Download The

REGULATION OF DENDRITIC MORPHOLOGY AND SYNAPSE FORMATION BY

THE INTELLECTUAL DISABILITY ASSOCIATED PALMITOYL ACYL

TRANSFERASES zDHHC15 and zDHHC9

Jordan J. Shimell

B.Sc., B.A., Simon Fraser University, 2011

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES

(Neuroscience)

THE UNIVERSITY OF BRITISH COLUMBIA

(Vancouver)

October 2019

The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled:

Regulation of dendritic morphology and synapse formation by the intellectual disability associated palmitoyl acyl transferases zDHHC15 and zDHHC9

submitted by Jordan J. Shimell in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Neuroscience

Examining Committee:

Shernaz Bamji, Cellular and Physiological Sciences Supervisor

Elizabeth Conibear, Biochemistry Supervisory Committee Member

Calvin Roskelly, Cellular and Physiological Sciences University Examiner

Stefan Taubert, Medical Genetics University Examiner

Additional Supervisory Committee Members:

Tim O’Connor, Cellular and Physiological Sciences Supervisory Committee Member

Lynn Raymond, Psychiatry Supervisory Committee Member

Abstract

Palmitoylation is a reversible post-translational modification that facilitates vesicular transport and subcellular localization of modified proteins. This process is catalyzed by a family of palmitoyl acyltransferases known as zDHHC enzymes and mounting evidence suggests that these enzymes play key roles in the development and function of neuronal connections.

Additionally, a number of zDHHCs have been associated with neurodevelopmental, neurological and neurodegenerative diseases. Loss-of-function variants in zDHHC15 and zDHHC9 are associated with intellectual disabilities; however, there is limited information on the function of these enzymes in the brain. This dissertation discusses work that demonstrates that zDHHC15 and zDHHC9 palmitoylation independently regulate dendritic arborization and are required for the formation and/or maintenance of excitatory (zDHHC15) and inhibitory (zDHHC9) synapses, thereby regulating the balance between excitation and inhibition. Loss of zDHHC15 function inhibits dendrite growth and decreases the palmitoylation and trafficking of PSD-95 into dendrites, leading to deficits in spine maturation. Loss of zDHHC9 function promotes dendritic retraction through aberrant palmitoylation of the small GTPase, Ras, and decreases the formation/maintenance of inhibitory synapses by decreasing the palmitoylation of the small

GTPase, TC10. As well, knocking out zDHHC9 in mice results in decreased palmitoylation of

Ras and TC10, and leads to elevated synaptic excitability and seizure-like activity. This work provides new insights into the function of zDHHC15 and zDHHC9 and provides a plausible mechanism for how loss-of-function mutations in these proteins may contribute to the etiology of intellectual disability.

iii

Lay Summary

Neurons are specialized nervous system cells with elongated processes called axons and dendrites. When axons from one neuron make contact with dendrites from another neuron, specialized junctions (synapses) are formed, which allow communication between neurons.

Dysfunction in dendrites and/or synapses are thought to underlie brain disorders, including intellectual disabilities. Neurons must transport proteins over long distances within dendrites and axons, which requires reliable transport mechanisms. One method to achieve this is the addition of a fatty acid, a process known as “palmitoylation”, which plays a central role in dendrite and synapse function. Indeed, approximately 41% of all synaptic proteins can be palmitoylated.

Mutations in two enzymes that regulate palmitoylation, zDHHC15 and zDHHC9, have been identified in patients with intellectual disability. In this dissertation, I determine that these enzymes are critical for dendrite growth and synapse formation, and provide a mechanism for how loss of these enzymes contributes to intellectual disability.

Preface

The work in Chapter 2, entitled “Regulation of Dendrite Morphology and Excitatory

Synapse Formation by zDHHC15” has been published as:

Shah, B.S.1, Shimell, J.J.1, & Bamji, S.X. Regulation of dendrite morphology and excitatory synapse formation by zDHHC15. Journal of Cell Science, doi:10.1242/jcs.230052.

