well be the case that in different subpopulations one or the

other system’s dysfunction predominates and/or is causative. In the end, it also

seems likely that abnormalities in both the glutamatergic and dopaminergic

system may underlie schizophrenia.

1.5.4 –Animal Models of Psychosis

A number of animal psychosis models have been developed, in large part to

assess antipsychotic efficacy, and none of them are 100% predictive of efficacy

in humans. Table 1.4 on the next page is a summary of the most commonly

used models (137).

Table 1.4 – Antipsychotic Animal Models*

Animal Model Antipsychotic Action in Typicals Active? Atypicals Active? Model conditioned avoidance suppression of conditioned √ √ response (CAR) avoidance without suppressing escape of unconditioned aversive stimulus apomorphine-induced reverses apomorphine- √ √ climbing behavior induced climbing behavior paw test in rats increase forelimb retraction Equipotent at More potent at time (FRT) and hindlimb prolonging FRT delaying HRT retraction time (HRT) and HRT amphetamine or inhibit hyperlocomotion √ √ apomorphine-induced hyperlocomotion amphetamine-induced reverse amphetamine- - Clozapine and isolation in monkeys induced social isolation quetiapine, possibly others NMDA antagonist- inhibit hyperlocomotion - √ induced hyperlocomotion apomorphine or normalize disruption √ √ amphetamine-induced disruption of prepulse inhibition (PPI)

DOI-induced disruption normalize disruption - √ of PPI isolation rearing- normalize disruption √ √ induced disruption of PPI in rats

NMDA-induced normalize disruption Chlorpromazine in √ disruption of PPI ketamine-induced disruption of PPI *(137)

PCP and related NMDA antagonists, which are used in two of the above models, appear to be unique in their ability to recapitulate the positive and negative symptoms that characterize schizophrenia (129). PCP-induced disruption of

38 prepulse inhibition (PPI) has been reported in both rodents and non-human

primates (138), (139). PPI – the inhibition of stimulus (usually a loud noise)

induced startle by a weaker preceding stimulus (a quieter noise) - is one of the

simplest ways to measure sensorimotor gating across species ranging from

rodents to man (140). Schizophrenics exhibit prominent perceptual gating

abnormalities, for which PPI may be the best surrogate measure (140), (141). As might be expected, it has been shown in numerous studies that schizophrenics exhibit significant PPI deficits (142), (143). PCP-induced disruption of PPI is a particularly attractive schizophrenia/psychosis model because it translates across species and mimics the PPI deficits seen in schizophrenia (138), (139), (144).

Furthermore, atypical antipsychotics such as clozapine preferentially normalize

PCP-induced disruption of PPI in both rodents and monkeys, whereas typical

antipsychotics have little to no effect (138), (145).

There is some evidence that 5-HT2A receptors are important in mediating atypical antipsychotic efficacy and reversing PCP-induced disruption of PPI (146), which is not surprising since clozapine and other atypical antipsychotics are characterized by a high 5-HT2A/2C affinity to D2 affinity ratio (147). For these

reasons, PCP-induced disruption of PPI represents a psychosis model with

arguably the most face validity, since PCP-induced disruption of PPI mimics the

disruption of PPI seen in human psychosis (140). Furthermore, this model

probably has greater construct validity, as atypical antipsychotics have the same

action (normalization of disrupted PPI) on PPI in schizophrenic humans as they

do on PCP-treated rats.

1.5.5 – Atypical Antipsychotics

1.5.5.1 - Atypical Antipsychotics are Potent 5-HT2A Antagonists

Pharmacotherapy of schizophrenia began over 50 years ago with the discovery of chlorpromazine, the first antipsychotic used to treat the disease (117). The

earlier generations of antipsychotics were effective in treating the more salient

symptoms exhibited by schizophrenic patients such as hallucinations and

delusions (222). These first generation antipsychotics are now referred to as

typical antipsychotics and are characterized primarily by high affinity for D2 dopamine receptors and a number of class-specific side effects (222). The class-specific side effects include extrapyramidal symptoms (EPS), which refer to the various movement disorders that can result from typical antipsychotic treatment – tardive dyskinesia, akathisia, and dystonia; elevation of serum prolactin; and neuroleptic malignant syndrome (NLS) (126).

The antipsychotic clozapine was discovered in 1958, but was largely unused for decades because of the occurrence of fatal agranulocytosis in a small percentage of patients (148), (222). A pair of studies in the mid-1970’s showed that clozapine had minimal EPS liability, suggesting that it might be unique (i.e., atypical) (149), (150). Decades after its initial discovery in 1958 (222), clozapine was shown to be characterized by a high affinity for 5-HT2A receptors (151), (152)

and a complex pharmacological profile (126); a lack of EPS symptoms (149),

(150); and an inability to elevate serum prolactin (117). Further subsequent

40 studies also showed that clozapine was particularly useful in treating refractory

schizophrenia (148), reducing suicidality in schizophrenics (153), and alleviating

some negative symptoms (154). Clozapine thus inspired the development of the

next generation of antipsychotics, the atypical antipsychotics - atypical due to their reduced EPS and serum prolactin elevation liabilities (155).

Attempts to develop new atypical antipsychotics resulted in the release of a number of new drugs to the market, including but not limited to risperidone, quetiapine, olanzapine, ziprasidone, zotepine, aripiprazole, amisulpride, and sertindole (126), (155). It should be noted that clozapine still appears to be therapeutically superior to all other antipsychotics, even compared to the newer atypicals, particularly with respect to treatment-resistant schizophrenia (148).

Though atypicality was initially defined by the reduced EPS liability and lack of

serum prolactin elevation that characterized second generation antipsychotics,

there is also a significant positive correlation between atypicality and the 5-

HT2A/D2 affinity ratio of a neuroleptic, suggesting a significant role for 5-HT2A receptors in conferring atypicality and the mediation of antipsychotic effects independent of D2 blockade (147). Not surprisingly, M100907, a 5-HT2A antagonist with negligible affinity for the D2 receptor, is effective in inhibiting

multiple glutamatergic-based psychosis models including PCP-induced disruption

of PPI (156) and PCP-induced hyperlocomotion (157). The fact that atypical

antipsychotics such as clozapine are characterized by high affinity for 5-HT2A receptors and preferentially normalize NMDA antagonist-induced disruption of

PPI suggests an interplay between the serotonergic and glutamatergic systems by way of 5-HT2A receptors. The mechanism by which this interplay takes place

is largely unexplored and unknown. Given the initial evidence that PSD-95

interacts with and regulates 5-HT2A receptors, the PSD-95-scaffolded

glutamatergic signaling complex may be the site of the functional interplay

between the serotonergic and glutamatergic systems.

1.6 – 5-HT2C Receptors

1.6.1 – 5-HT2C Overview and Neuroanatomy

5-HT2C receptors are of interest because they have been implicated in a number

of CNS disease processes and are thus therapeutic targets in some psychiatric

illnesses. They are closely related to 5-HT2A receptors, but have a distinct CNS

localization, with by far the highest levels of 5-HT2C receptors occurring in choroid

plexus, where they were first identified and designated the 5-HT1C receptor (158).

Numerous other brain regions express 5-HT2C receptor mRNA or protein,

including a number of cortical regions: 5-HT2C receptors have been located in

retrosplenial, piriform, entorhinal, frontal, cingulate, and parietal cortex (159-162).

They have also been reported in anterior olfactory nucleus, thalamus,

hypothalamus, amygdala, caudate-putamen (striatum), and CA1-CA3 and the

dentate gyrus of the hippocampus (159-162). Previous studies demonstrated that PSD-95 interacts with 5-HT2C receptors in vitro and in vivo, suggesting the

possibility that the receptors are localized at the postsynaptic density in neurons

(90). Compared to 5-HT2A receptors, less is known about 5-HT2C signaling and

how these receptors may (or may not) influence synaptic function. Some of the

more important points concerning what is known will be reviewed in the following

sections.

1.6.2 – 5-HT2C Receptor Function

1.6.2.1 – 5-HT2C Receptor Signaling

Figure 1.6 – 5-HT2C Signaling Pathways (all references contained in 1.6.2.1)

5-HT2C receptors have been shown to couple to PLCβ (measured by IP

accumulation) (163), and that was confirmed and further shown to take place via

Gαq without any involvement of Gβγ subunits some years later (164). 5-HT2C receptors have been shown to couple to AA as well (32), (56), and ligands exhibit functional selectivity at the 5-HT2C receptor with respect to the IP and AA

pathways, with the rank order efficacies changing depending on which functional

readout is measured (Table 1.5) (32).

Table 1.5 – Rank Order Efficacies of Five 5-HT2C Agonists in the IP and AA

Signaling Pathways (32)

Signaling Pathway Rank Order of Efficacies IP TFMPP = quipazine > bufotenin > DOI > LSD

AA bufotenin = DOI > quipazine = TFMPP > LSD

5-HT2C receptors activate the PLD pathway, both via Gα13 activation of RhoA

GTPase protein and through Gβγ via unidentified downstream mediators (165).

Like the 5-HT2A receptor, the 5-HT2C receptor also has an NPxxY motif, which

suggests that 5-HT2C may also interact directly with RhoA and ARF1 to activate

PLD, though this has not been shown to be the case. The 5-HT2C receptor also

activates the ERK1/2 pathway in a PLD-, PKC-, Raf/MEK-dependent and

tyrosine kinase-, PLC-, PI3K-, and endocytosis-independent manner (166). 5-

HT2C receptors can also couple to Gαi/o (167) and increase cGMP in choroid

plexus (168). Thus far, the evidence indicates that the signaling cascades for the

5-HT2A and 5-HT2C receptors are by and large very similar (compare Figure 1.1,

p. 13 to Figure 1.6, p. 43) – not surprising given their high sequence homology

(166).

1.6.2.2 – RNA Editing of the 5-HT2C Receptor

One way in which the 5-HT2C receptor is unique is the fact that it is the only

GPCR whose pre-mRNA is known to undergo RNA editing (169). RNA editing is performed by an adenosine deaminase that acts on RNA (ADAR) (170). The

enzyme deaminates adenosine to inosine, which is read by the translation

machinery as guanosine (170). RNA editing of the 5-HT2C receptor pre-mRNA

takes place at one of five sites, named A through E (Figure 1.7) (171). RNA

editing results in 32 possible transcripts, with 24 different protein products (170).

RNA editing affects 5-HT2C receptor function in a number of ways. It has been

shown that 5-HT2C-INI is highly constitutively active and that RNA editing silences

that constitutive activity, virtually abolishes the high affinity site, and greatly

reduces the efficiency of G protein-coupling (large increase in EC50) (172). Not

surprisingly, the 5-HT2C-mediated calcium response is shifted to the right in

highly edited isoforms – in other words, agonists have lower potency at edited

2+ isoforms – and the kinetics of the Ca response are altered (173).

Figure 1.7 – RNA Editing of the 5-HT2C Receptor pre-mRNA

Editing also appears to modulate signaling texture downstream of the 5-HT2C receptor – in contrast to 5-HT2C-INI, 5-HT2C-VGV does not appear to couple to Gα13,

or RhoA and PLD, which are downstream of 5-HT2C (174), (175).

1.6.2.3 – 5-HT2C Receptor Modulation of Synaptic Activity and Associated

Behaviors

The role of 5-HT2C receptors in regulating neuronal function is poorly

characterized. Local 5-HT2C antagonism has been shown to increase dopamine

(DA) efflux in the striatum, while systemic administration of mCPP, a 5-HT2C agonist, decreased striatal DA in vivo (176). Some evidence suggests that 5-

HT2C receptors are expressed on GABAergic interneurons of the ventral

tegmental area (VTA), and that activation of these receptors leads to a decrease

in the firing rate of DA neurons of the VTA, probably through GABAergic

inhibition (177), (178). In contrast, 5-HT2C agonists such as mCPP, MK212, and

RO600175 do not appear to lead to increased DA neuron firing in the substantia

nigra pars compacta (179).

Genetic evidence from 5-HT2C knockout mice suggests that long term

potentiation (LTP) is impaired in the dentate gyrus in the absence of 5-HT2C receptors (180). 5-HT2C knockout mice also exhibit defects in behaviors thought

to be mediated by the dentate gyrus, including specific defects in the morris

water maze task and a reduced aversion to novel environments (180).

Furthermore, 5-HT2C knockout mice are more prone than wildtype mice to spontaneous and audiogenic seizures (181), which are known to involve limbic recruitment (182), (183), suggesting a role for the 5-HT2C receptor in regulating

hippocampal neuronal excitability and synchrony. Consistent with the genetic

47 data, 5-HT2C agonists suppress hippocampal theta wave oscillations, which are characteristic of exploratory locomotor behaviors, and desynchronize the septo- hippocampal system, suggesting a role for 5-HT2C receptors in tonically inhibiting

hippocampal excitability (184), (185), which likely has relevance with respect to

limbic seizure susceptibility. 5-HT2C antagonists have the opposite effect,

promoting theta wave oscillations and septo-hippocampal synchrony (184).

Despite the accumulating evidence that 5-HT2C receptors modulate neuronal

function, little is known about how these receptors are targeted and trafficked.

Much of the aforementioned evidence suggests that the receptor is likely to be at

the synapse since activating or blocking 5-HT2C receptors appears to affect

electrical activity. How the 5-HT2C receptors function in a neuronal setting is of particular interest because they have shown promise as targets in the treatment

of a number of psychiatric disorders, particularly schizophrenia and obesity,

though it has been proposed that 5-HT2C receptors may also play a role in the etiology and treatment of OCD and depression as well (117), (186-189).

It has been suggested that 5-HT2C agonists may have potential as antipsychotics

with reduced EPS liability (117). Furthermore, more recent evidence shows that

a selective 5-HT2C agonist, WAY163909, has antipsychotic-like efficacy in a

number of animal models (187). WAY163909 also has efficacy in a number of

other models that have psychiatric relevance, including efficacy in reducing food

48 intake (obesity), anti-OCD-like activity, and anti-depressant-like activity (186),

(188). Lorcaserin, also a 5-HT2C-selective agonist, has shown efficacy in reducing food intake and is in clinical trials as a potential treatment for obesity

(190).

The aforementioned data suggest that the 5-HT2C receptor’s therapeutic potential

is likely related to its ability to influence neuronal excitability, as there is

accumulating evidence that drugs that target the receptor affect neuronal firing

rates and/or correlated activities (e.g., hippocampal synchrony). Thus, 5-HT2C receptors may functionally interact with the glutamatergic system in a PSD-95- dependent manner, thereby influencing neuronal synaptic activity. Identifying the subcellular location of 5-HT2C receptors in different brain regions may help future researchers to elucidate the mechanisms of action of therapies that target this receptor.

CHAPTER 2: Materials and Methods

2.1 – Materials

2.1.1 – Chemicals

Chemical reagents were purchased from Sigma-Aldrich (St. Louis, MO). N- ethoxycarbonyl-1,2-ethoxydihydroquinolone (EEDQ) was purchased from Acros

Organics (Geel, Belgium); 2,5-dimethoxy-4-iodoamphetamine (191) (191), phencyclidine (PCP), clozapine, SB206553 (5-HT2C inverse agonist), and 5-

hydroxytryptamine (5-HT) from Sigma-Aldrich; spiperone from Research

Biochemicals (RBI, Research Biochemicals, Köln, Germany); ritanserin from

Janssen Life Science Products (Beerse, Belgium); [ethylene-3H]-ketanserin (72

or 67 Ci/mmol) from Perkin-Elmer (Waltham, MA); [N6-methyl-3H]-mesulergine

(82 or 75 Ci/mmol) and [3H]-WAY100635 (74 Ci/mmol) from GE Healthcare

(Chalfont St. Giles, United Kingdom).

2.1.2 – Mice

PSD-95 knockout mice were generated by Seth G. Grant (Division of

Neuroscience, University of Edinburgh, Edinburgh EH8 9JZ, UK). All

experiments were approved by the Institutional Animal Care and Use Committee

at Case Western Reserve University or the University of North Carolina, Chapel

Hill. Mice were housed under standard conditions – 12 hour light/dark cycle and

food and water ad libitum.

2.1.3 – cDNA Constructs

Subcloning (vector, primers, etc.) is described in conjunction with the experiments in which each plasmid or plasmids were used.

2.1.4 – Antibodies

The following antibodies and dilutions were used: mouse anti-5-HT2A

(Pharmingen/BD Biosciences, Hamburg, Germany) - 1:500 (sections), 1:1000

(neurons); mouse anti-PSD-95 (Upstate Biotechnology, Lake Placid, NY) –

1:1000; mouse anti-5-HT2C D-12 (Santa Cruz Biotechnology, Santa Cruz, CA) –

1:500; rabbit anti-MAP2 (Chemicon, Temecula, CA) – 1:1000; rabbit anti-GFP

A11122 (Invitrogen, Carlsbad, CA) – 1:1000; rabbit anti-c-fos PC38 (Calbiochem,

San Diego, CA) – 1:1000; Alexa Fluor 488 goat anti-mouse or goat anti-rabbit,

and Alexa Fluor 594 goat anti-mouse or goat anti-rabbit (Invitrogen) – 1:200.

2.2 – Methods

2.2.1 – Immunochemistry

For immunochemistry on brain tissue sections, wildtype and knockout mice were

perfused with 4% paraformaldehyde in 1X PBS and their brains harvested and

placed overnight in 4% paraformaldehyde in 1X PBS at 40 C. Over the next night

brains were placed in 30% sucrose in 1X PBS until they sank, then frozen on dry ice and stored at -800 C. Sections (30 μM) were made on a Thermo Scientific

Richard-Allan MICROM HM 525 cryostat, either free-floating in 1X PBS (one or two sections per well in a 24 well plate) or thaw-mounted onto coated microscope

slides, and then they were permeabilized with 0.3% Triton in 1X PBS for 15-20

minutes. For immunochemistry on cultured cortical neurons, DIV 4 neurons were

washed twice with 1X PBS, fixed in 4% paraformaldehyde in 1X PBS for 30

minutes, then washed twice more with 1X PBS before permeabilizing with 0.3%

Triton in 1X PBS for 15-20 minutes. Blocking was performed using 5% milk in 1X

PBS for 1-2 hours. Primary antibodies were incubated in 5% milk in 1X PBS at

room temperature for 2 hours or overnight at 40 C while shaking. Sections were

then washed 3 times in 1X PBS (10 minutes for each wash). Secondary

antibodies were incubated in 5% milk in 1X PBS at room temperature for 1 hour

in the dark, while shaking. Sections were washed 3 times in 1X PBS (10 minutes

for each wash). Free-floating sections and neuronal cover slips were transferred

to a microscope slide and mounted (KPL Kirkegaard and Perry Laboratories,

Gaithersburg, MD) for fluorescence microscopic visualization.

2.2.2 – Saturation Radioligand Binding

For saturation binding assays, brain regions were microdissected and frozen on

dry ice, then stored at -800 C. A Tissue TearorTM (BioSpec Products, Bartlesville,

OK) was used to homogenize tissue (10 seconds, 15,000 rpm) in 2 ml of

standard binding buffer (SBB - 50 mM TrisHCl, pH 7.4; 10 mM MgCl2; 0.1 mM

EDTA). Homogenized tissue was spun for 10 min at 26,000 x g (40 C), and the

SBB removed. The pellet was resuspended in 1 ml of SBB and transferred to a

1.7 ml eppendorf tube, then spun at top speed in a microcentrifuge for 5 minutes at 40 C. The SBB was removed and the pellet was either used immediately for

binding or stored at -800 C until use. Saturation binding assays were performed

3 3 with the homogenized brain tissue and [ H]-ketanserin (5-HT2A; cortex); [ H]-

3 mesulergine + 100 nM spiperone (5-HT2C, hippocampus); or [ H]-WAY100635 (5-

HT1A, cortex, hippocampus), then incubated in SBB for 1.5 hours. The following

[3H] radioligand concentrations were used: 8 nM, 6 nM, 4 nM, 2 nM, 1.5 nM, 1.0

nM, 0.5 nM, 0.25 nM, all in duplicate for total and nonspecific (4 reactions at each concentration for each brain sample in which receptor binding was measured).

Nonspecific binding was determined by incubating the reactions with 8 μM ritanserin (5-HT2A and 5-HT2C) or 8 μM 5-HT (5-HT1A). For 5-HT2C measurements 3 hippocampal samples were pooled for each assay. Bradford protein assays were performed in order to normalize Bmax determinations to the

amount of protein in each assay. Reactions were harvested by vacuum filtration

through glass fiber filters (3X ice cold 50 mM Tris, pH 7.4; pH 6.9 at room

temperature) and measured by liquid scintillation using a Perkin-Elmer Tri-Carb

2800TR. Microsoft Excel and Graphpad Prism were used for all data analysis.

All saturation binding was analyzed using non-linear least squares fitting.

2.2.3 – EEDQ Time Course

EEDQ (dissolved EEDQ in 100% ethanol, then dilute 1:3 in saline) was injected i.p. at a dose of 10 mg/kg. Mice were sacrificed at 1, 2, 3, 5, 7, and 13 days

post-EEDQ treatment, and the 5-HT2A Bmax was measured by saturation binding.

Receptor synthesis was assumed to be a zero-order process and receptor trafficking a 1st-order process (192). Thus, the equation derived to model

receptor recovery was:

B = B 1 − e −kt [][]max t max ss ( )

where [Bmax]t is the amount of receptor at time t, [Bmax]ss is the steady state Bmax after the receptors have recovered, k is the catalytic rate constant for receptor

-1 turnover (inverse days, or d ), and t is the time at which [Bmax]t was measured in

days. This model was fit by non-linear least squares regression to a plot of the

average Bmax value at each time point.

2.2.4 – Quantitative RT PCR

Trizol (Invitrogen) was used to extract RNA from microdissected cortical tissue.

10 μg of RNA was treated with DNAse (DNA-free, Ambion, Austin, TX), and 2 μg

of the DNase-treated RNA was added to a reverse transcription reaction which

was performed using the SuperscriptTM III RNase H Reverse Transcriptase kit

(Invitrogen) with Oligo-(dT)12-18 primers (Invitrogen). IQ SYBR Green Supermix

(BioRAD, Richmond, CA) was used in conjunction with the 7300 RT PCR System

(Applied Biosystems, Foster City, CA) for quantitation. All steps were performed according to manufacturer’s instructions.

2.2.5 – Microarray Experiment

RNA was extracted from microdissected cortical tissue using Trizol (Invitrogen).

The gene chip assay was performed by the Gene Expression and Genotyping

Core Facility at the Case Comprehensive Cancer Center using the Affymetrix

Genechip® Mouse Genome 430 2.0 Array. Data was analyzed using the

Affymetrix Genechip Operating Software, version 1.4.0.036 according to the manual’s instructions.

2.2.6 – 5-HT2C mRNA Editing

Microdissected hippocampal tissue was pooled by genotype and used to

generate cDNA as described above. The following primers were used to generate a PCR fragment (containing the edited site) that was 327 base pairs in length that was then inserted into the BamHI/EcoRI sites of pcDNA3.

FORWARD PRIMER: 5’ AAA GGATCC TGT GCT ATT TTC AAC TGC GTC CAT

CAT G 3’; REVERSE PRIMER: 5’ AAA GAATTC CGG CGT AGG ACG TAG

ATC GTTAAG 3’ (171). Each bacterial colony resulting from transformation of the ligation product represents a transcript. Clones were miniprepped and sequenced to determine the extent of editing for each transcript.

2.2.7 – Cortical Neuronal Cultures

Cortical neurons were prepared from P0.5 mouse pups as described previously by others (193). Briefly, cortex was microdissected in Mg2+-containing Hank’s buffered salt solution (HBSS) under a dissecting microscope and incubated at

370 C for 20 minutes in neurobasal medium containing 0.1% papain and 0.02%

BSA. The supernatant was removed and the tissue was then mechanically

triturated in neurobasal medium with a glass Pasteur pipette. The supernatant

was transferred to a new sterile eppendorf, leaving the aggregates, and spun

down at 200 x g for 10 minutes. The supernatant was discarded and the pellet

0 resuspended in pre-equilibrated (to 37 C and 5% CO2) neurobasal medium

containing B27 supplement, antibiotics, and 0.5 mM glutamine and plated on

cover slips coated with low molecular weight poly-L-lysine. Immunochemical

experiments were performed at 4-5 DIV.

2.2.8 – Lentiviral Preparation

PSD-95 was cloned into FUGW (194) by ligating a BclI-digested PSD-95 PCR

fragment into the BamHI site 5’ to the GFP (FORWARD PRIMER: 5’ – AAA TGA

TCA ATG GAC TGT CTC TGT ATA GTG ACA ACC – 3’; REVERSE PRIMER: 5’

– AAA TGA TCA GAG TCT CTC TCG GGC TGG GAC CCA – 3’). Site-directed

mutagenesis was performed to mutate away the stop site that results from the

BclI-BamHI ligation at the 3’ end of PSD-95 and shift the reading frame so that

PSD-95 is in frame with GFP (SENSE PRIMER: 5’ – GCC CGA GAG AGA CTC

TTA TTT CCC CCG GGG GTA CCG GT – 3’; ANTISENSE PRIMER: 5’ – ACC

GGT ACC CCC GGG GGA AAT AAG AGT CTC TCT CGG GC – 3’). Fugene6

(50 μL Fugene6, 10 μg total DNA per 10 cm plate) was used to co-transfect

HEK293T cells with 3 plasmids (FUGW/Δ8.9 HIV-1/VSVG) in a ratio of 3.3/2.5/1.

Lentivirus-containing media was collected 48 hours later and filtered through a

0.45 μM filter to remove cellular debris. Lentivirus was aliquoted and frozen at

-800 C until use. Cortical neurons were infected with 20-50 μL GFP or PSD-95

GFP lentivirus at 2 DIV. Immunochemistry was performed at 5 DIV.

2.2.9 – MK-212-induced c-fos in Hippocampus

Mice were injected i.p. with 5 mg/kg MK-212 in 0.9% sterile NaCl or vehicle. 40 minutes later they were perfused with 4% paraformaldehyde. Frozen sections

(Bregma -1.34 mm to Bregma -2.7 mm) were thaw mounted onto frosted slides and then used for immunochemistry and subsequent c-fos quantitation.

2.2.10 – DOI-induced Head Twitch

The head twitch response elicited after administering hallucinogens to mice consists of a characteristic, rapid, rotational flick of the head, ears, and neck that is easily monitored and easily distinguishable from other head movements such as head shakes or head jerks. Mice were injected i.p. with one of three doses of

DOI: 0.3, 1, or 5 mg/kg. The number of head twitches was counted and recorded

57 in 5 minute bins for the half hour period immediately after injection. A subset of

the 5 mg/kg injections (N=7) were counted by two observers, one of whom was

blinded to the genotype. A comparison of the results produced by the two

different observers was not significantly different (data not shown). All the other

head twitch experiments were performed by one blinded observer.

2.2.11 - PPI

All PPI experiments were performed at the Mouse Behavioral Phenotyping

Laboratory Core Facility in the Neurodevelopmental Disorders Research Center

using the SR-Lab (San Diego Instruments). Briefly, mice were placed in a small,

plexiglass cylinder housed within a large sound-proofed chamber. The cylinder

is seated on a piezoelectric transducer which quantifies movement-induced

vibrations. The SR-Lab chamber also contains a light, fan, and loudspeaker for

acoustic stimuli. Calibration of 70 dB background sound levels and prepulse

acoustic stimuli was performed with a digital sound level meter (San Diego

Instruments). Each session consisted of a 5 minute habituation period followed

by 42 trials of 7 types – No Stimulation, 120 dB acoustic stimulus (AS50), and 5

different prepulse stimuli ranging from 4 dB over back ground (PP74) to 20 dB

over background (PP90). The trial types were performed in 6 sets of 7, with the

trial type order in each set randomized. Inter-trial intervals were 10-20 seconds,

with an average interval of 15 seconds. The AS50 was 40 ms long, while the prepulse stimulus was 20 ms long and occurred 100 ms before the onset of the

58 startle stimulus. The sample window for measuring startle amplitude was 65 ms.

The formula used to calculate % PPI was: ((AS50 – Startle After Prepulse)/AS50)

X 100. Mice were injected with vehicle, PCP, or clozapine plus PCP before being placed in the PPI chamber. When treated with vehicle or PCP, mice were immediately placed in the chamber. When treated with clozapine, mice were injected with antipsychotic 15 minutes before injecting PCP, after which mice were immediately placed in the chamber.

CHAPTER 3: PSD-95 Regulates 5-HT2A Receptor Function in vivo

3.1 – Introduction and Rationale

Previous studies demonstrated that PSD-95 interacts with 5-HT2A receptors in

vitro and in vivo (84), (116), (85). Additionally, ectopic expression of PSD-95 augments downstream signaling and inhibits the agonist-mediated internalization of the 5-HT2A receptor in vitro (84). Eliminating the Type I PDZ ligand motif abrogates both PSD-95 binding to the 5-HT2A receptor and the ability of PSD-95

to regulate receptor function in vitro (84), (116). What, if any, effect PSD-95

might have in vivo is unknown. The data suggest that PSD-95 is responsible for

proper synaptic membrane stabilization of 5-HT2A receptors. Thus, we predict

that, in the absence of PSD-95, 5-HT2A receptor expression would be reduced

due to synaptic membrane de-stabilization.

We examined several potential mechanisms which might account for the

hypothesized PSD-95-mediated modulation of 5-HT2A receptor expression.

These included: (1) non-specific serotonergic dysfunction; (2) PSD-95-mediated

regulation of 5-HT receptor transcription and/or widespread transcriptional

dysregulation; and (3) alterations in serotonin receptor turnover. Since we

hypothesized that PSD-95 modulates 5-HT2A expression by stabilizing the

receptor at the plasma membrane we predicted: (1) in the absence of PSD-95,

there is no change in the expression of 5-HT receptors that do not have PDZ

ligand motifs and (2) 5-HT2A gene transcription is unaltered in the absence PSD-

95, and there are few changes in global gene expression. Instead, we expected

that (3) the 5-HT2A receptor turnover rate increases due to receptor trafficking 60 abnormalities in the absence of PSD-95. Confirmation of the aforementioned

predictions would provide strong evidence that PSD-95 stabilizes 5-HT2A and 5-

HT2C receptors at the plasma membrane.

Though it is well known that 5-HT2A receptors are heavily expressed in the apical

dendrites of cortical pyramidal neurons (36), little is known about 5-HT2A targeting

mechanisms. Since PSD-95 interacts with 5-HT2A receptors and is also heavily

expressed at neuronal PSDs, we hypothesized that PSD-95 is critical for 5-HT2A targeting, for which is there is some initial evidence (116). Thus, we predict that

5-HT2A apical dendritic targeting is impaired in the absence of PSD-95, and

addback of PSD-95 rescues targeting.

Finally, evidence suggests that 5-HT2A interactions with PDZ domain-containing proteins play a critical role in mediating receptor signaling (84), (195). Thus, we

predict that, in the absence of PSD-95, 5-HT2A receptor signaling is impaired.

This impairment will be apparent at both the biochemical and behavioral level.

As a result, we predict a deficit in 5-HT2A-mediated behaviors and therapies

targeting the 5-HT2A receptor. More specifically, we predict that the response to

hallucinogens (5-HT2A agonists) is reduced, as is the efficacy of atypical

antipsychotics in glutamatergic psychosis models (5-HT2A antagonism being a

critical feature of atypical antipsychotics). The studies that will be discussed are

aimed at testing the predictions that result from our hypothesis. We hope to

show that PSD-95 regulates the targeting, turnover, and signaling of 5-HT2A

61 receptors, and that this regulation has important behavioral consequences with respect to hallucinogen action and psychosis.

3.2 – Results

3.2.1 – Genetic Deletion of PSD-95 Results in a Selective Loss of 5-HT2A

Receptors

3.2.1.1 – 5-HT2A Immunochemistry

In order to test the prediction that, in the absence of PSD-95, 5-HT2A receptor

expression would be reduced due to synaptic membrane de-stabilization, we

examined 5-HT2A receptor expression in PSD-95 wildtype and knockout mice.

Brains from PSD-95 wildtype and knockout mice were perfused, cryo-protected,

and frozen, and then 30 μM coronal sections were taken through prefrontal and

frontal cortical regions. As mentioned in the introduction, it has been known for

some time that 5-HT2A immunofluorescence is particularly dense in prefrontal and

frontal cortical apical dendrites (36), (35). As seen in Figure 3.1 in a high magnification image, PSD-95 knockout mice exhibit very little apical dendritic immunofluorescence in prefrontal cortex as compared to wildtype litter-mate controls.

Figure 3.1 – 5-HT2A Immunochemistry

5-HT2A and PSD-95 double-label immunochemistry in medial prefrontal

cortex of PSD-95 wildtype and knockout mice shows a large reduction in 5-

HT2A receptor expression in PSD-95 knockout mice (N=3). The red arrow

points to a stained apical dendrite in the wildtype image. Knockouts

exhibit a significant reduction in 5-HT2A immunostaining as compared wildtypes.

3.2.1.2 – Measuring 5-HT2A Receptor Density

The immunochemical microscopy images provided a qualitative indication that 5-

HT2A receptor expression was reduced in the absence of PSD-95. We also

63 performed saturation binding experiments with [3H]-ketanserin (8 μM ritanserin to

determine nonspecific) on microdissected and homogenized cortical tissue to

obtain a quantitative Bmax (total receptor expression) estimate of 5-HT2A levels in knockout mice (Figure 3.2). Consistent with the immunochemical data, quantitation showed a significant, 40% reduction in 5-HT2A expression in the cortices of PSD-95 knockout mice.

Figure 3.2 – 5-HT2A Bmax Measurements in PSD-95 Mice

Comparison of Bmax estimates for the 5-HT2A receptor in PSD-95 wildtype

* and knockout cortices (N=4). Bmax data are presented as means +/- SEM. p

< 0.05; one-tailed, unpaired t-test.

5-HT2A Cortex

3.2.2 – PSD-95 Does Not Modulate Expression of the 5-HT1A Receptor

In order to determine the mechanism by which PSD-95 modulates 5-HT2A expression, we first examined the possibility that genetic deletion of PSD-95 causes generalized serotonergic system dysfunction, leading in turn to a

64 reduction in serotonin receptor levels. We assessed this possibility by measuring

the expression of a related serotonin receptor which is also highly expressed in

cortical neurons but lacks a PDZ-ligand motif - the 5-HT1A receptor. As our

3 saturation binding experiments using [ H]-WAY100635 indicate, 5-HT1A expression levels were unchanged in PSD-95 knockout mice in cortical homogenates (Figure 3.3). These results indicate that genetic deletion of PSD-

95 does not lead to generalized serotonergic system dysfunction.

Figure 3.3 –5-HT1A Bmax Measurements in PSD-95 Mice

Comparison of Bmax estimates for the 5-HT1A receptor in PSD-95 wildtype

and knockout cortices (N=5). Bmax data are presented as means +/- SEM.

One-tailed, unpaired t-test revealed no significant difference between

genotypes. 5-HT1A Cortex

3.2.3 – PSD-95 Does Not Play a Prominent Role in Modulating Gene

Expression

3.2.3.1 – PSD-95 Does Not Alter 5-HT2A Receptor mRNA Levels

To examine the possibility that deleting PSD-95 leads to an alteration in

serotonin receptor gene transcription, we performed quantitative RT-PCR in

order to measure 5-HT2A receptor mRNA levels. Receptor mRNA levels were

normalized to β-actin mRNA levels. We found that normalized 5-HT2A mRNA levels are unchanged in cortex (Figure 3.4). Thus, PSD-95 does not affect 5-

HT2A receptor expression by modulating mRNA levels.

Figure 3.4 – 5-HT2A Receptor mRNA Levels in Cortex

Cortical 5-HT2A receptor mRNA levels normalized to β-actin mRNA levels

measured by quantitative RT-PCR (N=4 animals for each genotype; 11

measurements were performed for each animal). Normalized mRNA

measurements are presented as means +/- SEM. One-tailed, unpaired t-test

revealed no significant difference between genotypes.

5-HT2A mRNA Cortex

3.2.3.2 – Global Gene Expression Does Not Change Substantially in the

Absence of PSD-95

To further assess the role of PSD-95 in modulating mRNA levels more broadly, we performed whole-genome microarray analysis on cDNA prepared from PSD-

95 wildtype and knockout cortices. Overall, there were few differences in transcript levels, and only 28 genes (27 genes decreased, 1 gene increased) appear to be modulated greater than 2-fold in the absence of PSD-95 - none of which are G-protein coupled receptors (GPCRs) or are expected to modulate the expression of 5-HT receptors (Table 3.1).

Table 3.1 – All Genes Affected in PSD-95 Knockout Mice

Gene of Interest Accession # Change In Transcript Levels as % of Wildtype syndecan 4 BC005679.1 15.9 Nudel BC021434.1 33.0 Arc NM_018790.1 36.6 egr2 X06746.1 38.6 Per1 AF022992.1 42 Tieg NM_013692.1 45.1 Arf3 NM_007478.1 46.7 myla NM_010858.1 46.7 edr AJ007909.1 46.7 Jun-B NM_008416.1 46.7 Gng3lg AK013851.1 50 Dnajb5 AF088983.1 50 Nr4a1 NM_010444.1 50 Per2 NM_011066.1 50 Tgtp NM_011579.1 20.3 Homer1a AF093257.1 35.4 Matn4 NM_013592.1 42.1 H2-Q7 M29881.1 42.1 Nptx2 NM_016789.1 45.1 Ptprs D28531.1 45.1 Igtp NM_018738.1 48.3 MHC H-2Dr M34962.1 48.3 Fmnl NM_019679.1 48.3 Igfbp5 NM_010518.1 50 Inhba NM_008380.1 50 Kcnk4 NM_008431.1 50 Dcl AF155820.1 50 Ywhaz BF608615 200

Thus, the whole genome microarray data are more consistent with a role for

PSD-95 in post-transcriptional/post-translational regulation of 5-HT2A receptors.

Interestingly, 6 of the 28 genes, out of approximately 45,000 transcripts on the microarray, have previously been reported to be induced after hallucinogen administration (Table 3.2) (69), (68). In one study of transcripts induced by 5-

HT2A agonists, only 3 of 13 transcripts shown to be changed by agonist

administration were specific to hallucinogenic agonists (69). 2 of these 3 genes,

egr2 and per1, are down-regulated in the absence of PSD-95 according to our

microarray data, which is consistent with a possible role of PSD-95 in mediating

some 5-HT2A signaling pathways, particularly those related to hallucinogen

actions.