1These authors contributed equally to this work.

All experiments were conceived by BSS, JJS, and SXB, and all experiments were jointly conducted by BSS and JJS. Experiments by BSS and JJS were done in equal partnership with equal intellectual contribution. Bhavin Shah performed the developmental time course

Western blots, localization immunochemistry, biotinylation assays, dendritic imaging, spine analysis, and synaptic imaging. Jordan Shimell performed colocalization experiments with PSD-

95, gephyrin, and giantin, Western blots for protein expression, live imaging of dendritic growth dynamics, palmitoylation assays, and fluorescence recovery after photobleaching (FRAP) experiments. Jordan Shimell and Bhavin Shah performed all analysis, data curation, figure creation, and manuscript preparation equally.

The work in Chapter 3, entitled “The X-linked Intellectual Disability Gene, zDHHC9, is Essential for Dendrite Outgrowth and Inhibitory Synapse Formation” is accepted in principle and will be published in Cell Reports as:

Shimell, J.J., Shah, B.S., Cain, S.M., Thouta, S., Jovellar, D.B., Brigidi, G.S., Kass, J., Tatarnikov, I., Kuhlmann, N., Milnerwood, A., Snutch, T.P. & Bamji, S.X. The X-linked intellectual disability gene, zDHHC9, is essential for dendrite outgrowth and inhibitory synapse formation. (accepted in principle at Cell Reports).

Experiments were conceived by JJS, GSB, BSS and SXB, and conducted by JJS with the following exceptions: DBJ assisted with immunos and imaging for colocalization and imaging and quantification for synapse density. BSS helped perform palmitoylation assays for TC10 in HEK Cells. SMC, ST, JK, and TPC designed and performed the in vivo electrophysiology experiments. IT, NK, and AM designed and performed in vitro electrophysiological experiments. v

JJS and SXB wrote the manuscript with contribution for electrophysiology sections from SMC, ST, and TPC.

Ethics Certificate Numbers: The animal studies presented in this dissertation were performed with ethical approval from the UBC Animal Care Committee (certificates #A15-0081, #A14-0338, #A18-0331, #A16-0288, and #A19-0137).