Table 3.2 – Genes of Interest Affected in PSD-95 Knockout Mice

Gene of Interest Alternate Names Downregulation as % of Wildtype Arcb rg3.1 36.6 egr2a,b krox20; ngf1b; zfp-25; zfp- 38.6 6 per1a rigui 42.0 Jun-Ba - 46.7 Nr4a1a N-10; gfrp; gfrp1; hbr-1; 50 hmr; np10; tr3; nur77; tis1 Homer1ab,* - 35.3 a(69) b(68) *the gene previously reported to be upregulated after hallucinogen administration is ania3, a closely related isoform that differs only in the 5’ UTR and a few amino acids at the C-terminus

3.2.4 –5-HT2A Receptor Turnover Rate is Accelerated in the Absence of

PSD-95

Our data clearly point to the fourth prediction that PSD-95 is exerting its effect on

5-HT2A receptors by regulating their trafficking/turnover. Implicit in our hypothesis is that, in the absence of PSD-95, 5-HT2A receptors will have greater access to

intracellular trafficking machinery, or will enter alternative trafficking pathways,

leading to higher rates of receptor turnover. To assess the rates of receptor

turnover in PSD-95 wildtype and knockout animals, we took advantage of the

properties of N-ethoxycarbonyl-1,2-ethoxydihydroquinolone (EEDQ), which binds

irreversibly to 5-HT2A receptors, occluding them from recognition by their ligands

after EEDQ treatment. By treating mice with EEDQ and modeling the rate of

receptor recovery over time (see Methods section 2.2.3 for more details about

the mathematical model), one can measure the rate of 5-HT2A receptor turnover

in vivo (Figure 3.5) (192). For these studies we injected mice once with 10 mg/kg of EEDQ, a dose that achieves approximately 90% irreversible blockade of 5-HT-

2A receptors (data not shown), and performed saturation binding experiments at

different time after EEDQ treatment to measure the recovery rate of 5-HT2A receptors. If 5-HT2A receptors in knockout mice have a higher rate of turnover,

then the rate constant of recovery (192) should be higher in these mice.

Figure 3.5 – Modeling in vivo Receptor Turnover Using EEDQ

Consistent with this prediction, the modeled receptor recovery in PSD-95 wildtypes and knockouts (Figure 3.6) showed that the rate constant, k (d-1), was substantially higher in knockout mice (Figure 3.7). These findings indicate that genetic deletion of PSD-95 accelerates 5-HT2A receptor turnover in vivo.

Figure 3.6 – Modeling 5-HT2A Turnover Kinetics

Fitted curves for modeling 5-HT2A receptor turnover kinetics in PSD-95

wildtype and knockout mice, respectively, (N=3-4 at each data point).

Visual inspection shows that steady state levels for the 5-HT2A receptor are

reached sooner in the absence of PSD-95, suggesting accelerated turnover.

Figure 3.7 – 5-HT2A Receptor Turnover Rate Constant Comparison

Comparison of the kinetic rate constant k (d-1) of receptor turnover. Rate

constant, k, is a non-linear least squares fitted parameter of an equation

modeling receptor recovery (see Methods section 2.2.3 for details), +/- SEM.

*p < 0.05; one-tailed, unpaired t-test.

3.2.5 – PSD-95 is Required for Normal Expression and Polarized Sorting of

5-HT2A Receptors to Pyramidal Neuron Apical Dendrites

3.2.5.1 – 5-HT2A Receptor Expression and Dendritic Targeting is Attenuated

in Neurons Prepared From PSD-95 Knockout Mice

Another important aspect of our hypothesis focuses on 5-HT2A receptors and the prediction that PSD-95 is crucial for proper targeting to the apical dendrites.

Previous studies showed that mutating the PDZ ligand motif prevents dendritic

targeting of the 5-HT2A receptor in vitro (116). In order to determine if PSD-95 is one of the PDZ domain proteins responsible for the preferential dendritic targeting of 5-HT2A receptors, we examined the ability of 5-HT2A receptors to be

sorted to neuronal dendrites in cortical neurons prepared from PSD-95 wildtype

and knockout mice.

For these studies, we performed confocal immunofluorescence studies of mouse

cortical neurons for 5-HT2A receptors and the dendritic marker microtubule-

associated protein 2 (MAP2) at 4-5 DIV (196). As Figures 3.8-3.9 illustrate,

neurons prepared from PSD-95 knockout animals exhibit significantly lower 5-

HT2A receptor expression in both the neuronal soma and dendrites - a finding

consistent with our in vivo data. In order to examine the impact of PSD-95 on

dendritic trafficking, we also calculated a 5-HT2A receptor cell body/dendrite

expression (CB/D) ratio. If dendritic targeting is impaired in PSD-95 knockout

neurons, we predicted that the CB/D ratio should be higher in these neurons, as

impairment of 5-HT2A trafficking to dendrites should result in a relative

72 accumulation of receptors in the neuronal cell body. As predicted, Figure 3.10 confirms that the CB/D ratio is higher in PSD-95 knockout neurons.

Figure 3.8 – 5-HT2A and MAP2 Immunochemistry in Cortical Neurons

Representative images of double-label immunochemistry performed on

P0.5 cortical neurons of PSD-95 wildtype and knockout mice. The red

arrows highlight the same dendritic process in all 3 images of each neuron.

5-HT2A MAP2 overlay T W - 5 9 D S P

5-HT2A MAP2 overlay O K - 5 9 D S P

Figure 3.9 – Quantitative Comparison of 5-HT2A Expression in Cultured

Cortical Neurons

Comparison of 5-HT2A receptor expression, normalized to MAP2

expression, in cell bodies and dendrites. N=3 animals for each genotype,

17 neurons from each animal, for a total of 51 neurons measured per

genotype. Data are presented as the mean +/- the SEM. ***p < 0.001; one-

tailed, unpaired t-test.

Cell Body Dendrite

* ** * **

Figure 3.10 – Quantitative Comparison of 5-HT2A Dendritic Targeting in

Cultured Cortical Neurons

A cell body to dendritic (CB/D) expression ratio is compared for PSD-95

wildtype and knockout neurons to examine whether or not there is a 5-HT2A trafficking defect in knockout neurons. The increase in the CB/D ratio in

PSD-95 knockout neurons suggests an impairment in dendritic targeting.

For B, C, and D, N=3 animals for each genotype, 17 neurons from each animal, for a total of 51 neurons measured per genotype. Data are presented as the mean +/- the SEM. ***p < 0.001; one-tailed, unpaired t-test.

CB/D Ratio * **

3.2.5.2 – Lentiviral Addback of PSD-95 to Knockout Cortical Neurons

Rescues Targeting and Expression of 5-HT2A Receptors

If PSD-95 is essential for 5-HT2A expression and sorting to the dendrites,

addback of PSD-95 should increase receptor expression in both the neuronal

soma and dendrites. Furthermore, adding back PSD-95 should decrease the

CB/D ratio, representing an increase in dendritic targeting of receptor. To assess

the effect of PSD-95 addback on 5-HT2A expression and targeting, we generated

PSD-95-GFP lentivirus and a control GFP lentivirus and infected cortical

75 neuronal cultures prepared from PSD-95 knockout animals (Figure 3.11). PSD-

95-GFP expression led to an approximately 2-fold increase in cell body 5-HT2A expression and an approximately 5-fold increase in dendritic 5-HT2A expression

as compared to GFP expression in neurons prepared from the same knockout

animals (Figure 3.12). Furthermore, as predicted, the CB/D ratio is greatly

decreased in knockout neurons expressing PSD-95-GFP as compared to those

expressing the control GFP (Figure 3.13).

Figure 3.11 – 5-HT2A Immunochemistry in PSD-95 Knockout Cortical

Neurons Infected with GFP or PSD-95-GFP Lentivirus

Representative images of double-label immunochemistry performed on

P0.5 cortical neurons of PSD-95 knockout mice infected with either GFP

lentivirus (top two rows of panels) or PSD-95-GFP lentivirus (bottom two

rows of panels). Knockout neurons from each animal were plated in two

wells, one for GFP lentiviral infection and the other for PSD-95-GFP

lentiviral infection. The yellow arrows highlight dendritic 5-HT2A receptor expression in an infected neuron. White arrows highlight 5-HT2A receptor expression in an uninfected neuron. GFP-infected neurons display low overall 5-HT2A expression and low dendritic targeting. In contrast, PSD-95-

GFP-infected neurons display a dramatic increase in overall 5-HT2A receptor expression and substantially more receptor appears to be targeted to the dendritic compartment, both in comparison to control GFP-

76 infected neurons and in comparison to uninfected neurons in the same field.

5-HT2A GFP overlay

5-HT2A PSD95-GFP overlay

Figure 3.12 – Quantitative Comparison of 5-HT2A Expression in GFP- and

PSD-95-GFP- infected PSD-95 Knockout Cortical Neurons

Comparison of 5-HT2A receptor expression in GFP- or PSD-95-GFP- infected

PSD-95 knockout neuronal cell bodies and dendrites. Expression is

normalized to GFP or PSD-95-GFP. N=3 animals for each genotype, and 10

infected neurons from each lentiviral infection were measured (120

neurons total, as each animal was used to produce neurons for infection

with both lentiviruses). Data are presented as the mean +/- the SEM. *p <

0.05, ***p < 0.001; one-tailed, unpaired t-test.

* ***

Figure 3.13 – Quantitative Comparison of 5-HT2A Dendritic Targeting in

GFP- and PSD-95-GFP- infected Cortical Neurons

Comparison of the CB/D ratio in GFP- and PSD-95-GFP- infected neurons in

order to assess 5-HT2A receptor trafficking. N=3 animals for each

genotype, and 10 infected neurons from each lentiviral infection were

measured (120 neurons total, as each animal was used to produce neurons

for infection with both lentiviruses). Data are presented as the mean +/- the

SEM. ***p < 0.001; one-tailed, unpaired t-test.

***

3.2.6 – PSD-95 Mediates Hallucinogen Actions in vivo

We also predicted that the alterations in 5-HT2A expression induced by deleting

PSD-95 should lead to a reduction in hallucinogen actions in vivo. Although a

number of animal models have been proposed for studying hallucinogen action in

rodents (39), head twitch behavior has been shown to be the most specific for

79 hallucinogenic action in that non-hallucinogenic 5-HT2A agonists such as lisuride

do not induce the behavior (57). PSD-95 wildtype and knockout mice were

injected with three different doses of the prototypical 5-HT2A hallucinogen 2,5-

dimethoxy-4-iodoamphetamine (191) (191). We found no difference in the total

number of head twitches over a 30 minute period between wildtype and knockout mice at 0.3 mg/kg or 1 mg/kg DOI (Figure 3.14). The time course of the head twitch behavior was also similar at the two different doses (Figures 3.15 and

3.16). At 5 mg/kg DOI, however, we found that there was a large and significant decrease in DOI-induced head twitch in PSD-95 knockout animals as compared to wildtype mice (Figures 3.14 and 3.17). Such a dosage effect is consistent with a difference in total 5-HT2A receptor expression between PSD-95 wildtypes in

knockouts. In order to provide further evidence that the dosage effect is due to a

reduction in 5-HT2A receptor expression in PSD-95 knockouts, we performed

additional head twitch experiments in five pairs of mice at both 0.3 mg/kg and 5

mg/kg DOI, with a one week recovery period between administrations of the

drug. As predicted, we found a significantly greater increase in head twitch

response in the wildtype mice with the larger second dose (Figure 3.18).

Figure 3.14 – DOI-induced Head Twitch in PSD-95 Mice

Head twitch behavior in PSD-95 wildtype and knockout mice after i.p. injection of one of three different doses of the hallucinogen DOI: 0.3 mg/kg,

1 mg/kg, 5 mg/kg; or saline (N=3). Data are given as means +/- the SEM. *p

< 0.05; one-tailed, unpaired t-test.

Figure 3.15 – 0.3 mg/kg DOI-induced Head Twitch Time Course (N=7). Data are given as means +/- the SEM. *p < 0.05; one-tailed, unpaired t-test.

Figure 3.16 – 1 mg/kg DOI-induced Head Twitch Time Course (N=6). Data

are given as means +/- the SEM.

Figure 3.17 – 5 mg/kg DOI-induced Head Twitch Time Course (N=11). Data

are given as means +/- the SEM. *p < 0.05, **p < 0.01, ***p < 0.001; one-tailed, unpaired t-test.

** * ** ** *** *

Figure 3.18 – The Dose Dependent Increase in Head Twitch is Greater in

PSD-95 Wildtype Mice

Five pairs of mice were injected with DOI at two different doses one week apart. The 5 mg/kg data is expressed as the percent of the number of head twitches seen at the 0.3 mg/kg dose in the same mouse. PSD-95 wildtype

mice exhibit a significantly larger increase in head twitch at the higher dose relative to the lower dose, suggesting a Bmax effect. Data are given as

means +/- the SEM. *p < 0.05; one-tailed, unpaired t-test. *

3.2.7 – Deletion of PSD-95 Renders Clozapine “Propsychotic”

It has been recently demonstrated that synaptic and behavioral measures of

dopamine-mediated synaptic plasticity are also altered by genetic deletion of

PSD-95 (100). We thus hypothesized that the prototypical atypical antipsychotic

83 drug clozapine, whose actions are mediated via inverse agonism at 5-HT2A and

5-HT2C receptors (147), (197) and by weak D2/D3/D4-dopamine antagonism

(198), might have an altered activity in PSD-95 knockout mice. In this regard, the phencyclidine (PCP)-induced disruption of prepulse inhibition (PPI) is a well- accepted pharmacological model of schizophrenia (138), (139). Importantly, clozapine preferentially normalizes PCP-induced disruption of PPI in both rodents and monkeys, while typical antipsychotics like haloperidol are much less potent (138), (145). Given the evidence that 5-HT2A receptors are important in

mediating clozapine’s reversal of PCP-induced disruption of PPI (146), we

predicted that clozapine would exhibit an altered ability to inhibit PCP-induced

disruption of PPI in PSD-95 knockout mice.

In order to test this prediction, we injected littermate pairs of PSD-95 wildtype

and knockout mice with vehicle, PCP, or clozapine plus PCP, followed by PPI assessment. PCP significantly disrupted PPI at all prepulse levels in wildtypes, and at two of the four prepulse levels in PSD-95 knockout mice (Figures 3.20 and

3.21). Clozapine normalized the PCP-induced deficit of PPI in wildtype mice while having no significant effect in PSD-95 knockout mice. Significantly, at 8 dB and 16 dB, clozapine potentiated the PPI-disrupting actions of PCP. As a control, we also examined the raw startle response data and found no significant effect of genotype or treatment on startle response (AS50) (Figure 3.19). Thus, genetic deletion of PSD-95 abolishes clozapine’s antipsychotic actions.

Figure 3.19 – Comparison of Raw Acoustic Startle Responses

Baseline startle response to 50 dB stimulus and no stimulus, along with corresponding startle responses after prepulse stimuli of 4, 8, 12, 16 dB.

There is no significant difference in baseline startle response between

PSD-95 wildtype and knockout mice. Data are given as means +/- the SEM.

There is no significant difference by two-way repeated measures ANOVA

followed by Bonferroni post-tests.

Figure 3.20 – Measuring the Effect of Clozapine on PCP-induced Disruption of PPI in PSD-95 Mice

PPI in PSD-95 wildtype and knockout mice after injection of vehicle, 6 mg/kg PCP, or 0.5 mg/kg clozapine plus 6 mg/kg PCP. At all four prepulses

PCP significantly disrupted PPI in PSD-95 wildtype mice. PCP significantly disrupted PPI at 4 and 12 dB in PSD-95 knockout mice. In PSD-95 wildtype mice, clozapine co-injection with PCP normalized the disruption of PPI at 4,

12, and 16 dB, with a trend towards normalization at 8 dB. In knockout mice, clozapine potentiated PCP disruption of PPI at 8 and 16 dB and had no antipsychotic effect at 4 and 12 dB. Data are given as means +/- the

SEM. *p < 0.05, **p < 0.01, ***p < 0.001; Two-way repeated measures ANOVA

followed by Bonferroni post-tests.

** ** *** * * * * ** ***

** ** ** **

Figure 3.21 – Clozapine Effect on PCP-induced Disruption of PPI at 16 dB

Prepulse in PSD-95 Mice

PPI at 16 dB illustrating the contrast between the antipsychotic effect of

clozapine in wildtypes and the “pro-psychotic”, potentiating effect of

clozapine in knockouts.

clozapine NORMALIZES DISRUPTION of PPI by PCP in WTs

PCP DISRUPTS PPI in WTs clozapine POTENTIATES DISRUPTION of PPI by PCP in KOs

16 dB

Given that clozapine is also a D2 antagonist, an alternative explanation of the

data is that dopaminergic dysfunction is responsible for the lack of antipsychotic

efficacy clozapine exhibited in the absence of PSD-95. If that were the case, one

would predict that even high doses of clozapine would lack efficacy. On the

other hand, we hypothesized that the serotonergic dysfunction we have shown in

87 the absence of PSD-95 is responsible for clozapine’s lack of efficacy. Thus, we

predicted that high doses of clozapine would normalize PCP-induced disruption of PPI via D2 antagonism. This prediction results from the observation that

typical antipsychotics inhibit PCP-induced disruption of PPI, but much less

potently than drugs with 5-HT2A antagonist properties. In order to test our

prediction, we assessed the ability of 1.0 and 1.5 mg/kg clozapine to normalize

disruption of PPI by PCP. As the data show, there is dose-dependent recovery

of the antipsychotic efficacy of clozapine in the absence of PSD-95 (Figure 3.22).

Figure 3.22 – Effect of Increasing Doses of Clozapine on PCP-induced

Disruption of PPI

PPI at 8, 12, and 16 dB illustrating the effect of increasing doses of

clozapine on PCP-induced disruption of PPI. The arrow illustrates

particularly well a dose-dependent increase of clozapine on PPI, which is

pro-psychotic at the lowest-dose of 0.5 mg/kg clozapine, ineffective at 1.0

mg/kg, and trending towards antipsychotic at 1.5 mg/kg.

0.5 MG/KG CLOZAPINE 100 90 80 70 60 WT VEH 50 WT PCP WT PCP + CLOZ 40

%PPI KO VEH 30 KO PCP 20 KO PCP + CLOZ 10 0 8 dB 12 dB 16 dB -10 Prepulse Level -20

3.3 – Discussion

3.3.1 – Major Findings

As described at length in the introduction, GPCRs couple to multiple G protein- dependent and G protein-independent pathways. Furthermore, functional selectivity has been demonstrated for a wide range of GPCRs and ligands. The

most important functional selectivity observations that have been made in the last

decade or so are: (1) ligands can differentially activate downstream pathways at the same receptor due to the fact that GPCRs exist in a probabilistic conformational distribution, with different ligands altering the distribution in different ways. Different sub-distributions within a conformational ensemble are thought to correspond to different downstream pathways. (2) Receptor environment also plays a critical role in determining the downstream signaling texture available to a receptor – as a result, a receptor can have different downstream signatures in different cell lines. Two possibilities can explain this finding. First, it could be due to differential subcellular compartmentalization of receptors, which in turn determines available regulatory and signaling partners.

Second, it could be due to the fact that regulatory partners in a subcellular compartment influence the available conformations that can be sampled before and after ligand binding. Either or both of the aforementioned possibilities may play a role in determining the environment-dependent texture of downstream signaling.

Applied to the 5-HT2A receptor field, functional selectivity has the potential to

explain some of the more vexing questions that have troubled 5-HT2A researchers. First, it has been observed that almost all antagonists at the 5-HT2A receptor, including atypical antipsychotics, induce downregulation (termed paradoxical downregulation), a finding that is impossible to reconcile with classical receptor theory. By proposing that receptors exist in conformational ensembles, rather than the binary model of classical theory (inactive or active state), functional selectivity suggests a possible explanation. In classical GPCR theory, it is the active state, or some sequence of events initiated by the active state, that promotes internalization and hence downregulation over the long-term.

Functional selectivity suggests instead that sub-distributions exist that correspond to the different downstream signaling pathways, and other sometimes overlapping (with each other and with those corresponding to signaling pathways) sub-distributions correspond to the different possible routes of internalization/trafficking. Thus, a drug could theoretically shift the conformational ensemble away from those conformations that promote downstream signaling while sparing or increasing the probabilities of some or all of those that promote internalization. Second, it is also an oddity of 5-HT2A receptor activity that most, but not all, agonists are hallucinogenic. This, too, is irreconcilable with classical theory but compatible with functional selectivity. The prediction would be that hallucinogenic 5-HT2A agonists induce/stabilize a subset

of conformations that are not seen after non-hallucinogenic 5-HT2A agonist receptor activation.

An important step in the process of understanding how 5-HT2A receptors

function with respect to hallucinogenic actions and psychosis is to characterize

the relevant subcellular environment or environments. Though different

conformational sub-distributions are thought to correspond to different

downstream receptor-related events, the subcellular localization of the receptor

likely plays a role also. As a hypothetical example, the signaling initiated by

some conformational sub-distribution is dependent on downstream effector

proteins, which may not be present at all the subcellular locations in which a

receptor is found. In such an instance, despite the ability of a ligand to access

the appropriate conformations corresponding to the hypothetical pathway in

question, activation of that pathway will be subcellular location-dependent. We

are most interested in the 5-HT2A signaling pathways that are relevant with

respect to hallucinogen action and psychosis. Thus, it is critical to identify the

subcellular locale or locales through which the 5-HT2A receptor modulates

psychosis-related behavior and hallucinogenic activities.

Broadly, it is increasingly apparent that proteins are directed to specialized

subcellular locations in large part by a litany of scaffolding proteins, which would

suggest that the 5-HT2A receptor should be scaffolded by one or more proteins at

one or more subcellular locations in neurons. Recent evidence suggesting that

cortical 5-HT2A receptors are required for hallucinogen actions, possibly by

facilitating corticocortical activity, points the way towards identifying at least one of these subcellular locations (57). Consistent with this data, the primary

92 neuroanatomical site of expression of 5-HT2A receptors is the apical dendrites of

cortical pyramidal neurons, particularly in layer V pyramidal neurons (35), (36).

Furthermore, a wide range of evidence supports altered glutamatergic signaling

in neocortex as playing a key role in mediating the effects of hallucinogens on consciousness (74). As described in more detail in the introduction, apical dendritic activity has been implicated as forming the neural basis for cognition and consciousness (199), (73), and it is thought that corticocortical connections, which are primarily composed of synaptic contacts at apical dendrites (200), are important in generating and shaping the neural activity that underlies consciousness (72). Overall, the data strongly implicate the apical dendrites of cortical pyramidal neurons as being the most important neuroanatomical location of 5-HT2A receptors with respect to hallucinogen action and psychosis.

Data showing that the 5-HT2A receptor interacts with the PDZ domain-containing

PSD-95, thought to be the major scaffolding protein of the postsynaptic

glutamatergic signaling complex (84), suggests this complex as an important site

for the 5-HT2A receptor’s actions. Importantly, PSD-95 is heavily expressed at

postsynaptic locations, which is consistent with what is known about 5-HT2A localization and function. Also consistent with those findings, the 5-HT2A receptor’s PDZ ligand is necessary for dendritic targeting of the receptor (116).

This led to our hypothesis that PSD-95 is a critical regulator of 5-HT2A function in

vivo. Broadly stated, the resulting prediction was that, in the absence of PSD-95,

5-HT2A receptors should be mis-localized and mis-regulated in neuronal settings

93 and in vivo, and that behavioral consequences would result. Confirming this broad prediction would implicate the PSD-95-scaffolded glutamatergic signaling complex as an important site for hallucinogen action and psychosis.

More specifically, we hypothesized that PSD-95 is a critical regulator of 5-HT2A receptor trafficking, stabilizing the receptor at the membrane, and predicted that

5-HT2A expression would be reduced in vivo in the absence of PSD-95. Thus,

we performed studies examining receptor expression and showed that 5-HT2A levels were indeed reduced, and we further showed that this reduction was due to an effect of PSD-95 on trafficking and not other possible mechanisms (nonspecific effects on the serotonergic system or modulation of mRNA levels). We also showed that apical dendritic targeting of 5-HT2A receptors to postsynaptic

densities is significantly impaired in cortical neurons prepared from PSD-95

knockout mice. Consistent with this data, DOI-induced head twitch behavior, the

behavioral correlate of hallucinogen action, is reduced at the highest dose of

drug administered. Moreover, we found that the addback of PSD-95 into

previously PSD-95 knockout neurons rescues both the deficient expression and

targeting phenotype. Together, the data suggest a role for 5-HT2A receptors in

regulating glutamatergic signaling. In order to better test the hypothesis that 5-

HT2A receptors play such a role, we made the prediction that the efficacy of

atypical antipsychotics – which is mediated in part by 5-HT2A receptors – should

be reduced in glutamatergic models of psychosis.

As discussed in detail in the introduction, it has been known for some time that

PCP, a non-competitive NMDA receptor antagonist, induces psychotic and

‘deficit’ states that are nearly indistinguishable from the positive and negative

symptoms of schizophrenia (129), (136), (201). Furthermore, clozapine and

other drugs with potent 5-HT2A inverse agonist actions ameliorate PCP-induced

PPI deficits (138), (145), (202), (146). Finally, PSD-95 knockout or deletion of one of the PDZ domains results in abnormalities in LTP, a phenotype related to glutamatergic dysfunction, as would be expected with disruption of the PSD-95- scaffolded glutamatergic signaling complex (203), (100).

Since, in the absence of PSD-95, glutamatergic signaling is abnormal, and 5-

HT2A receptors are mis-targeted and mis-trafficked, we predicted that there would

be abnormalities in the ability of the prototypical atypical antipsychotic clozapine

to alleviate PCP-induced psychotic-like behaviors in mice. We found that clozapine treatment lacked antipsychotic efficacy in a glutamatergic psychosis model in PSD-95 knockout mice at doses that were effective in wildtype mice.

Furthermore, the data implicate serotonergic dysfunction, rather than dopaminergic dysfunction, as higher doses of clozapine are efficacious in PSD-

95 knockout mice, presumably through D2 antagonism, which is known to be less

potent at inhibiting PCP-induced disruption of PPI.

Our data provide a mechanism whereby 5-HT2A receptors can be targeted to a

cortical, postsynaptic site of action and trafficked and regulated appropriately

95 once they have arrived. In fact, our studies have provided the first candidate subcellular locus for 5-HT2A-mediated hallucinogen action and 5-HT2A-related effects on psychosis - the PSD-95-scaffolded macromolecular signaling complex of cortical neurons. Given the evidence that hallucinogenic action involves alterations in synaptic activity, our data further suggest the possibility that hallucinogens may act by affecting glutamatergic signaling complex function through functional interactions with one or more components of the postsynaptic signaling scaffold via 5-HT2A receptor activation. Our use of a glutamatergic-

based psychosis model, PCP-induced disruption of PPI, provides further

evidence of this interplay between 5-HT2A receptors (and thus the serotonergic

system) and the glutamatergic system, confirming that this interplay has

important behavioral consequences. These findings have wider relevance

towards understanding the functional selectivity seen at 5-HT2A receptors (e.g.,

hallucinogenic vs. non-hallucinogenic agonists, paradoxical downregulation),

which is likely related to the unique local environment of 5-HT2A receptors localized at the PSD-95-scaffolded complex. It may very well be the case that the unique trafficking and signaling signature of 5-HT2A receptors associated with

the PSD-95-scaffolded glutamatergic complex will help other researchers explain

why some 5-HT2A agonists are hallucinogens and others are not, and how

antagonists of 5-HT2A receptors exert their antipsychotic efficacy – very long-

standing questions in the field.

CHAPTER 4: PSD-95 Regulates 5-HT2C Receptor Function in vivo

4.1 – Introduction and Rationale

Previous studies demonstrated that PSD-95 interacts with 5-HT2C (90) receptors

in vitro and in vivo. Additionally, ectopic expression of PSD-95 modulates

surface expression and promotes desensitization of 5-HT2C receptors (195) in

vitro. Mutations in the type I PDZ ligand motif abrogate PSD-95 binding in vitro

(85). Analogous findings were described in the previous chapter with respect to

the closely related 5-HT2A receptor. What, if any, effect PSD-95 might have on 5-

HT2C receptors in vivo is unknown. The data suggest that PSD-95 is responsible

for proper synaptic membrane stabilization of 5-HT2C receptors. Thus, we predicted that, in the absence of PSD-95, 5-HT2C expression would be reduced due to synaptic membrane de-stabilization.

As with the 5-HT2A receptor, we examined several potential mechanisms which

might account for the PSD-95-mediated modulation of 5-HT2C receptor expression. These included: (1) non-specific effects on serotonin receptor expression and function; (2) PSD-95-mediated regulation of 5-HT receptor transcription and/or a generalized disruption of the machinery essential for neuronal regulation of receptors; (3) PSD-95-mediated alterations in serotonin

receptor mRNA editing and (4) alterations in serotonin receptor turnover.

Mechanism (3) is unique to the 5-HT2C receptor. Since we hypothesized that

PSD-95 modulates 5-HT2C expression by stabilizing the receptor at the

membrane we predicted: (1) in the absence of PSD-95, there is no change in the

97 expression of 5-HT receptors that do not have PDZ ligand motifs; (2) 5-HT2C gene transcription is unaltered in the absence PSD-95; (3) genetic deletion of

PSD-95 does not alter 5-HT2C mRNA editing; and (4) the 5-HT2C receptor turnover rate increases in the absence of PSD-95. Due to the comparatively low

expression of 5-HT2C receptors in vivo (approximately 10-fold lower than 5-HT2A

receptors), it was not possible to accurately examine the turnover rate of 5-HT2C

receptors by using EEDQ. However, by ruling out mechanisms (1) – (3), we

hoped to implicate mechanism (4) as the only remaining possibility.

Finally, evidence suggests that 5-HT2C interactions with PDZ domain-containing proteins play a critical role in mediating receptor signaling (195). Thus, we

predicted that, in the absence of PSD-95, 5-HT2C receptor signaling would be

altered. More specifically, with respect to the 5-HT2C receptor, we predicted that

5-HT2C agonist-induced neuronal activity is reduced in the absence of PSD-95.

The studies that will be discussed were aimed at testing the aforementioned predictions. We here show that PSD-95 regulates the targeting, turnover, and signaling of 5-HT2C receptors, and that this regulation has important

consequences with respect to the regulation of synaptic activity by 5-HT2C receptors.

4.2 – Results

4.2.1 – Deletion of PSD-95 Results in a Selective Loss of 5-HT2C Receptors

4.2.1.1 – 5-HT2C Immunochemistry

To test the prediction that PSD-95 knockout mice exhibit a reduction in 5-HT2C expression due to membrane de-stabilization of the receptor, we visualized 5-

HT2C receptor expression in PSD-95 wildtype and knockout mice. In order to do so, striatal and hippocampal 5-HT2C receptor expression was examined immunohistochemically in PSD-95 wildtype and knockout mice. As shown in Fig

4.1, PSD-95 knockout animals displayed large decrements of striatal and hippocampal 5-HT2C receptors as assessed by a 5-HT2C-selective antibody.

Figure 4.1 – 5-HT2C Immunochemistry in PSD-95 Mice

5-HT2C immunochemistry in PSD-95 wildtype and knockout striatum and

hippocampus reveals that 5-HT2C receptor expression is greatly reduced in the absence of PSD-95 in both regions (N=3).

4.2.1.2 – 5-HT2C Saturation Binding

The immunohistochemical microscopy images provided a qualitative indication that 5-HT2C receptor expression was reduced in the absence of PSD-95.

Saturation binding experiments were performed with [3H]-mesulergine to estimate

5-HT2C receptor expression levels quantitatively. Bmax estimates were obtained

by performing [3H]-mesulergine saturation binding in the presence of 100 nM

spiperone to block 5-HT2A receptor binding. The experiment demonstrated a

significant 72% reduction in 5-HT2C receptor expression levels in the

hippocampus in the absence of PSD-95 (Figure 4.2).

Figure 4.2 – 5-HT2C Receptor Bmax in PSD-95 Wildtype and Knockout Mice

Comparison of Bmax estimates for the 5-HT2C receptor in PSD-95 wildtype

and knockout hippocampi (N=3; tissue from three animals was pooled for each measurement, for a total of 9 animals). There is a large reduction in 5-

HT2C receptor density in PSD-95 knockout mice. Bmax data are presented as

means +/- SEM. ***p < 0.001; one-tailed, unpaired t-test.

5-HT2C Hippocampus

***

100

4.2.2 – PSD-95 Does Not Modulate Expression of the 5-HT1A Receptor

To identify the mechanism(s) by which PSD-95 modulates 5-HT2C expression, we

followed the same approach as we did for the 5-HT2A receptor. Thus, we first

examined the possibility that genetic deletion of PSD-95 causes generalized

serotonergic system dysfunction in areas of heavy 5-HT2C expression, leading to

a reduction in serotonin receptor expression in those regions. We studied this

first possibility by measuring the expression of the 5-HT1A receptor, which is also

highly expressed in hippocampus, but lacks a PDZ-ligand motif. As our

3 saturation binding experiments using [ H]-WAY100635 indicate, 5-HT1A expression levels were unchanged in PSD-95 knockout mice in the hippocampus

(Figure 4.3). These results indicate that genetic deletion of PSD-95 does not lead to a generalized serotonergic system dysfunction.

Figure 4.3 – 5-HT1A Receptor Bmax in PSD-95 Wildtype and Knockout Mice

Comparison of Bmax estimates for the 5-HT1A receptor in PSD-95 wildtype

and knockout hippocampi (N=6). Bmax data are presented as means +/-

SEM. One-tailed, unpaired t-test revealed no significant difference between genotypes. 5-HT1A Hippocampus

101

4.2.3 – PSD-95 Does Not Play a Prominent Role in Modulating Gene

Expression

4.2.3.1 – PSD-95 Does Not Alter 5-HT2C Receptor Levels Via Transcriptional

or Post-transcriptional Mechanisms

To determine if deleting PSD-95 leads to an alteration in 5-HT receptor gene

transcription, we performed quantitative RT-PCR in order to measure 5-HT2C receptor mRNA levels. Receptor mRNA levels were normalized to β-actin levels.

Our measurements show that 5-HT2C mRNA levels in the hippocampus are unaffected by genetic deletion of PSD-95 (Figure 4.4). Thus, PSD-95 does not modulate 5-HT2C receptor expression by regulating mRNA levels.

Figure 4.4 – 5-HT2C Receptor mRNA Levels in Hippocampus

Hippocampal 5-HT2C receptor mRNA levels normalized to β-actin mRNA

levels measured by quantitative RT-PCR (N=4 animals for each genotype;

five measurements for each animal). Normalized mRNA measurements are

presented as means +/- SEM. One-tailed, unpaired t-test revealed no

significant difference between genotypes.

5-HT2C mRNA Hippocampus

102

4.2.3.2 – PSD-95 Does Not Modulate RNA Editing of the 5-HT2C Receptor

The 5-HT2C receptor undergoes mRNA editing which profoundly modulates its

constitutive activity, G-protein coupling efficiency, and expression (172), (173). It

is therefore conceivable that changes in 5-HT2C receptor expression are

secondary to altered editing of 5-HT2C mRNAs. To examine this possibility, we

examined RNA editing at all possible sites in PSD-95 wildtype and knockout

hippocampal tissue, and we found that there is no change in the frequency of

editing at any of the five sites (Figure 4.5). Furthermore, there is no significant

change in the proportions of 14 of the 15 different isoforms detected in the PSD-

95 knockout mice as compared to wildtypes (Figure 4.6). An increase in PSD-95

knockout mice of the VSI isoform is inconsistent with a role for mRNA editing in down-regulating 5-HT2C receptors in PSD-95 knockout mice. These findings

indicate that neither transcriptional nor post-transcriptional mechanisms (i.e.,

RNA editing) can account for the large effect that genetic deletion of PSD-95 has

on the expression of 5-HT2A and 5-HT2C receptors.

103

Figure 4.5 – 5-HT2C mRNA Editing Frequencies in PSD-95 Mice by Site

Data are plotted as the frequency of editing events expressed as a fraction

of the total, +/- the SEM. One-way ANOVA followed by Newman-Keuls post-

hoc tests revealed no significant difference between genotypes.

Figure 4.6 – Frequencies of Edited Isoforms in PSD-95 Mice (wildtypes,

N=94; knockouts N=93).

15 isoforms were detected, and 14 of them were not significantly altered in

the absence of PSD-95. Data are plotted as the isoform frequency

expressed as a fraction of the total, +/- the SEM. *p < 0.05, **p < 0.01; One-

way ANOVA followed by Newman-Keuls post-hoc tests revealed no significant difference between genotypes.

104

4.2.4 – PSD-95 is Required for 5-HT2C Signaling in vivo

Having provided strong evidence that PSD-95 regulates the expression of 5-HT2C receptors, we next examined the consequences of knocking out PSD-95 on 5-

HT2C signaling in vivo. We predicted that 5-HT2C signaling would be impaired. It

is well established that c-fos is an immediate early gene (IEG) that is transcribed

after GPCR activation (204) and is useful as a general marker of neuronal activity

(205). To examine the consequences of genetic deletion of PSD-95 on signaling

downstream of the 5-HT2C receptor and on neural activity, we treated mice with

MK-212, a 5-HT2C-selective agonist (190), and measured induction of c-fos in the hippocampus. Notably, we found that the number of c-fos-positive cells after

MK-212 treatment was greatly reduced in PSD-95 knockout animals in a number of hippocampal subregions (Figure 4.7). Our finding that MK-212 induces the largest c-fos response in the dentate gyrus (DG) region of the hippocampus is in accordance with prior studies (206). We also found that c-fos was induced in 5-

HT2C-expressing neurons, suggesting that 5-HT2C-activation is inducing the IEG protein directly, rather than indirectly in surrounding neurons (Figure 4.8). This decrease in c-fos induction seen in all hippocampal regions measured was highly significant (Figure 4.9) and indicates that genetic deletion of PSD-95 greatly attenuates 5-HT2C signaling in vivo.

105

Figure 4.7 – MK-212 Induction of c-fos in the Hippocampus of PSD-95 Mice

5-HT2C and c-fos double-label immunochemistry in the hippocampus of

PSD-95 wildtype and knockout mice after MK-212 treatment (N=3).

Representative images of CA1, CA2, CA3, and DG are shown. There are fewer c-fos-positive cells in the PSD-95 knockout mice treated with MK-212 in all the examined regions. 5-HT2C c-fos merge PSD-95 WT CA1 PSD-95 KO PSD-95 WT PSD-95 CA2 PSD-95 KO

106

5-HT2C c-fos merge PSD-95 WT CA3 PSD-95 KO PSD-95 PSD-95 WT DG PSD-95 KO

107

Figure 4.8 – c-fos Induction in 5-HT2C-Expressing Cells in Hippocampus

Higher magnification image of CA1 in order to examine co-localization of 5-

HT2C receptors and c-fos. 5-HT2C receptor co-localizes with c-fos, suggesting that 5-HT2C is inducing this IEG directly, rather than indirectly in surrounding neurons. 5-HT2C c-fos merge PSD-95 WT CA1 PSD-95 KO PSD-95

108

Figure 4.9 – c-fos Quantitation in PSD-95 Wildtype and Knockout

Hippocampus

Analysis of c-fos induction was performed by counting the number of c-

fos-positive cells in CA1, CA2, CA3, and dentate gyrus (DG). Data are

presented as the mean number of c-fos-positive cells +/- the SEM. c-fos counts were performed separately in the hippocampus of each hemisphere

(two values for each section analyzed). Every seventh section was

analyzed, for a total of six sections per animal. *p < 0.05, **p < 0.01, ***p <

0.001; one-tailed, unpaired t-test.