Table of Contents

Abstract ...... 3

Lay Summary ...... 4

Preface ...... 5

Table of Contents ...... 7

List of Tables ...... 13

List of Figures ...... 14

List of Abbreviations ...... 17

Acknowledgements ...... 20

Chapter 1: Introduction ...... 1

1.1 The Hippocampus as a Model System ...... 1

1.2 Post-Translational Modification of Proteins: Lipidation ...... 4

1.3 Protein Palmitoylation ...... 5

1.3.1 Palmitoylation Enzymes ...... 7

1.3.2 Quantitative Assessment of Protein Palmitoylation ...... 10

1.3.3 Functional Consequences of Palmitoylation ...... 13

1.4 Palmitoylation in the Nervous System ...... 14

1.4.1 Involvement of Palmitoylation in Neurodevelopmental and Neurological Disease

and Disorder ...... 15

1.5 Intellectual Disability ...... 15

1.6 zDHHC15 ...... 17

1.7 zDHHC9 ...... 18

1.8 zDHHC15 and zDHHC9 in Neuronal Connectivity ...... 20 vii

1.9 Dendrite Development ...... 20

1.9.1 Dendritic Arborization ...... 22

1.9.1.1 Transcription Factors in Dendritic Development ...... 23

1.9.1.2 Extracellular Signals in Control of Dendritic Arborization ...... 24

1.9.1.3 Intracellular Mechanisms Regulating Dendritic Arborization ...... 26

1.9.2 Palmitoylation in Dendritic Arborization ...... 27

1.10 Protein Trafficking in Dendrites ...... 28

1.11 Synapse Structure and Function...... 31

1.11.1 Synaptogenesis ...... 32

1.11.2 Excitatory Synapses ...... 33

1.11.2.1 Composition of the Excitatory Presynaptic Compartment ...... 34

1.11.2.2 Composition of the Excitatory Postsynaptic Compartment ...... 36

1.11.2.2.1 PSD-95 ...... 39

1.11.2.2.2 AMPA Receptors ...... 40

1.11.2.2.3 NMDA Receptors ...... 41

1.11.2.3 Ras...... 43

1.11.3 Inhibitory Synapses ...... 43

1.11.3.1 Interneurons ...... 45

1.11.3.2 Composition of the Inhibitory Presynaptic Compartment ...... 45

1.11.3.3 Composition of the Inhibitory Postsynaptic Compartment ...... 46

1.11.3.3.1 Gephyrin ...... 47

1.11.3.3.2 GABAR & GlyR ...... 48

1.11.3.3.3 Collybistin ...... 50 viii

1.11.4 Excitatory/Inhibitory Ratio ...... 51

1.12 Small G Proteins (GTPases) ...... 53

1.12.1 Structural Similarities of Small G Proteins ...... 54

1.12.2 Ras Proteins ...... 55

1.12.3 Rho Proteins ...... 56

1.12.4 Ras and TC10 in Neurodevelopmental Disorders ...... 57

1.13 Thesis Objective and Hypothesis ...... 59

Chapter 2: Regulation of dendrite morphology and excitatory synapse formation by zDHHC15 ...... 61

2.1 Introduction ...... 62

2.2 Materials and Methods ...... 63

2.2.1 Antibodies ...... 63

2.2.2 Plasmids and Primers ...... 64

2.2.3 Animals ...... 65

2.2.4 Primary Culture from Sprague-Dawley Rats ...... 65

2.2.5 Transfections (Primary Hippocampal Cultures/HEK293T Cells) ...... 66

2.2.6 Immunocytochemistry ...... 67

2.2.7 Biotinylation ...... 68

2.2.8 Acyl-RAC (palmitoylation) assay...... 69

2.2.9 Western blot analysis ...... 69

2.2.10 Imaging ...... 70

2.2.11 Fluorescence Recovery after Photobleaching (FRAP) ...... 70

2.2.12 Image Analysis and Quantification ...... 71 ix

2.2.12.1 Dendrite Length and Quantification ...... 71

2.2.12.2 Spine Analysis ...... 71

2.2.12.3 Synapse Density, Puncta Size Analysis, and Colocalization ...... 72

2.2.13 Statistical Analysis ...... 72

2.3 Results and Discussion ...... 73

2.3.1 zDHHC15 is Expressed During Early Stages of Brain Development, and in Cultured

Excitatory and Inhibitory Hippocampal Neurons ...... 73

2.3.2 zDHHC15 Promotes Dendritic Outgrowth and Arborization ...... 76

2.3.3 zDHHC15 Promotes Spine Maturation and the Formation of Excitatory Synapses 80

2.3.4 zDHHC15 Promotes the Trafficking of PSD-95 into Dendrites ...... 82

Chapter 3: The X-linked Disability Gene, zDHHC9, is Essential for Dendrite Outgrowth and Inhibitory Synapse Formation ...... 87