* **

* * * * **

109

4.3 – Discussion

4.3.1 – Major Findings

The 5-HT2C receptor, though not involved in mediating the main effects of

hallucinogens, resembles the 5-HT2A receptor in a number of important respects.

Like 5-HT2A receptors, 5-HT2C receptors often undergo paradoxical

downregulation after chronic antagonist treatment (207). Ligands can also

exhibit functional selectivity at 5-HT2C receptors (32). The 5-HT2C receptor is

unique among GPCRs because its RNA undergoes post-transcriptional editing,

which in turn can alter the amino acid translated at 3 positions in intracellular loop

two (170). This editing has been shown to alter the downstream signaling texture

(174), (175).

Also like the 5-HT2A receptor, there is converging evidence, much of it described

in greater detail in the introduction, that 5-HT2C receptor signaling can modulate

electrical activity in the brain. 5-HT2C knockout mice are prone to audiogenic seizures (181); 5-HT2C agonists suppress theta wave oscillations, inhibit

hippocampal excitability, and de-synchronize the septo-hippocampal system

(184), (185); whereas 5-HT2C antagonists promote theta wave oscillations and

septo-hippocampal synchrony (184). These findings in particular implicate 5-

HT2C receptors in regulating synaptic activity.

There is also strong evidence that agonists or antagonists at 5-HT2C receptors

may be of benefit in the treatment of some psychiatric illnesses. The compound

110

[(7bR,10aR)-1,2,3,4,8,9,10,10a-octahydro-7bH-cyclopenta-

[b][1,4]diazepino[6,7,1hi]indole] (WAY163909) is a 5-HT2C agonist that has been

shown to be effective in a number of pre-clinical models of antipsychotic efficacy

(Table 1.4) (187). A number of 5-HT2C agonists are also being developed as safe and effective anorectic agents. For example, WAY163909 has also been shown to reduce food intake in preclinical models (208), and lorcaserin has been shown to be effective in humans in phase II trials. 5-HT2C receptors have also

been implicated in the etiology of obsessive-compulsive disorder (OCD) (209)

and 5-HT2C agonists have been proposed as potential therapies for OCD (189).

Together, the data suggest that 5-HT2C receptors play an important role modulating electrical activity, possibly by functional interactions with glutamatergic synaptic function. As with 5-HT2A receptors, the question of how 5-

HT2C receptors might modulate synaptic function is largely unexplored. The finding that 5-HT2C receptors interact with PSD-95 (85) and that interactions with

PDZ domain-containing proteins affect 5-HT2C function in vitro (195) suggests

that PSD-95 may mediate the 5-HT2C receptor’s interactions with synaptic

glutamatergic activity. Our hypothesis closely paralleled that of the previous

chapter. More specifically, we hypothesized that PSD-95 is a critical regulator of

neuronal 5-HT2C membrane stability, trafficking, and function in vivo. We thus

predicted that 5-HT2C expression would be reduced in the absence of PSD-95

due to an effect on receptor trafficking, rather than one on general serotonergic

function, mRNA levels, and/or RNA editing.

111

We began by examining 5-HT2C receptor expression in vivo. We showed that 5-

HT2C expression was almost undetectable in striatum and hippocampus in PSD-

95 knockout mice. Quantitative estimates of 5-HT2C expression suggested a

72% reduction in 5-HT2C expression in hippocampus in the absence of PSD-95.

The reduction in expression suggests that PSD-95 stabilizes the 5-HT2C receptor at the plasma membrane. Analogous to the experiments performed to study the

5-HT2A receptor, we showed that, in the absence of PSD-95, 5-HT1A expression

and RNA editing were unaltered in hippocampus, pointing to abnormal trafficking

as being responsible for the reduction in 5-HT2C expression. Trafficking

experiments could not be performed due to the extremely low expression of 5-

HT2C receptor protein compared to that of 5-HT2A receptors. Nonetheless, the

EEDQ experiment performed to examine trafficking of the closely related 5-HT2A receptor, in conjunction with ruling out other explanations for the reductions in 5-

HT2A and 5-HT2C receptor expression, clearly point to trafficking as the likely

explanation. We also predicted that 5-HT2C receptor signaling would be impaired in the absence of PSD-95 – in particular, its modulation of neuronal activation.

Thus, we showed that 5-HT2C-agonist-induced c-fos protein induction (a marker

of neuronal activation) is dramatically reduced in PSD-95 knockout mice,

confirming a role for PSD-95 in regulating the 5-HT2C receptor’s ability to

modulate neuronal activity.

112

Taken altogether, the data suggest that the 5-HT2C receptor is located at PSD-

95-scaffolded postsynaptic glutamatergic signaling complexes. This subcellular localization is critical for normal receptor membrane stabilization, trafficking, and downstream signaling. The discovery of an in vivo role for PSD-95 in regulating

5-HT2C function suggests the possibility that this may be an important site for mediating the therapeutic effects of 5-HT2C-targeted drugs, for example with respect to treating obesity, OCD, and/or psychosis.

113

CHAPTER 5: Future Directions

5.1 – Regulation of Cortical 5-HT2A Receptor Function, Mechanisms of

Hallucinogen Action, and the Basis for Psychosis-Related Signaling Events

5.1.1 – Exploring the Relative Importance of PDZ Domain-Mediated

Interactions With Respect to Neuronal 5-HT2A Receptor Function

The findings presented in Chapter 3 represent the first direct evidence supporting

the long hypothesized link between 5-HT2A signaling events and glutamatergic

signaling at cortical postsynaptic densities. Our findings suggest that

hallucinogenic 5-HT2A agonists mediate their hallucinogenic actions via signaling

events that take place at the subpopulation of receptors localized at the PSD-95-

scaffolded PSD, a subcellular microdomain that specializes in regulating glutamatergic signaling at the synapse. Furthermore, our research suggests that the antipsychotic efficacy mediated by clozapine via the 5-HT2A receptor involves

the same subpopulation of apical dendritic receptors.

One weakness of the studies in this dissertation is that 5-HT2A receptors are free to interact via their PDZ ligand motif with other MAGUKs besides PSD-95, such

as PSD-93, SAP97, and SAP102 (86). PSD-93 and PSD-95 are enriched in

PSDs, whereas SAP97 and SAP102 are found in dendrites and axons and are heavily expressed both in the cytoplasm and at synapses (86). Evidence suggests that PSD-93 and PSD-95 may be more specifically associated with synaptic function, whereas SAP97 and SAP102 may play a role in trafficking

(86). Overall, however, the different roles of the aforementioned MAGUKs are

114

not well characterized. Thus, in the absence of PSD-95, other MAGUKs may be

able to at least partially compensate for any loss of function, or one or more of

them may normally play a role in hallucinogenic signaling. Notably, the head

twitch response to DOI administration was only reduced 35%, which suggests

one or more of three possibilities, two of which were just mentioned: 1) other

MAGUKs compensate for PSD-95 in its absence, and can scaffold networks

capable of mediating hallucinogenic signaling pathways; 2) multiple MAGUKs

may be involved in mediating hallucinogenic signaling; and 3) other non-MAGUK

scaffolding proteins are involved in mediating hallucinogen signaling pathways,

either alone or in concert with PSD-95 (Table 5.1).

Table 5.1 – Regulation of 5-HT2A-mediated Hallucinogen-Related Signaling

Regulation of 5-HT2A Receptor-Mediated Hallucinogenic Signaling in vivo

MAGUKs compensate for PSD-95 due to genetic deletion (but normally do not play a role in hallucinogenic signaling)

Multiple MAGUKs are normally involved in mediating hallucinogen actions through the 5- HT2A receptor

Non-MAGUK scaffolding proteins/interacting partners (i.e., RSK2, Cav-1) may play an important role in mediating hallucinogen actions

To determine which of the aforementioned possibilities is most consistent with in

vivo 5-HT2A function, an important first step would be to study in vivo the function

of a 5-HT2A mutant that cannot interact with PDZ domain-containing proteins.

Previous and current studies in our laboratory have characterized such a mutant,

5-HT2A-GFP-AAA, in which the C-terminal SCV is mutated to AAA and a GFP tag

115

is included to facilitate receptor visualization. 5-HT2A-GFP-AAA has been shown

not to interact with PSD-95 (84) and does not traffic to apical dendrites in

cultured neurons (116). If possibilities 1) and/or 2) describe the in vivo role of 5-

HT2A-PDZ protein interactions in regulating 5-HT2A function, then the 5-HT2A-

related dysfunction in a 5-HT2A-GFP-AAA mouse should be more pronounced

than that seen in PSD-95 knockout mice. More specifically, steady state 5-HT2A expression would be expected to be lower, turnover rate higher, and hallucinogen-induced head twitch lower (as compared to PSD-95 knockout mice). Thus the creation and characterization of mutant mice expressing only 5-

HT2A-GFP-AAA receptors would be useful in determining the overall extent to

which receptor-PDZ protein interactions are critical in regulating 5-HT2A receptor function. Furthermore, by comparing their phenotype to that of PSD-95 knockout mice, one can determine the contribution other MAGUKs are making to receptor function. The most conclusive way to distinguish between possibilities 1) and 2), however, would be to also perform the same experiments in mouse knockouts of other MAGUKs. Possibility 3) can only be examined by performing analogous experiments in mice with genetic deletions of other interacting partners of interest, such as Caveolin-1. Careful characterization of 5-HT2A-interacting

proteins that are likely to play a role in regulating receptor function should

eventually provide a clearer picture of the relative contributions of these different

proteins to different aspects of 5-HT2A receptor function.

116

5.1.2 – 5-HT2A Receptor Functional Selectivity as it Relates to Hallucinogen

Action

A related set of questions that need to be answered in the future revolve around the relative contributions of the local environment of the Gαq-coupled 5-HT2A receptor as compared to the receptor itself in terms of conferring the ability to mediate hallucinogenic agonist signaling. Though it is now thought that canonical Gαq signaling is not responsible for mediating hallucinogenic signaling,

whether 5-HT2A receptor-mediated hallucinogenic signaling is due to the cellular

context in which the receptor is expressed or intrinsic to the receptor remains an

unresolved question. Our data provide evidence that the cellular context, via

auxiliary proteins such as PSD-95, contributes significantly to 5-HT2A receptor

hallucinogenic signaling. Nonetheless, the data suggest two main possibilities:

A) the local environment, which is essentially determined by the PDZ ligand motif, is primarily what confers the ability of the 5-HT2A receptor to mediate

hallucinogenic signaling, probably by regulating substrate specificity (210) or B)

the local environment acts in concert with unique signaling characteristics of the

5-HT2A receptor to mediate hallucinogenic signaling. Technology that would facilitate the differentiation between A) and B) has recently been developed within our laboratory.

Designer Receptors Exclusively Activated by Designer Drugs (DREADDs) based

on the Gαq-coupled M3 muscarinic receptor have been engineered by screening

large libraries of random receptor mutants for variants that have acquired the

117 ability to be activated by clozapine-N-oxide (CNO), an inert (at any receptor) metabolite of clozapine, and have also lost the ability to be activated by the endogenous ligand, acetylcholine (211). By adding the 5-HT2A PDZ ligand motif to the M3 DREADD (M3-DREADD-2APDZ) and driving receptor expression via the 5-HT2A promoter using BAC transgenic technology (212), one should be able to target an alternative Gαq-coupled receptor to the same subcellular locations as the 5-HT2A receptor. If administration of CNO to an M3-DREADD-2APDZ mouse induces head twitch, such a result would argue strongly for possibility A), since hallucinogenic signaling appears to be a characteristic of any (or at least multiple) Gαq-coupled GPCRs, just as long as they are targeted appropriately – in other words, the local environment confers hallucinogenic signaling properties.

On the other hand, a lack of head twitch response would point to B), suggesting that the 5-HT2A receptor has unique, intrinsic signaling properties that are at least as critical as the local environment in mediating hallucinogenic action.

5.1.3 – Hallucinogenic Signaling Events Downstream of 5-HT2A Activation

Another remaining mystery is the identity of the downstream events that mediate hallucinogen actions. Previous research presented in the introduction in combination with the findings described in Chapter 3 suggests some directions and possibilities. First, there is some evidence that per1, egr-1, and egr-2 are induced selectively by 5-HT2A hallucinogenic agonists, and not by non- hallucinogenic agonists. Our initial data are consistent with that finding, as PSD-

95 knockout mouse cortex exhibits lower levels of both per1 and egr-2

118

transcripts, suggesting an impaired ability to mediate critical upstream signaling

events. Future experiments to confirm whether or not per1, egr-1, and egr-2

induction is impaired will be interesting in this respect. Though these IEGs do

not represent the signaling events responsible for mediating the actions of

hallucinogens, they may be downstream of those signaling events. Thus, one

more of these IEGs may be useful in assessing the impact of future candidate signaling cascades upon hallucinogen actions. Signaling cascades that are critical mediators of hallucinogen actions should induce per1, egr-1, and egr-2.

5.1.4 – Exploring the Extent of Macromolecular Disruption and How

Different Proteins in the Extended PSD-95 Scaffolded Network Modulate 5-

HT2A Receptor Function

A final set of points relates to establishing more firmly the importance of the PSD-

95 scaffold in regulating 5-HT2A receptor signaling, the role of the scaffold with

respect to psychotic-like behaviors, and the importance of ionotropic glutamate

receptor signaling in mediating hallucinogen actions. It follows from our data that, if the PSD-95 scaffolded macromolecular network plays a prominent role in regulating 5-HT2A receptor function, then other proteins that are directly or

indirectly scaffolded are likely to modulate 5-HT2A-mediated events. Confirming

such modulation would provide further evidence that 5-HT2A receptors are regulated by and participate in the large PSD-95-scaffolded signaling complex.

Such experiments could also provide further evidence supporting the newly discovered role for the PSD-95 complex in psychosis.

119

An example of a specific testable prediction involves the Gαq-coupled group I

metabotropic glutamate receptors (mGluRs), mGluR1 and mGluR5 (213). As

detailed in Figure 1.4 and the references contained therein, mGluR1/5 are

scaffolded indirectly in a large macromolecular complex that includes PSD-95.

Agonists of mGluR1 (214) and mGluR5 (215), (216) have been shown to

potentiate NMDA and AMPA responses. Surprisingly, mGluR1 (217) and

mGluR5 (218) knockout mice exhibit impaired PPI, and mGluR5 antagonists

augment PCP-induced deficits in PPI (219), though mGluR1 antagonists have no

effect on MK-801-induced deficits in PPI (220). Thus, it is worth exploring

whether or not mGluR1 and mGluR5 agonists and antagonists can modulate

head twitch behavior and PCP-induced disruptions of PPI, both in the presence and absence of PSD-95. It is difficult to predict whether or not 5-HT2A receptors

will be complexed with group I mGluRs in the absence of PSD-95, as they could

conceivably still associate indirectly via RSK2-Shank-Homer (Figure 1.4). Our

data, however, suggest that PSD-95 plays an important role in trafficking and

targeting 5-HT2A receptors, without which 5-HT2A may be unable to associate

indirectly with group I mGluRs. Given our data, we would predict that drugs

targeting group I mGluRs will be unable to modulate hallucinogen-induced head

twitch behavior or PCP-induced disruption of PPI in the absence of PSD-95, but

the other possible outcomes would also be informative (Table 5.2).

120

Table 5.2 – Possible Outcomes of a Group I mGluR Experiment in the

Presence and Absence of PSD-95

Outcome (in PSD-95 Conclusion Knockout Mice)

Group I mGluR drugs do Without PSD-95, 5-HT2A receptors are unable to complex not modulate either indirectly with group I mGluRs, and/or group I mGluRs are no head twitch or PCP- longer able to communicate with the ionotropic glutamate induced disruption of system, suggesting that PSD-95 mediates the interplay PPI between metabotropic glutamate signaling on the one hand and serotonergic (head twitch) and ionotropic glutamate signaling on the other (NMDA antagonist-induced disruption of PPI)

Group I mGluR drugs 5-HT2A receptors are still able to complex indirectly with group modulate only head I mGluRs, possibly through RSK2-Shank-Homer, thus twitch allowing metabotropic glutamate receptors to modulate 5- HT2A-mediated events, but group I mGluRs cannot complex with ionotropic glutamate receptors in the absence of PSD- 95; further suggests that hallucinogen action is not dependent on ionotropic glutamatergic modulation

Group I mGluR drugs In the absence of PSD-95, 5-HT2A receptors are unable to modulate only PCP- traffic to apical dendrites to localize near group I mGluRs, but induced disruption of group I mGluRs are still modulating the ionotropic glutamate PPI system, possibly through other MAGUKs

Group I mGluR drugs 5-HT2A receptors are still able to complex indirectly with group modulate both head I mGluRs, and group I mGluRs are still communicating with twitch and PCP-induced the ionotropic glutamate system, possibly through other disruption of PPI MAGUKs

Clearly, many questions remain concerning how the 5-HT2A receptor is regulated,

how it mediates hallucinogen actions, how it affects psychotic behaviors, and the

identity of the biochemical mediators linking metabotropic serotonin and

ionotropic glutamate neurotransmission. The findings presented in this dissertation begin to address these questions, and they may inspire experiments that can potentially generate the more specific answers which have eluded

researchers for some time. The evidence presented in Chapters 3 and 4

121 suggests PSD-95 is a central link between the serotonergic and glutamatergic system, though it does not rule out other links. By continuing where the studies presented herein leave off, future researchers in the field should be able to answer some of the most pressing questions in multiple fields of inquiry. Our hope is that, armed with some of the most recent data to guide their hypotheses, their efforts will be as fruitful as ours have been.

122

BIBLIOGRAPHY

1. Kroeze, W.K., D.J. Sheffler, and B.L. Roth, G-protein-coupled receptors at a glance. J Cell Sci, 2003. 116(Pt 24): p. 4867-9. 2. Nathans, J. and D.S. Hogness, Isolation, sequence analysis, and intron- exon arrangement of the gene encoding bovine rhodopsin. Cell, 1983. 34(3): p. 807-14. 3. Kristiansen, K., Molecular mechanisms of ligand binding, signaling, and regulation within the superfamily of G-protein-coupled receptors: molecular modeling and mutagenesis approaches to receptor structure and function. Pharmacol Ther, 2004. 103(1): p. 21-80. 4. Palczewski, K., T. Kumasaka, T. Hori, C.A. Behnke, H. Motoshima, B.A. Fox, I. Le Trong, D.C. Teller, T. Okada, R.E. Stenkamp, M. Yamamoto, and M. Miyano, Crystal structure of rhodopsin: A G protein-coupled receptor. Science, 2000. 289(5480): p. 739-45. 5. Cherezov, V., D.M. Rosenbaum, M.A. Hanson, S.G. Rasmussen, F.S. Thian, T.S. Kobilka, H.J. Choi, P. Kuhn, W.I. Weis, B.K. Kobilka, and R.C. Stevens, High-resolution crystal structure of an engineered human beta2- adrenergic G protein-coupled receptor. Science, 2007. 318(5854): p. 1258-65. 6. Rasmussen, S.G., H.J. Choi, D.M. Rosenbaum, T.S. Kobilka, F.S. Thian, P.C. Edwards, M. Burghammer, V.R. Ratnala, R. Sanishvili, R.F. Fischetti, G.F. Schertler, W.I. Weis, and B.K. Kobilka, Crystal structure of the human beta2 adrenergic G-protein-coupled receptor. Nature, 2007. 450(7168): p. 383-7. 7. Rosenbaum, D.M., V. Cherezov, M.A. Hanson, S.G. Rasmussen, F.S. Thian, T.S. Kobilka, H.J. Choi, X.J. Yao, W.I. Weis, R.C. Stevens, and B.K. Kobilka, GPCR engineering yields high-resolution structural insights into beta2-adrenergic receptor function. Science, 2007. 318(5854): p. 1266-73. 8. Murakami, M. and T. Kouyama, Crystal structure of squid rhodopsin. Nature, 2008. 453(7193): p. 363-7. 9. Hermans, E., Biochemical and pharmacological control of the multiplicity of coupling at G-protein-coupled receptors. Pharmacol Ther, 2003. 99(1): p. 25-44. 10. Pierce, K.L., R.T. Premont, and R.J. Lefkowitz, Seven-transmembrane receptors. Nat Rev Mol Cell Biol, 2002. 3(9): p. 639-50. 11. Ferguson, S.S., Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev, 2001. 53(1): p. 1-24. 12. Roth, B.L., E.P. Palvimaki, S. Berry, N. Khan, N. Sachs, A. Uluer, and M.S. Choudhary, 5-Hydroxytryptamine2A (5-HT2A) receptor desensitization can occur without down-regulation. J Pharmacol Exp Ther, 1995. 275(3): p. 1638-46. 13. Berry, S.A., M.C. Shah, N. Khan, and B.L. Roth, Rapid agonist-induced internalization of the 5-hydroxytryptamine2A receptor occurs via the endosome pathway in vitro. Mol Pharmacol, 1996. 50(2): p. 306-13. 123

14. Gray, J.A. and B.L. Roth, Paradoxical trafficking and regulation of 5- HT(2A) receptors by agonists and antagonists. Brain Res Bull, 2001. 56(5): p. 441-51. 15. Gray, J.A., D.J. Sheffler, A. Bhatnagar, J.A. Woods, S.J. Hufeisen, J.L. Benovic, and B.L. Roth, Cell-type specific effects of endocytosis inhibitors on 5-hydroxytryptamine(2A) receptor desensitization and resensitization reveal an arrestin-, GRK2-, and GRK5-independent mode of regulation in human embryonic kidney 293 cells. Mol Pharmacol, 2001. 60(5): p. 1020- 30. 16. Lefkowitz, R.J. and E.J. Whalen, beta-arrestins: traffic cops of cell signaling. Curr Opin Cell Biol, 2004. 16(2): p. 162-8. 17. Lefkowitz, R.J., K. Rajagopal, and E.J. Whalen, New roles for beta- arrestins in cell signaling: not just for seven-transmembrane receptors. Mol Cell, 2006. 24(5): p. 643-52. 18. Luttrell, L.M., S.S. Ferguson, Y. Daaka, W.E. Miller, S. Maudsley, G.J. Della Rocca, F. Lin, H. Kawakatsu, K. Owada, D.K. Luttrell, M.G. Caron, and R.J. Lefkowitz, Beta-arrestin-dependent formation of beta2 adrenergic receptor-Src protein kinase complexes. Science, 1999. 283(5402): p. 655- 61. 19. Schmid, C.L., K.M. Raehal, and L.M. Bohn, Agonist-directed signaling of the serotonin 2A receptor depends on beta-arrestin-2 interactions in vivo. Proc Natl Acad Sci U S A, 2008. 105(3): p. 1079-84. 20. Stephenson, R.P., A modification of receptor theory. Br J Pharmacol Chemother, 1956. 11(4): p. 379-93. 21. Weiss, J.M., P.H. Morgan, M.W. Lutz, and T.P. Kenakin, The cubic ternary complex receptor-occupancy model. III. resurrecting efficacy. J Theor Biol, 1996. 181(4): p. 381-97. 22. De Lean, A., J.M. Stadel, and R.J. Lefkowitz, A ternary complex model explains the agonist-specific binding properties of the adenylate cyclase- coupled beta-adrenergic receptor. J Biol Chem, 1980. 255(15): p. 7108- 17. 23. Costa, T. and A. Herz, Antagonists with negative intrinsic activity at delta opioid receptors coupled to GTP-binding proteins. Proc Natl Acad Sci U S A, 1989. 86(19): p. 7321-5. 24. Samama, P., S. Cotecchia, T. Costa, and R.J. Lefkowitz, A mutation- induced activated state of the beta 2-adrenergic receptor. Extending the ternary complex model. J Biol Chem, 1993. 268(7): p. 4625-36. 25. Kenakin, T., Principles: receptor theory in pharmacology. Trends Pharmacol Sci, 2004. 25(4): p. 186-92. 26. Urban, J.D., W.P. Clarke, M. von Zastrow, D.E. Nichols, B. Kobilka, H. Weinstein, J.A. Javitch, B.L. Roth, A. Christopoulos, P.M. Sexton, K.J. Miller, M. Spedding, and R.B. Mailman, Functional selectivity and classical concepts of quantitative pharmacology. J Pharmacol Exp Ther, 2007. 320(1): p. 1-13. 27. Kenakin, T., Inverse, protean, and ligand-selective agonism: matters of receptor conformation. Faseb J, 2001. 15(3): p. 598-611.

124

28. Kenakin, T., Functional selectivity through protean and biased agonism: who steers the ship? Mol Pharmacol, 2007. 72(6): p. 1393-401. 29. Roth, B.L. and D.M. Chuang, Multiple mechanisms of serotonergic signal transduction. Life Sci, 1987. 41(9): p. 1051-64. 30. Gray, J.A., A. Bhatnagar, V.V. Gurevich, and B.L. Roth, The interaction of a constitutively active arrestin with the arrestin-insensitive 5-HT(2A) receptor induces agonist-independent internalization. Mol Pharmacol, 2003. 63(5): p. 961-72. 31. Spengler, D., C. Waeber, C. Pantaloni, F. Holsboer, J. Bockaert, P.H. Seeburg, and L. Journot, Differential signal transduction by five splice variants of the PACAP receptor. Nature, 1993. 365(6442): p. 170-5. 32. Berg, K.A., S. Maayani, J. Goldfarb, C. Scaramellini, P. Leff, and W.P. Clarke, Effector pathway-dependent relative efficacy at serotonin type 2A and 2C receptors: evidence for agonist-directed trafficking of receptor stimulus. Mol Pharmacol, 1998. 54(1): p. 94-104. 33. Mottola, D.M., J.D. Kilts, M.M. Lewis, H.S. Connery, Q.D. Walker, S.R. Jones, R.G. Booth, D.K. Hyslop, M. Piercey, R.M. Wightman, C.P. Lawler, D.E. Nichols, and R.B. Mailman, Functional selectivity of dopamine receptor agonists. I. Selective activation of postsynaptic dopamine D2 receptors linked to adenylate cyclase. J Pharmacol Exp Ther, 2002. 301(3): p. 1166-78. 34. Mailman, R.B., GPCR functional selectivity has therapeutic impact. Trends Pharmacol Sci, 2007. 28(8): p. 390-6. 35. Willins, D.L., A.Y. Deutch, and B.L. Roth, Serotonin 5-HT2A receptors are expressed on pyramidal cells and interneurons in the rat cortex. Synapse, 1997. 27(1): p. 79-82. 36. Jakab, R.L. and P.S. Goldman-Rakic, 5-Hydroxytryptamine2A serotonin receptors in the primate cerebral cortex: possible site of action of hallucinogenic and antipsychotic drugs in pyramidal cell apical dendrites. Proc Natl Acad Sci U S A, 1998. 95(2): p. 735-40. 37. Nocjar, C., B.L. Roth, and E.A. Pehek, Localization of 5-HT(2A) receptors on dopamine cells in subnuclei of the midbrain A10 cell group. Neuroscience, 2002. 111(1): p. 163-76. 38. McKenna, D.J. and J.M. Saavedra, Autoradiography of LSD and 2,5- dimethoxyphenylisopropylamine psychotomimetics demonstrates regional, specific cross-displacement in the rat brain. Eur J Pharmacol, 1987. 142(2): p. 313-5. 39. Nichols, D.E., Hallucinogens. Pharmacol Ther, 2004. 101(2): p. 131-81. 40. Roth, B.L., K. Baner, R. Westkaemper, D. Siebert, K.C. Rice, S. Steinberg, P. Ernsberger, and R.B. Rothman, Salvinorin A: a potent naturally occurring nonnitrogenous kappa opioid selective agonist. Proc Natl Acad Sci U S A, 2002. 99(18): p. 11934-9. 41. Gaddum, J.H. and K.A. Hameed, Drugs which antagonize 5- hydroxytryptamine. Br J Pharmacol Chemother, 1954. 9(2): p. 240-8.

125

42. Woolley, D.W. and E. Shaw, A Biochemical and Pharmacological Suggestion About Certain Mental Disorders. Proc Natl Acad Sci U S A, 1954. 40(4): p. 228-31. 43. Shaw, E. and D.W. Woolley, Some serotoninlike activities of lysergic acid diethylamide. Science, 1956. 124(3212): p. 121-2. 44. Anden, N.E., H. Corrodi, K. Fuxe, and T. Hokfelt, Evidence for a central 5- hydroxytryptamine receptor stimulation by lysergic acid diethylamide. Br J Pharmacol, 1968. 34(1): p. 1-7. 45. Aghajanian, G.K., W.E. Foote, and M.H. Sheard, Lysergic acid diethylamide: sensitive neuronal units in the midbrain raphe. Science, 1968. 161(842): p. 706-8. 46. Aghajanian, G.K. and H.J. Hailgler, Hallucinogenic indoleamines: Preferential action upon presynaptic serotonin receptors. Psychopharmacol Commun, 1975. 1(6): p. 619-29. 47. Glennon, R.A., R. Young, and J.A. Rosecrans, Antagonism of the effects of the hallucinogen DOM and the purported 5-HT agonist quipazine by 5- HT2 antagonists. Eur J Pharmacol, 1983. 91(2-3): p. 189-96. 48. Glennon, R.A., M. Titeler, and J.D. McKenney, Evidence for 5-HT2 involvement in the mechanism of action of hallucinogenic agents. Life Sci, 1984. 35(25): p. 2505-11. 49. Ismaiel, A.M., J. De Los Angeles, M. Teitler, S. Ingher, and R.A. Glennon, Antagonism of 1-(2,5-dimethoxy-4-methylphenyl)-2-aminopropane stimulus with a newly identified 5-HT2- versus 5-HT1C-selective antagonist. J Med Chem, 1993. 36(17): p. 2519-25. 50. Schreiber, R., M. Brocco, and M.J. Millan, Blockade of the discriminative stimulus effects of DOI by MDL 100,907 and the 'atypical' antipsychotics, clozapine and risperidone. Eur J Pharmacol, 1994. 264(1): p. 99-102. 51. Roth, B.L., T. Nakaki, D.M. Chuang, and E. Costa, Aortic recognition sites for serotonin (5HT) are coupled to phospholipase C and modulate phosphatidylinositol turnover. Neuropharmacology, 1984. 23(10): p. 1223- 5. 52. Roth, B.L., T. Nakaki, D.M. Chuang, and E. Costa, 5-Hydroxytryptamine2 receptors coupled to phospholipase C in rat aorta: modulation of phosphoinositide turnover by phorbol ester. J Pharmacol Exp Ther, 1986. 238(2): p. 480-5. 53. Kagaya, A., M. Mikuni, I. Kusumi, H. Yamamoto, and K. Takahashi, Serotonin-induced acute desensitization of serotonin2 receptors in human platelets via a mechanism involving protein kinase C. J Pharmacol Exp Ther, 1990. 255(1): p. 305-11. 54. Egan, C.T., K. Herrick-Davis, K. Miller, R.A. Glennon, and M. Teitler, Agonist activity of LSD and lisuride at cloned 5HT2A and 5HT2C receptors. Psychopharmacology (Berl), 1998. 136(4): p. 409-14. 55. Roth, B.L., M.S. Choudhary, N. Khan, and A.Z. Uluer, High-affinity agonist binding is not sufficient for agonist efficacy at 5-hydroxytryptamine2A receptors: evidence in favor of a modified ternary complex model. J Pharmacol Exp Ther, 1997. 280(2): p. 576-83.

126

56. Felder, C.C., R.Y. Kanterman, A.L. Ma, and J. Axelrod, Serotonin stimulates phospholipase A2 and the release of arachidonic acid in hippocampal neurons by a type 2 serotonin receptor that is independent of inositolphospholipid hydrolysis. Proc Natl Acad Sci U S A, 1990. 87(6): p. 2187-91. 57. Gonzalez-Maeso, J., N.V. Weisstaub, M. Zhou, P. Chan, L. Ivic, R. Ang, A. Lira, M. Bradley-Moore, Y. Ge, Q. Zhou, S.C. Sealfon, and J.A. Gingrich, Hallucinogens recruit specific cortical 5-HT(2A) receptor- mediated signaling pathways to affect behavior. Neuron, 2007. 53(3): p. 439-52. 58. Kurrasch-Orbaugh, D.M., J.C. Parrish, V.J. Watts, and D.E. Nichols, A complex signaling cascade links the serotonin2A receptor to phospholipase A2 activation: the involvement of MAP kinases. J Neurochem, 2003. 86(4): p. 980-91. 59. Johnson-Farley, N.N., S.B. Kertesy, G.R. Dubyak, and D.S. Cowen, Enhanced activation of Akt and extracellular-regulated kinase pathways by simultaneous occupancy of Gq-coupled 5-HT2A receptors and Gs- coupled 5-HT7A receptors in PC12 cells. J Neurochem, 2005. 92(1): p. 72-82. 60. Hershenson, M.B., T.S. Chao, M.K. Abe, I. Gomes, M.D. Kelleher, J. Solway, and M.R. Rosner, Histamine antagonizes serotonin and growth factor-induced mitogen-activated protein kinase activation in bovine tracheal smooth muscle cells. J Biol Chem, 1995. 270(34): p. 19908-13. 61. Quinn, J.C., N.N. Johnson-Farley, J. Yoon, and D.S. Cowen, Activation of extracellular-regulated kinase by 5-hydroxytryptamine(2A) receptors in PC12 cells is protein kinase C-independent and requires calmodulin and tyrosine kinases. J Pharmacol Exp Ther, 2002. 303(2): p. 746-52. 62. Gooz, M., P. Gooz, L.M. Luttrell, and J.R. Raymond, 5-HT2A receptor induces ERK phosphorylation and proliferation through ADAM-17 tumor necrosis factor-alpha-converting enzyme (TACE) activation and heparin- bound epidermal growth factor-like growth factor (HB-EGF) shedding in mesangial cells. J Biol Chem, 2006. 281(30): p. 21004-12. 63. Greene, E.L., O. Houghton, G. Collinsworth, M.N. Garnovskaya, T. Nagai, T. Sajjad, V. Bheemanathini, J.S. Grewal, R.V. Paul, and J.R. Raymond, 5-HT(2A) receptors stimulate mitogen-activated protein kinase via H(2)O(2) generation in rat renal mesangial cells. Am J Physiol Renal Physiol, 2000. 278(4): p. F650-8. 64. Robertson, D.N., M.S. Johnson, L.O. Moggach, P.J. Holland, E.M. Lutz, and R. Mitchell, Selective interaction of ARF1 with the carboxy-terminal tail domain of the 5-HT2A receptor. Mol Pharmacol, 2003. 64(5): p. 1239- 50. 65. Mitchell, R., D. McCulloch, E. Lutz, M. Johnson, C. MacKenzie, M. Fennell, G. Fink, W. Zhou, and S.C. Sealfon, Rhodopsin-family receptors associate with small G proteins to activate phospholipase D. Nature, 1998. 392(6674): p. 411-4.

127

66. Regina, M.J., R.C. Bucelli, J.C. Winter, and R.A. Rabin, Cellular mechanisms of serotonin 5-HT2A receptor-mediated cGMP formation: the essential role of glutamate. Brain Res, 2004. 1003(1-2): p. 168-75. 67. Garcia, E.E., R.L. Smith, and E. Sanders-Bush, Role of G(q) protein in behavioral effects of the hallucinogenic drug 1-(2,5-dimethoxy-4- iodophenyl)-2-aminopropane. Neuropharmacology, 2007. 52(8): p. 1671- 7. 68. Nichols, C.D. and E. Sanders-Bush, A single dose of lysergic acid diethylamide influences gene expression patterns within the mammalian brain. Neuropsychopharmacology, 2002. 26(5): p. 634-42. 69. Gonzalez-Maeso, J., T. Yuen, B.J. Ebersole, E. Wurmbach, A. Lira, M. Zhou, N. Weisstaub, R. Hen, J.A. Gingrich, and S.C. Sealfon, Transcriptome fingerprints distinguish hallucinogenic and nonhallucinogenic 5-hydroxytryptamine 2A receptor agonist effects in mouse somatosensory cortex. J Neurosci, 2003. 23(26): p. 8836-43. 70. McCudden, C.R., M.D. Hains, R.J. Kimple, D.P. Siderovski, and F.S. Willard, G-protein signaling: back to the future. Cell Mol Life Sci, 2005. 62(5): p. 551-77. 71. Boyer, J.L., G.L. Waldo, and T.K. Harden, Beta gamma-subunit activation of G-protein-regulated phospholipase C. J Biol Chem, 1992. 267(35): p. 25451-6. 72. Tononi, G. and G.M. Edelman, Consciousness and complexity. Science, 1998. 282(5395): p. 1846-51. 73. Laberge, D. and R. Kasevich, The apical dendrite theory of consciousness. Neural Netw, 2007. 20(9): p. 1004-1020. 74. Aghajanian, G.K. and G.J. Marek, Serotonin and hallucinogens. Neuropsychopharmacology, 1999. 21(2 Suppl): p. 16S-23S. 75. Marek, G.J., R.A. Wright, J.C. Gewirtz, and D.D. Schoepp, A major role for thalamocortical afferents in serotonergic hallucinogen receptor function in the rat neocortex. Neuroscience, 2001. 105(2): p. 379-92. 76. Scruggs, J.L., S. Patel, M. Bubser, and A.Y. Deutch, DOI-Induced activation of the cortex: dependence on 5-HT2A heteroceptors on thalamocortical glutamatergic neurons. J Neurosci, 2000. 20(23): p. 8846- 52. 77. Serra Catafau, J., F. Rubio, and J. Peres Serra, Peduncular hallucinosis associated with posterior thalamic infarction. J Neurol, 1992. 239(2): p. 89- 90. 78. Noda, S., M. Mizoguchi, and A. Yamamoto, Thalamic experiential hallucinosis. J Neurol Neurosurg Psychiatry, 1993. 56(11): p. 1224-6. 79. Aghajanian, G.K. and G.J. Marek, Serotonin model of schizophrenia: emerging role of glutamate mechanisms. Brain Res Brain Res Rev, 2000. 31(2-3): p. 302-12. 80. Celada, P., M.V. Puig, L. Diaz-Mataix, and F. Artigas, The Hallucinogen DOI Reduces Low-Frequency Oscillations in Rat Prefrontal Cortex: Reversal by Antipsychotic Drugs. Biol Psychiatry, 2008.