3.1 Introduction ...... 88

3.2 Methods and Materials ...... 90

3.2.1 Key Resources Table ...... 90

3.2.2 Experimental Model and Subject Details ...... 98

3.2.2.1 Cell Lines (HEK293T cells) ...... 98

3.2.2.2 Animal Ethics...... 99

3.2.2.3 Primary Culture from Sprague-Dawley Rats ...... 99

3.2.2.4 zDHHC9 Knockout Mice ...... 100

3.2.3 Method Details ...... 101

3.2.3.1 Transfection (Primary Hippocampal Cultures / HEK293T Cells) ...... 101

3.2.3.2 Immunocytochemistry ...... 102 x

3.2.3.3 Biotinylation ...... 102

3.2.3.4 Palmitoylation Assays (ABE and Acyl-RAC) ...... 103

3.2.3.5 Western Blot Analysis ...... 104

3.2.4 Confocal Imaging...... 105

3.2.5 Quantification ...... 107

3.2.5.1 Dendritic Complexity (Sholl Analysis) ...... 107

3.2.5.2 Total Dendritic Length ...... 107

3.2.5.3 Synaptic Density and Total Synapses ...... 108

3.2.6 Electrophysiological Methods ...... 109

3.2.6.1 In Vitro Electrophysiological Recordings ...... 109

3.2.6.2 Brain Slice Preparation ...... 110

3.2.6.3 Acute Brain Slice Electrophysiology ...... 111

3.2.6.4 In Vivo Local Field Potential Recordings ...... 112

3.2.7 Statistical Analysis ...... 113

3.3 Results ...... 114

3.3.1 zDHHC9 is Expressed in Neurons and Localized to Golgi Satellites Associated with

Excitatory and Inhibitory Synapses ...... 114

3.3.2 zDHHC9 Promotes Dendritic Outgrowth and Maintenance ...... 116

3.3.3 zDHHC9-Mediated Palmitoylation of Ras is Essential for Dendritic Outgrowth .. 120

3.3.4 zDHHC9 is Required to mMaintain the Balance Between Excitatory and Inhibitory

Synapses ...... 124

3.3.5 zDHHC9 Promotes Inhibitory Synapse Formation Through the Palmitoylation of

TC10 130 xi

3.3.6 zDHHC9 Knockout Mice Exhibit Enhanced Synaptic Excitability ...... 134

3.3.7 zDHHC9 Knockout Mice Display Seizure-Like Cortical Spike Activity ...... 137

3.4 Discussion ...... 139

Chapter 4: Conclusion ...... 143

4.1 The Role of Palmitoylation in Excitatory and Inhibitory Synapse Balance ...... 145

4.2 Palmitoylation in Intellectual Disability ...... 148

4.3 Potential for Palmitoylation as a Therapeutic Target ...... 150

4.4 Epilepsy...... 151

4.5 zDHHC9 and Ras Signaling in Neurons ...... 152

4.6 Significance, Strengths, and Limitations ...... 154

4.7 Final Remarks ...... 157

Bibliography ...... 159

xii

List of Tables

Table 1.1 PAT and Thioesterase Involvement in Disease Processes Affecting the Nervous

System ...... 16

xiii

List of Figures

Figure 1.1 The Hippocampus and Tri-Synaptic Loop ...... 3

Figure 1.2 Major Lipid Modifications of Proteins ...... 4

Figure 1.3 Protein Palmitoylation ...... 6

Figure 1.4 Mammalian zDHHC Family ...... 8

Figure 1.5 Schematic Structures of Mammalian zDHHC Enzymes ...... 9

Figure 1.6 ABE and Acyl-RAC Assays ...... 12

Figure 1.7 Chromosomal Location and Known Mutations in zDHHC9 ...... 20

Figure 1.8 Establishment of Polarity and Stages of Neuronal Development in Hippocampal

Neurons ...... 21

Figure 1.9 Dual Modes of ER-to-Golgi Transport...... 29

Figure 1.10 Electrical vs Chemical Synapses ...... 31

Figure 1.11 Schematic Represenation of an Excitatory Synapse ...... 34

Figure 1.12 Classification of the Most Common Spine Morphologies ...... 37

Figure 1.13 Molecular Architecture of the Excitatory Synapse ...... 38

Figure 1.14 Schematic Representation of an Inhibitory Synapse ...... 44

Figure 1.15 Interneurons ...... 46

Figure 1.16 Molecular Archtitecture of the Inhibitory Synapse ...... 47

Figure 1.17 Schematic Representation of a Typical Small GTPase ...... 54

Figure 1.18 The Major Ras Pathways and Implications in RASopathies ...... 58

Figure 2.1 zDHHC15 Antibody Validation ...... 74

Figure 2.2 zDHHC15 is Expressed During Early Stages of Neocortical Development and in