128

81. Puig, M.V., P. Celada, L. Diaz-Mataix, and F. Artigas, In vivo modulation of the activity of pyramidal neurons in the rat medial prefrontal cortex by 5- HT2A receptors: relationship to thalamocortical afferents. Cereb Cortex, 2003. 13(8): p. 870-82. 82. Manford, M. and F. Andermann, Complex visual hallucinations. Clinical and neurobiological insights. Brain, 1998. 121 ( Pt 10): p. 1819-40. 83. Behrendt, R.P. and C. Young, Hallucinations in schizophrenia, sensory impairment, and brain disease: a unifying model. Behav Brain Sci, 2004. 27(6): p. 771-87; discussion 787-830. 84. Xia, Z., J.A. Gray, B.A. Compton-Toth, and B.L. Roth, A direct interaction of PSD-95 with 5-HT2A serotonin receptors regulates receptor trafficking and signal transduction. J Biol Chem, 2003. 278(24): p. 21901-8. 85. Becamel, C., S. Gavarini, B. Chanrion, G. Alonso, N. Galeotti, A. Dumuis, J. Bockaert, and P. Marin, The serotonin 5-HT2A and 5-HT2C receptors interact with specific sets of PDZ proteins. J Biol Chem, 2004. 279(19): p. 20257-66. 86. Kim, E. and M. Sheng, PDZ domain proteins of synapses. Nat Rev Neurosci, 2004. 5(10): p. 771-81. 87. Oschkinat, H., A new type of PDZ domain recognition. Nat Struct Biol, 1999. 6(5): p. 408-10. 88. Doyle, D.A., A. Lee, J. Lewis, E. Kim, M. Sheng, and R. MacKinnon, Crystal structures of a complexed and peptide-free membrane protein- binding domain: molecular basis of peptide recognition by PDZ. Cell, 1996. 85(7): p. 1067-76. 89. Harris, B.Z. and W.A. Lim, Mechanism and role of PDZ domains in signaling complex assembly. J Cell Sci, 2001. 114(Pt 18): p. 3219-31. 90. Becamel, C., G. Alonso, N. Galeotti, E. Demey, P. Jouin, C. Ullmer, A. Dumuis, J. Bockaert, and P. Marin, Synaptic multiprotein complexes associated with 5-HT(2C) receptors: a proteomic approach. Embo J, 2002. 21(10): p. 2332-42. 91. Sheng, M. and C. Sala, PDZ domains and the organization of supramolecular complexes. Annu Rev Neurosci, 2001. 24: p. 1-29. 92. Craven, S.E., A.E. El-Husseini, and D.S. Bredt, Synaptic targeting of the postsynaptic density protein PSD-95 mediated by lipid and protein motifs. Neuron, 1999. 22(3): p. 497-509. 93. Lim, I.A., D.D. Hall, and J.W. Hell, Selectivity and promiscuity of the first and second PDZ domains of PSD-95 and synapse-associated protein 102. J Biol Chem, 2002. 277(24): p. 21697-711. 94. Hering, H. and M. Sheng, Direct interaction of Frizzled-1, -2, -4, and -7 with PDZ domains of PSD-95. FEBS Lett, 2002. 521(1-3): p. 185-9. 95. DeMarco, S.J. and E.E. Strehler, Plasma membrane Ca2+-atpase isoforms 2b and 4b interact promiscuously and selectively with members of the membrane-associated guanylate kinase family of PDZ (PSD95/Dlg/ZO-1) domain-containing proteins. J Biol Chem, 2001. 276(24): p. 21594-600.

129

96. Burkhardt, C., M. Muller, A. Badde, C.C. Garner, E.D. Gundelfinger, and A.W. Puschel, Semaphorin 4B interacts with the post-synaptic density protein PSD-95/SAP90 and is recruited to synapses through a C-terminal PDZ-binding motif. FEBS Lett, 2005. 579(17): p. 3821-8. 97. Larsson, M., G. Hjalm, A.M. Sakwe, A. Engstrom, A.S. Hoglund, E. Larsson, R.C. Robinson, C. Sundberg, and L. Rask, Selective interaction of megalin with postsynaptic density-95 (PSD-95)-like membrane- associated guanylate kinase (MAGUK) proteins. Biochem J, 2003. 373(Pt 2): p. 381-91. 98. Sun, Y., A. Savanenin, P.H. Reddy, and Y.F. Liu, Polyglutamine-expanded huntingtin promotes sensitization of N-methyl-D-aspartate receptors via post-synaptic density 95. J Biol Chem, 2001. 276(27): p. 24713-8. 99. Niethammer, M., J.G. Valtschanoff, T.M. Kapoor, D.W. Allison, T.M. Weinberg, A.M. Craig, and M. Sheng, CRIPT, a novel postsynaptic protein that binds to the third PDZ domain of PSD-95/SAP90. Neuron, 1998. 20(4): p. 693-707. 100. Yao, W.D., R.R. Gainetdinov, M.I. Arbuckle, T.D. Sotnikova, M. Cyr, J.M. Beaulieu, G.E. Torres, S.G. Grant, and M.G. Caron, Identification of PSD- 95 as a regulator of dopamine-mediated synaptic and behavioral plasticity. Neuron, 2004. 41(4): p. 625-38. 101. Kornau, H.C., L.T. Schenker, M.B. Kennedy, and P.H. Seeburg, Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science, 1995. 269(5231): p. 1737-40. 102. Furukawa, H., S.K. Singh, R. Mancusso, and E. Gouaux, Subunit arrangement and function in NMDA receptors. Nature, 2005. 438(7065): p. 185-92. 103. Gouaux, E., Structure and function of AMPA receptors. J Physiol, 2004. 554(Pt 2): p. 249-53. 104. Sheng, M. and E. Kim, The Shank family of scaffold proteins. J Cell Sci, 2000. 113 ( Pt 11): p. 1851-6. 105. Thomas, G.M., G.R. Rumbaugh, D.B. Harrar, and R.L. Huganir, Ribosomal S6 kinase 2 interacts with and phosphorylates PDZ domain- containing proteins and regulates AMPA receptor transmission. Proc Natl Acad Sci U S A, 2005. 102(42): p. 15006-11. 106. Sheng, M., Molecular organization of the postsynaptic specialization. Proc Natl Acad Sci U S A, 2001. 98(13): p. 7058-61. 107. Taylor, C.W., P.C. da Fonseca, and E.P. Morris, IP(3) receptors: the search for structure. Trends Biochem Sci, 2004. 29(4): p. 210-9. 108. Naisbitt, S., E. Kim, J.C. Tu, B. Xiao, C. Sala, J. Valtschanoff, R.J. Weinberg, P.F. Worley, and M. Sheng, Shank, a novel family of postsynaptic density proteins that binds to the NMDA receptor/PSD- 95/GKAP complex and cortactin. Neuron, 1999. 23(3): p. 569-82. 109. Hsueh, Y.P. and M. Sheng, Requirement of N-terminal cysteines of PSD- 95 for PSD-95 multimerization and ternary complex formation, but not for binding to potassium channel Kv1.4. J Biol Chem, 1999. 274(1): p. 532-6.

130

110. Romorini, S., G. Piccoli, M. Jiang, P. Grossano, N. Tonna, M. Passafaro, M. Zhang, and C. Sala, A functional role of postsynaptic density-95- guanylate kinase-associated protein complex in regulating Shank assembly and stability to synapses. J Neurosci, 2004. 24(42): p. 9391- 404. 111. Xiao, B., J.C. Tu, and P.F. Worley, Homer: a link between neural activity and glutamate receptor function. Curr Opin Neurobiol, 2000. 10(3): p. 370- 4. 112. Tu, J.C., B. Xiao, J.P. Yuan, A.A. Lanahan, K. Leoffert, M. Li, D.J. Linden, and P.F. Worley, Homer binds a novel proline-rich motif and links group 1 metabotropic glutamate receptors with IP3 receptors. Neuron, 1998. 21(4): p. 717-26. 113. Tu, J.C., B. Xiao, S. Naisbitt, J.P. Yuan, R.S. Petralia, P. Brakeman, A. Doan, V.K. Aakalu, A.A. Lanahan, M. Sheng, and P.F. Worley, Coupling of mGluR/Homer and PSD-95 complexes by the Shank family of postsynaptic density proteins. Neuron, 1999. 23(3): p. 583-92. 114. Kew, J.N. and J.A. Kemp, Ionotropic and metabotropic glutamate receptor structure and pharmacology. Psychopharmacology (Berl), 2005. 179(1): p. 4-29. 115. Hall, R.A., R.T. Premont, and R.J. Lefkowitz, Heptahelical receptor signaling: beyond the G protein paradigm. J Cell Biol, 1999. 145(5): p. 927-32. 116. Xia, Z., S.J. Hufeisen, J.A. Gray, and B.L. Roth, The PDZ-binding domain is essential for the dendritic targeting of 5-HT2A serotonin receptors in cortical pyramidal neurons in vitro. Neuroscience, 2003. 122(4): p. 907-20. 117. Gray, J.A. and B.L. Roth, The pipeline and future of drug development in schizophrenia. Mol Psychiatry, 2007. 12(10): p. 904-22. 118. Harrison, P.J. and D.R. Weinberger, Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence. Mol Psychiatry, 2005. 10(1): p. 40-68; image 5. 119. Walsh, T., J.M. McClellan, S.E. McCarthy, A.M. Addington, S.B. Pierce, G.M. Cooper, A.S. Nord, M. Kusenda, D. Malhotra, A. Bhandari, S.M. Stray, C.F. Rippey, P. Roccanova, V. Makarov, B. Lakshmi, R.L. Findling, L. Sikich, T. Stromberg, B. Merriman, N. Gogtay, P. Butler, K. Eckstrand, L. Noory, P. Gochman, R. Long, Z. Chen, S. Davis, C. Baker, E.E. Eichler, P.S. Meltzer, S.F. Nelson, A.B. Singleton, M.K. Lee, J.L. Rapoport, M.C. King, and J. Sebat, Rare structural variants disrupt multiple genes in neurodevelopmental pathways in schizophrenia. Science, 2008. 320(5875): p. 539-43. 120. Stone, J.L., M.C. O'Donovan, H. Gurling, G.K. Kirov, D.H. Blackwood, A. Corvin, N.J. Craddock, M. Gill, C.M. Hultman, P. Lichtenstein, A. McQuillin, C.N. Pato, D.M. Ruderfer, M.J. Owen, D. St Clair, P.F. Sullivan, P. Sklar, S.M. Purcell Leader, J. Korn, S. Macgregor, D.W. Morris, C.T. O'Dushlaine, M.J. Daly, P.M. Visscher, P.A. Holmans, S.M. Purcell, E.M. Scolnick, P. Sklar Leader, N.M. Williams, L. Georgieva, I. Nikolov, N. Norton, H. Williams, D. Toncheva, V. Milanova, E.F. Thelander, P.

131

Sullivan, E. Kenny, J.L. Waddington, K. Choudhury, S. Datta, J. Pimm, S. Thirumalai, V. Puri, R. Krasucki, J. Lawrence, D. Quested, N. Bass, D. Curtis, C. Crombie, G. Fraser, S. Leh Kwan, N. Walker, W.J. Muir, K.A. McGhee, B. Pickard, P. Malloy, A.W. Maclean, M. Van Beck, M.T. Pato, H. Medeiros, F. Middleton, C. Carvalho, C. Morley, A. Fanous, D. Conti, J.A. Knowles, C. Paz Ferreira, A. Macedo, M. Helena Azevedo, S.A. McCarroll, M. Daly, K. Chambert, C. Gates, S.B. Gabriel, S. Mahon, and K. Ardlie, Rare chromosomal deletions and duplications increase risk of schizophrenia. Nature, 2008. 121. Stefansson, H., D. Rujescu, S. Cichon, O.P. Pietilainen, A. Ingason, S. Steinberg, R. Fossdal, E. Sigurdsson, T. Sigmundsson, J.E. Buizer- Voskamp, T. Hansen, K.D. Jakobsen, P. Muglia, C. Francks, P.M. Matthews, A. Gylfason, B.V. Halldorsson, D. Gudbjartsson, T.E. Thorgeirsson, A. Sigurdsson, A. Jonasdottir, A. Bjornsson, S. Mattiasdottir, T. Blondal, M. Haraldsson, B.B. Magnusdottir, I. Giegling, H.J. Moller, A. Hartmann, K.V. Shianna, D. Ge, A.C. Need, C. Crombie, G. Fraser, N. Walker, J. Lonnqvist, J. Suvisaari, A. Tuulio-Henriksson, T. Paunio, T. Toulopoulou, E. Bramon, M. Di Forti, R. Murray, M. Ruggeri, E. Vassos, S. Tosato, M. Walshe, T. Li, C. Vasilescu, T.W. Muhleisen, A.G. Wang, H. Ullum, S. Djurovic, I. Melle, J. Olesen, L.A. Kiemeney, B. Franke, R.S. Kahn, D. Linszen, J. van Os, D. Wiersma, R. Bruggeman, W. Cahn, I. Germeys, L. de Haan, L. Krabbendam, C. Sabatti, N.B. Freimer, J.R. Gulcher, U. Thorsteinsdottir, A. Kong, O.A. Andreassen, R.A. Ophoff, A. Georgi, M. Rietschel, T. Werge, H. Petursson, D.B. Goldstein, M.M. Nothen, L. Peltonen, D.A. Collier, D. St Clair, and K. Stefansson, Large recurrent microdeletions associated with schizophrenia. Nature, 2008. 122. Willner, P., The dopamine hypothesis of schizophrenia: current status, future prospects. Int Clin Psychopharmacol, 1997. 12(6): p. 297-308. 123. Snyder, S.H., Catecholamines in the brain as mediators of amphetamine psychosis. Arch Gen Psychiatry, 1972. 27(2): p. 169-79. 124. Lieberman, J.A., J.M. Kane, and J. Alvir, Provocative tests with psychostimulant drugs in schizophrenia. Psychopharmacology (Berl), 1987. 91(4): p. 415-33. 125. Creese, I., D.R. Burt, and S.H. Snyder, Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs. Science, 1976. 192(4238): p. 481-3. 126. Roth, B.L., D.J. Sheffler, and W.K. Kroeze, Magic shotguns versus magic bullets: selectively non-selective drugs for mood disorders and schizophrenia. Nat Rev Drug Discov, 2004. 3(4): p. 353-9. 127. Snyder, S.H., Amphetamine psychosis: a "model" schizophrenia mediated by catecholamines. Am J Psychiatry, 1973. 130(1): p. 61-7. 128. Bell, D.S., Comparison of Amphetamine Psychosis and Schizophrenia. Br J Psychiatry, 1965. 111: p. 701-7. 129. Jentsch, J.D. and R.H. Roth, The neuropsychopharmacology of phencyclidine: from NMDA receptor hypofunction to the dopamine

132

hypothesis of schizophrenia. Neuropsychopharmacology, 1999. 20(3): p. 201-25. 130. Goldberg, T.E., L.B. Bigelow, D.R. Weinberger, D.G. Daniel, and J.E. Kleinman, Cognitive and behavioral effects of the coadministration of dextroamphetamine and haloperidol in schizophrenia. Am J Psychiatry, 1991. 148(1): p. 78-84. 131. Angrist, B., E. Peselow, M. Rubinstein, J. Corwin, and J. Rotrosen, Partial improvement in negative schizophrenic symptoms after amphetamine. Psychopharmacology (Berl), 1982. 78(2): p. 128-30. 132. Javitt, D.C. and S.R. Zukin, Recent advances in the phencyclidine model of schizophrenia. Am J Psychiatry, 1991. 148(10): p. 1301-8. 133. Steinpreis, R.E., The behavioral and neurochemical effects of phencyclidine in humans and animals: some implications for modeling psychosis. Behav Brain Res, 1996. 74(1-2): p. 45-55. 134. Luby, E.D., B.D. Cohen, G. Rosenbaum, J.S. Gottlieb, and R. Kelley, Study of a new schizophrenomimetic drug; sernyl. AMA Arch Neurol Psychiatry, 1959. 81(3): p. 363-9. 135. Lodge, D. and N.A. Anis, Effects of phencyclidine on excitatory amino acid activation of spinal interneurones in the cat. Eur J Pharmacol, 1982. 77(2- 3): p. 203-4. 136. Olney, J.W., J.W. Newcomer, and N.B. Farber, NMDA receptor hypofunction model of schizophrenia. J Psychiatr Res, 1999. 33(6): p. 523-33. 137. Geyer, M.A. and B. Ellenbroek, Animal behavior models of the mechanisms underlying antipsychotic atypicality. Prog Neuropsychopharmacol Biol Psychiatry, 2003. 27(7): p. 1071-9. 138. Geyer, M.A., K. Krebs-Thomson, D.L. Braff, and N.R. Swerdlow, Pharmacological studies of prepulse inhibition models of sensorimotor gating deficits in schizophrenia: a decade in review. Psychopharmacology (Berl), 2001. 156(2-3): p. 117-54. 139. Linn, G.S. and D.C. Javitt, Phencyclidine (PCP)-induced deficits of prepulse inhibition in monkeys. Neuroreport, 2001. 12(1): p. 117-20. 140. Swerdlow, N.R., D.L. Braff, and M.A. Geyer, Animal models of deficient sensorimotor gating: what we know, what we think we know, and what we hope to know soon. Behav Pharmacol, 2000. 11(3-4): p. 185-204. 141. McGhie, A. and J. Chapman, Disorders of attention and perception in early schizophrenia. Br J Med Psychol, 1961. 34: p. 103-16. 142. Braff, D.L., M.A. Geyer, and N.R. Swerdlow, Human studies of prepulse inhibition of startle: normal subjects, patient groups, and pharmacological studies. Psychopharmacology (Berl), 2001. 156(2-3): p. 234-58. 143. Braff, D., C. Stone, E. Callaway, M. Geyer, I. Glick, and L. Bali, Prestimulus effects on human startle reflex in normals and schizophrenics. Psychophysiology, 1978. 15(4): p. 339-43. 144. Swerdlow, N.R., D.L. Braff, N. Taaid, and M.A. Geyer, Assessing the validity of an animal model of deficient sensorimotor gating in schizophrenic patients. Arch Gen Psychiatry, 1994. 51(2): p. 139-54.