Golgi Compartments of Cultured Excitatory and Inhibitory Neurons...... 75 xiv

Figure 2.3 Validation of Synaptic Markers and zDHHC15 shRNA-mediated Changes in

Excitatory Synaptic Density ...... 77

Figure 2.4 zDHHC15 Promotes Dendritic Growth and Arborization ...... 78

Figure 2.5 zDHHC15 Knockdown Leads to Inhibition of Dendritic Outgrowth ...... 79

Figure 2.6 zDHHC15 Knockdown Leads to Inhibition of Mature Spine Formation ...... 81

Figure 2.7 zDHHC15 Knockdown Decreases Excitatory Synapse Density in Hippocampal

Neurons and Disrupts PSD-95 Trafficking into Dendrites ...... 83

Figure 2.8 Model of zDHHC15 Function ...... 86

Figure 3.1 zDHHC9 is Localized to Golgi Membranes Associated with Excitatory and

Inhibitory Synapses ...... 115

Figure 3.2 zDHHC9 Colocalizes with Giantin and ERGIC ...... 116

Figure 3.3 shRNA Validation of zDHHC9 in Hippocampal Neurons and HEK293T Cells ...... 117

Figure 3.4 Representative Confocal Images, Neuron Masking, ImageJ Tracing, and Sholl

Analysis...... 117

Figure 3.5 zDHHC9 Promotes Dendrite Outgrowth and Maintenance ...... 119

Figure 3.6 Validation of Pooled Ras shRNA ...... 120

Figure 3.7 zDHHC9 Enhances Dendritic Growth Through Palmitoylation of Ras-GTPase ...... 123

Figure 3.8 Pre- and Post-Synaptic Markers of Excitatory and Inhibitory Synapses in

Hippocampal Neurons ...... 125

Figure 3.9 zDHHC9 is Required to Maintain Excitatory/Inhibitory Synapse Balance ...... 126

Figure 3.10 Total Synapses for zDHHC9 and Ras Calculated Using Synapse Density and Total

Dendritic Length ...... 127

Figure 3.11 Loss of zDHHC9 Affects GABA γ2 Clustering...... 128 xv

Figure 3.12 In Vitro Electrophysiological Recordings ...... 129

Figure 3.13 zDHHC9 Promotes Inhibitory Synapse Formation through the Palmitoylation of the

Small GTPase, TC10 ...... 133

Figure 3.14 Total TC10 Synapses ...... 133

Figure 3.15 zDHHC9 Knockout Mice Demonstrate Reduced Palmitoylation of Ras and TC10 135

Figure 3.16 Intrinsic Action Potential Firing and Membrane Properties of CA1 Neurons in Wild-

Type and zDHHC9 Knockout Mice ...... 137

Figure 3.17 zDHHC9 Knockout Mice Exhibiti Increased Both Spontaneous and Miniature

Synaptic Activity and Seizure-Like Events ...... 138

Figure 3.18 Model for zDHHC9 Function ...... 142

Figure 4.1 Unidimensional vs Multidimensional Views of E:I Balance ...... 146

Figure 4.2 Genes Screened for in X-Linked Intellectual Disability (XLID) ...... 149

Figure 4.3 Relative Expression of ZDHHC15 and ZDHHC9 from Ben-Barres RNAseq...... 157