133

145. Linn, G.S., S.S. Negi, S.V. Gerum, and D.C. Javitt, Reversal of phencyclidine-induced prepulse inhibition deficits by clozapine in monkeys. Psychopharmacology (Berl), 2003. 169(3-4): p. 234-9. 146. Yamada, S., M. Harano, N. Annoh, K. Nakamura, and M. Tanaka, Involvement of serotonin 2A receptors in phencyclidine-induced disruption of prepulse inhibition of the acoustic startle in rats. Biol Psychiatry, 1999. 46(6): p. 832-8. 147. Meltzer, H.Y., S. Matsubara, and J.C. Lee, Classification of typical and atypical antipsychotic drugs on the basis of dopamine D-1, D-2 and serotonin2 pKi values. J Pharmacol Exp Ther, 1989. 251(1): p. 238-46. 148. Remington, G., Understanding antipsychotic "atypicality": a clinical and pharmacological moving target. J Psychiatry Neurosci, 2003. 28(4): p. 275-84. 149. Matz, R., W. Rick, H. Thompson, and S. Gershon, Clozapine--a potential antipsychotic agent without extrapyramidal manifestations. Curr Ther Res Clin Exp, 1974. 16(7): p. 687-95. 150. Matz, R., W. Rick, D. Oh, H. Thompson, and S. Gershon, Clozapine--a potential antipsychotic agent without extrapyramidal manifestations. Psychopharmacol Bull, 1975. 11(1): p. 14. 151. Fink, H., R. Morgenstern, and W. Oelssner, Clozapine--a serotonin antagonist? Pharmacol Biochem Behav, 1984. 20(4): p. 513-7. 152. Reynolds, G.P., N.J. Garrett, N. Rupniak, P. Jenner, and C.D. Marsden, Chronic clozapine treatment of rats down-regulates cortical 5-HT2 receptors. Eur J Pharmacol, 1983. 89(3-4): p. 325-6. 153. Meltzer, H.Y. and G. Okayli, Reduction of suicidality during clozapine treatment of neuroleptic-resistant schizophrenia: impact on risk-benefit assessment. Am J Psychiatry, 1995. 152(2): p. 183-90. 154. Meltzer, H.Y. and S.R. McGurk, The effects of clozapine, risperidone, and olanzapine on cognitive function in schizophrenia. Schizophr Bull, 1999. 25(2): p. 233-55. 155. Kapur, S. and G. Remington, Atypical antipsychotics: new directions and new challenges in the treatment of schizophrenia. Annu Rev Med, 2001. 52: p. 503-17. 156. Varty, G.B., V.P. Bakshi, and M.A. Geyer, M100907, a serotonin 5-HT2A receptor antagonist and putative antipsychotic, blocks dizocilpine-induced prepulse inhibition deficits in Sprague-Dawley and Wistar rats. Neuropsychopharmacology, 1999. 20(4): p. 311-21. 157. Maurel-Remy, S., K. Bervoets, and M.J. Millan, Blockade of phencyclidine- induced hyperlocomotion by clozapine and MDL 100,907 in rats reflects antagonism of 5-HT2A receptors. Eur J Pharmacol, 1995. 280(2): p. R9- 11. 158. Pazos, A., D. Hoyer, and J.M. Palacios, The binding of serotonergic ligands to the porcine choroid plexus: characterization of a new type of serotonin recognition site. Eur J Pharmacol, 1984. 106(3): p. 539-46. 159. Pompeiano, M., J.M. Palacios, and G. Mengod, Distribution of the serotonin 5-HT2 receptor family mRNAs: comparison between 5-HT2A

134

and 5-HT2C receptors. Brain Res Mol Brain Res, 1994. 23(1-2): p. 163- 78. 160. Molineaux, S.M., T.M. Jessell, R. Axel, and D. Julius, 5-HT1c receptor is a prominent serotonin receptor subtype in the central nervous system. Proc Natl Acad Sci U S A, 1989. 86(17): p. 6793-7. 161. Clemett, D.A., T. Punhani, M.S. Duxon, T.P. Blackburn, and K.C. Fone, Immunohistochemical localisation of the 5-HT2C receptor protein in the rat CNS. Neuropharmacology, 2000. 39(1): p. 123-32. 162. Lopez-Gimenez, J.F., L.H. Tecott, J.M. Palacios, G. Mengod, and M.T. Vilaro, Serotonin 5- HT (2C) receptor knockout mice: autoradiographic analysis of multiple serotonin receptors. J Neurosci Res, 2002. 67(1): p. 69-85. 163. Conn, P.J., E. Sanders-Bush, B.J. Hoffman, and P.R. Hartig, A unique serotonin receptor in choroid plexus is linked to phosphatidylinositol turnover. Proc Natl Acad Sci U S A, 1986. 83(11): p. 4086-8. 164. Chang, M., L. Zhang, J.P. Tam, and E. Sanders-Bush, Dissecting G protein-coupled receptor signaling pathways with membrane-permeable blocking peptides. Endogenous 5-HT(2C) receptors in choroid plexus epithelial cells. J Biol Chem, 2000. 275(10): p. 7021-9. 165. McGrew, L., M.S. Chang, and E. Sanders-Bush, Phospholipase D activation by endogenous 5-hydroxytryptamine 2C receptors is mediated by Galpha13 and pertussis toxin-insensitive Gbetagamma subunits. Mol Pharmacol, 2002. 62(6): p. 1339-43. 166. Werry, T.D., K.J. Gregory, P.M. Sexton, and A. Christopoulos, Characterization of serotonin 5-HT2C receptor signaling to extracellular signal-regulated kinases 1 and 2. J Neurochem, 2005. 93(6): p. 1603-15. 167. Cussac, D., A. Newman-Tancredi, D. Duqueyroix, V. Pasteau, and M.J. Millan, Differential activation of Gq/11 and Gi(3) proteins at 5- hydroxytryptamine(2C) receptors revealed by antibody capture assays: influence of receptor reserve and relationship to agonist-directed trafficking. Mol Pharmacol, 2002. 62(3): p. 578-89. 168. Kaufman, M.J., P.R. Hartig, and B.J. Hoffman, Serotonin 5-HT2C receptor stimulates cyclic GMP formation in choroid plexus. J Neurochem, 1995. 64(1): p. 199-205. 169. Burns, C.M., H. Chu, S.M. Rueter, L.K. Hutchinson, H. Canton, E. Sanders-Bush, and R.B. Emeson, Regulation of serotonin-2C receptor G- protein coupling by RNA editing. Nature, 1997. 387(6630): p. 303-8. 170. Werry, T.D., R. Loiacono, P.M. Sexton, and A. Christopoulos, RNA editing of the serotonin 5HT(2C) receptor and its effects on cell signalling, pharmacology and brain function. Pharmacol Ther, 2008. 119(1): p. 7-23. 171. Du, Y., M.T. Davisson, K. Kafadar, and K. Gardiner, A-to-I pre-mRNA editing of the serotonin 2C receptor: comparisons among inbred mouse strains. Gene, 2006. 382: p. 39-46. 172. Niswender, C.M., S.C. Copeland, K. Herrick-Davis, R.B. Emeson, and E. Sanders-Bush, RNA editing of the human serotonin 5-hydroxytryptamine

135

2C receptor silences constitutive activity. J Biol Chem, 1999. 274(14): p. 9472-8. 173. Price, R.D. and E. Sanders-Bush, RNA editing of the human serotonin 5- HT(2C) receptor delays agonist-stimulated calcium release. Mol Pharmacol, 2000. 58(4): p. 859-62. 174. Price, R.D., D.M. Weiner, M.S. Chang, and E. Sanders-Bush, RNA editing of the human serotonin 5-HT2C receptor alters receptor-mediated activation of G13 protein. J Biol Chem, 2001. 276(48): p. 44663-8. 175. McGrew, L., R.D. Price, E. Hackler, M.S. Chang, and E. Sanders-Bush, RNA editing of the human serotonin 5-HT2C receptor disrupts transactivation of the small G-protein RhoA. Mol Pharmacol, 2004. 65(1): p. 252-6. 176. Alex, K.D., G.J. Yavanian, H.G. McFarlane, C.P. Pluto, and E.A. Pehek, Modulation of dopamine release by striatal 5-HT2C receptors. Synapse, 2005. 55(4): p. 242-51. 177. Prisco, S., S. Pagannone, and E. Esposito, Serotonin-dopamine interaction in the rat ventral tegmental area: an electrophysiological study in vivo. J Pharmacol Exp Ther, 1994. 271(1): p. 83-90. 178. Di Giovanni, G., V. Di Matteo, V. La Grutta, and E. Esposito, m- Chlorophenylpiperazine excites non-dopaminergic neurons in the rat substantia nigra and ventral tegmental area by activating serotonin-2C receptors. Neuroscience, 2001. 103(1): p. 111-6. 179. Di Giovanni, G., V. Di Matteo, M. Di Mascio, and E. Esposito, Preferential modulation of mesolimbic vs. nigrostriatal dopaminergic function by serotonin(2C/2B) receptor agonists: a combined in vivo electrophysiological and microdialysis study. Synapse, 2000. 35(1): p. 53- 61. 180. Tecott, L.H., S.F. Logue, J.M. Wehner, and J.A. Kauer, Perturbed dentate gyrus function in serotonin 5-HT2C receptor mutant mice. Proc Natl Acad Sci U S A, 1998. 95(25): p. 15026-31. 181. Tecott, L.H., L.M. Sun, S.F. Akana, A.M. Strack, D.H. Lowenstein, M.F. Dallman, and D. Julius, Eating disorder and epilepsy in mice lacking 5- HT2c serotonin receptors. Nature, 1995. 374(6522): p. 542-6. 182. Wieraszko, A. and T.N. Seyfried, Influence of audiogenic seizures on synaptic facilitation in mouse hippocampal slices is mediated by N-methyl- D-aspartate receptor. Epilepsia, 1993. 34(6): p. 979-84. 183. Pereira, M.G., D.L. Gitai, M.L. Paco-Larson, J.B. Pesquero, N. Garcia- Cairasco, and C.M. Costa-Neto, Modulation of B1 and B2 kinin receptors expression levels in the hippocampus of rats after audiogenic kindling and with limbic recruitment, a model of temporal lobe epilepsy. Int Immunopharmacol, 2008. 8(2): p. 200-5. 184. Hajos, M., W.E. Hoffmann, and R.J. Weaver, Regulation of septo- hippocampal activity by 5-hydroxytryptamine(2C) receptors. J Pharmacol Exp Ther, 2003. 306(2): p. 605-15. 185. Buzsaki, G., Theta oscillations in the hippocampus. Neuron, 2002. 33(3): p. 325-40.

136

186. Dunlop, J., A.L. Sabb, H. Mazandarani, J. Zhang, S. Kalgaonker, E. Shukhina, S. Sukoff, R.L. Vogel, G. Stack, L. Schechter, B.L. Harrison, and S. Rosenzweig-Lipson, WAY-163909 [(7bR, 10aR)- 1,2,3,4,8,9,10,10a-octahydro-7bH-cyclopenta-[b][1,4]diazepino[6,7,1h i]indole], a novel 5-hydroxytryptamine 2C receptor-selective agonist with anorectic activity. J Pharmacol Exp Ther, 2005. 313(2): p. 862-9. 187. Marquis, K.L., A.L. Sabb, S.F. Logue, J.A. Brennan, M.J. Piesla, T.A. Comery, S.M. Grauer, C.R. Ashby, Jr., H.Q. Nguyen, L.A. Dawson, J.E. Barrett, G. Stack, H.Y. Meltzer, B.L. Harrison, and S. Rosenzweig-Lipson, WAY-163909 [(7bR,10aR)-1,2,3,4,8,9,10,10a-octahydro-7bH-cyclopenta- [b][1,4]diazepino[ 6,7,1hi]indole]: A novel 5-hydroxytryptamine 2C receptor-selective agonist with preclinical antipsychotic-like activity. J Pharmacol Exp Ther, 2007. 320(1): p. 486-96. 188. Dunlop, J., K.L. Marquis, H.K. Lim, L. Leung, J. Kao, C. Cheesman, and S. Rosenzweig-Lipson, Pharmacological profile of the 5-HT(2C) receptor agonist WAY-163909; therapeutic potential in multiple indications. CNS Drug Rev, 2006. 12(3-4): p. 167-77. 189. Sard, H., G. Kumaran, C. Morency, B.L. Roth, B.A. Toth, P. He, and L. Shuster, SAR of psilocybin analogs: discovery of a selective 5-HT 2C agonist. Bioorg Med Chem Lett, 2005. 15(20): p. 4555-9. 190. Thomsen, W.J., A.J. Grottick, F. Menzaghi, H. Reyes-Saldana, S. Espitia, D. Yuskin, K. Whelan, M. Martin, M. Morgan, W. Chen, H. Al-Shama, B. Smith, D. Chalmers, and D. Behan, Lorcaserin, A Novel Selective Human 5-HT2C Agonist: In Vitro and In Vivo Pharmacological Characterization. J Pharmacol Exp Ther, 2008. 191. Imamura, F., S. Maeda, T. Doi, and Y. Fujiyoshi, Ligand binding of the second PDZ domain regulates clustering of PSD-95 with the Kv1.4 potassium channel. J Biol Chem, 2002. 277(5): p. 3640-6. 192. Pinto, W. and G. Battaglia, Comparative recovery kinetics of 5- hydroxytryptamine 1A, 1B, and 2A receptor subtypes in rat cortex after receptor inactivation: evidence for differences in receptor production and degradation. Mol Pharmacol, 1994. 46(6): p. 1111-9. 193. Ahlemeyer, B. and E. Baumgart-Vogt, Optimized protocols for the simultaneous preparation of primary neuronal cultures of the neocortex, hippocampus and cerebellum from individual newborn (P0.5) C57Bl/6J mice. J Neurosci Methods, 2005. 149(2): p. 110-20. 194. Lois, C., E.J. Hong, S. Pease, E.J. Brown, and D. Baltimore, Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors. Science, 2002. 295(5556): p. 868-72. 195. Gavarini, S., C. Becamel, C. Altier, P. Lory, J. Poncet, J. Wijnholds, J. Bockaert, and P. Marin, Opposite effects of PSD-95 and MPP3 PDZ proteins on serotonin 5-hydroxytryptamine2C receptor desensitization and membrane stability. Mol Biol Cell, 2006. 17(11): p. 4619-31. 196. Caceres, A., G. Banker, O. Steward, L. Binder, and M. Payne, MAP2 is localized to the dendrites of hippocampal neurons which develop in culture. Brain Res, 1984. 315(2): p. 314-8.

137

197. Rauser, L., J.E. Savage, H.Y. Meltzer, and B.L. Roth, Inverse agonist actions of typical and atypical antipsychotic drugs at the human 5- hydroxytryptamine(2C) receptor. J Pharmacol Exp Ther, 2001. 299(1): p. 83-9. 198. Roth, B.L., E. Lopez, S. Beischel, R.B. Westkaemper, and J.M. Evans, Screening the receptorome to discover the molecular targets for plant- derived psychoactive compounds: a novel approach for CNS drug discovery. Pharmacol Ther, 2004. 102(2): p. 99-110. 199. LaBerge, D., Apical dendrite activity in cognition and consciousness. Conscious Cogn, 2006. 15(2): p. 235-57. 200. Spratling, M.W., Cortical region interactions and the functional role of apical dendrites. Behav Cogn Neurosci Rev, 2002. 1(3): p. 219-28. 201. Javitt, D.C., Glutamate as a therapeutic target in psychiatric disorders. Mol Psychiatry, 2004. 9(11): p. 984-97, 979. 202. Carlsson, M.L., P. Martin, M. Nilsson, S.M. Sorensen, A. Carlsson, S. Waters, and N. Waters, The 5-HT2A receptor antagonist M100907 is more effective in counteracting NMDA antagonist- than dopamine agonist- induced hyperactivity in mice. J Neural Transm, 1999. 106(2): p. 123-9. 203. Migaud, M., P. Charlesworth, M. Dempster, L.C. Webster, A.M. Watabe, M. Makhinson, Y. He, M.F. Ramsay, R.G. Morris, J.H. Morrison, T.J. O'Dell, and S.G. Grant, Enhanced long-term potentiation and impaired learning in mice with mutant postsynaptic density-95 protein. Nature, 1998. 396(6710): p. 433-9. 204. Lo, R.K. and Y.H. Wong, Transcriptional activation of c-Fos by constitutively active Galpha(16)QL through a STAT1-dependent pathway. Cell Signal, 2006. 18(12): p. 2143-53. 205. Chaudhuri, A., Neural activity mapping with inducible transcription factors. Neuroreport, 1997. 8(13): p. iii-vii. 206. Campbell, B.M. and K.M. Merchant, Serotonin 2C receptors within the basolateral amygdala induce acute fear-like responses in an open-field environment. Brain Res, 2003. 993(1-2): p. 1-9. 207. Van Oekelen, D., W.H. Luyten, and J.E. Leysen, 5-HT2A and 5-HT2C receptors and their atypical regulation properties. Life Sci, 2003. 72(22): p. 2429-49. 208. Smith, B.M., J.M. Smith, J.H. Tsai, J.A. Schultz, C.A. Gilson, S.A. Estrada, R.R. Chen, D.M. Park, E.B. Prieto, C.S. Gallardo, D. Sengupta, P.I. Dosa, J.A. Covel, A. Ren, R.R. Webb, N.R. Beeley, M. Martin, M. Morgan, S. Espitia, H.R. Saldana, C. Bjenning, K.T. Whelan, A.J. Grottick, F. Menzaghi, and W.J. Thomsen, Discovery and structure-activity relationship of (1R)-8-chloro-2,3,4,5-tetrahydro-1-methyl-1H-3- benzazepine (Lorcaserin), a selective serotonin 5-HT2C receptor agonist for the treatment of obesity. J Med Chem, 2008. 51(2): p. 305-13. 209. Tsaltas, E., D. Kontis, S. Chrysikakou, H. Giannou, A. Biba, S. Pallidi, A. Christodoulou, A. Maillis, and A. Rabavilas, Reinforced spatial alternation as an animal model of obsessive-compulsive disorder (OCD):

138

investigation of 5-HT2C and 5-HT1D receptor involvement in OCD pathophysiology. Biol Psychiatry, 2005. 57(10): p. 1176-85. 210. Tsui, J. and R.C. Malenka, Substrate localization creates specificity in calcium/calmodulin-dependent protein kinase II signaling at synapses. J Biol Chem, 2006. 281(19): p. 13794-804. 211. Armbruster, B.N., X. Li, M.H. Pausch, S. Herlitze, and B.L. Roth, Evolving the lock to fit the key to create a family of G protein-coupled receptors potently activated by an inert ligand. Proc Natl Acad Sci U S A, 2007. 104(12): p. 5163-8. 212. Heintz, N., BAC to the future: the use of bac transgenic mice for neuroscience research. Nat Rev Neurosci, 2001. 2(12): p. 861-70. 213. Valenti, O., P.J. Conn, and M.J. Marino, Distinct physiological roles of the Gq-coupled metabotropic glutamate receptors Co-expressed in the same neuronal populations. J Cell Physiol, 2002. 191(2): p. 125-37. 214. Skeberdis, V.A., J. Lan, T. Opitz, X. Zheng, M.V. Bennett, and R.S. Zukin, mGluR1-mediated potentiation of NMDA receptors involves a rise in intracellular calcium and activation of protein kinase C. Neuropharmacology, 2001. 40(7): p. 856-65. 215. Ugolini, A., M. Corsi, and F. Bordi, Potentiation of NMDA and AMPA responses by the specific mGluR5 agonist CHPG in spinal cord motoneurons. Neuropharmacology, 1999. 38(10): p. 1569-76. 216. Awad, H., G.W. Hubert, Y. Smith, A.I. Levey, and P.J. Conn, Activation of metabotropic glutamate receptor 5 has direct excitatory effects and potentiates NMDA receptor currents in neurons of the subthalamic nucleus. J Neurosci, 2000. 20(21): p. 7871-9. 217. Brody, S.A., F. Conquet, and M.A. Geyer, Disruption of prepulse inhibition in mice lacking mGluR1. Eur J Neurosci, 2003. 18(12): p. 3361-6. 218. Brody, S.A., S.C. Dulawa, F. Conquet, and M.A. Geyer, Assessment of a prepulse inhibition deficit in a mutant mouse lacking mGlu5 receptors. Mol Psychiatry, 2004. 9(1): p. 35-41. 219. Henry, S.A., V. Lehmann-Masten, F. Gasparini, M.A. Geyer, and A. Markou, The mGluR5 antagonist MPEP, but not the mGluR2/3 agonist LY314582, augments PCP effects on prepulse inhibition and locomotor activity. Neuropharmacology, 2002. 43(8): p. 1199-209. 220. Pietraszek, M., A. Gravius, D. Schafer, T. Weil, D. Trifanova, and W. Danysz, mGluR5, but not mGluR1, antagonist modifies MK-801-induced locomotor activity and deficit of prepulse inhibition. Neuropharmacology, 2005. 49(1): p. 73-85. 221. LUISADA, P. V.: The phencyclidine psychosis: Phenomenology and treatment. In Phencyclidine (PCP) Abuse: An Appraisal, NIDA Monograph Series 21, ed. by R. C. Petersen and R. C. Stillman, pp. 241–253, US Government Printing Office, Washington, DC, 1978. 222. Roth BL, Scheffler D, Potkin SG (2005) Atypical antipsychotic drug actions: unitary or multiple mechanisms for “atypicality”? Clin Neurosci Res 3:108 –117.

139

223. Furchgott, RF. The use of β-haloalkylamines in the differentiation of receptors and in determination of dissociation constants of receptor- agonist complexes. Advances in Drug Research, vol. 3, ed. By N.J. Harper and A.B. Simmonds, pp 21-55, Academic Press, London, 1966.

140