xvi

List of Abbreviations

ABE acyl-biotin exchange ABHD α/β-hydrolase domains Acyl-RAC acyl resin assisted capture AD Alzheimer’s Disease AKAP A-kinase-anchoring protein ALS amyotrophic lateral sclerosis AMPAR α-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid type receptors AIS axon initial segment ANOVA analysis of variance APP amyloid precursor protein APT acyl-protein thioesterase CBF cleaved bound fraction CA Cornu Ammonis CA(1-3) Cornu Ammonis area 1-3 CAM cell adhesion molecule Ca2+ calcium ion CDC42 cell division control protein 42 CNS central nervous system CoA Coenzyme A CRD cysteine-rich domain CUF cleaved unbound fraction DG dentate gyrus DHHC aspartate-histidine-histidine-cysteine motif DIV days in vitro DPG aspartate-proline-glycine motif E embryonic day E:I excitatory-to-inhibitory EPSC excitatory postsynaptic current EPSP excitatory postsynaptic potential ER endoplasmic reticulum ERES endoplasmic reticulum exit sites ERGIC endoplasmic reticulum-golgi intermediate compartment ERK extracellular regulated kinase FKBP FK506 binding protein FRAP fluorescence recovery after photobleaching GABA γ-aminobutyric acid GABAR GABA receptor GAD65 L-glutamic acid decarboxylase 65 GAP GTP-ase activating protein GCP16 golgi colocalized protein 16 GDP guanine diphosphate GEF guanine nucleotide exchange factor GFP green fluorescent protein xvii

GluA(1-4) glutamate receptor subunits 1-4 GPI glycosylphosphatidylinositol GTP guanine triphosphate GRIP1 glutamate receptor interacting protein 1 HD Huntington’s Disease HEK human embryonic kidney HIP14 Huntington interacting protein 14 Htt Huntingtin H-Ras Harvey rat sarcoma viral oncogene homolog ID intellectual disability IntDen integrated density IPSC inhibitory postsynaptic current IPSP inhibitory postsynaptic potential JNK c-Jun N-terminal kinase K+ potassium ion K-Ras Kirsten sarcoma viral oncogene homolog KO knockout LIMK1 LIM Kinase 1 LTD long-term depression LTP long-term potentiation MAGUK membrane-associated guanylate kinase MAPK mitogen associated protein kinase Mg2+ magnesium ion Na+ sodium ion NH2OH hydroxylamine NL neuroligin NMDAR N-methyl-D-aspartate type receptor N-Ras Neuroblastoma rat sarcoma viral oncogene homology Nrx neurexin N-WASP neuronal Wiskott-Aldrich Syndrome protein OPC oligodendrocyte precursor cell PaCCT palmitoyltransferase conserved C-terminus motif PAK p21-activated kinase PAT palmitoyl acyltransferase PBF preserved bound fraction PDZ PSD-95/Disc large/Zone-occludens PCR polymerase chain reaction PC12 rat phaeochromocytoma cells PFA paraformaldehyde PM plasma membrane PPT palmitoyl-protein thioesterases PSD postsynaptic density PSD-95 postsynaptic density 05 PUF preserved bound fraction PV parvalbumin xviii

Rab Ras related in brain Raf rapidly accelerated fibrosarcoma Ran Ras-related nuclear protein Ras rat sarcoma Rho Ras homologue RNAi ribonucleic acid interference ROI region of interest Sar1/Arf secretion-associated Ras-related 1 / ADP ribosylation factor SDS-PAGE sodium-dodecyl sulfate and polyacrylamide gel electrophoresis shRNA short hairpin ribonucleic acid SH3 Src family homology domain 3 SNARE SNAP receptor (t-SNARE – target SNARE, v-snare – vesicle SNARE) SNP single nucleotide polymorphism SOD1 superoxide dismutase 1 SST somatostatin SV synaptic vesicle TMD transmembrane domain TTX tetrodotoxin TTxE threonine-threonine-x-glutamine motif (where x is any amino acid) VGAT vesicular GABA transporter VGlut(1/2/3) vesicular glutamate transporter (1/2/3) XLID X-linked intellectual disability XLMR X-linked mental retardation zDHHC zinc-finger DHHC (aspartate-histidine-histidine-cysteine) domain 2BP 2-bromopalmitate

xix

Acknowledgements

Numerous people deserve to have their names listed here for their continued help and support throughout this journey. First and foremost, I would like to thank my supervisor, colleague, and friend, Dr. Shernaz Bamji. Shernaz introduced me to the wonderful world of palmitoylation, and has given me all the opportunities I could hope for to learn, understand, and grow as a scientist. Her belief and support in me along the way has been unwavering. Her guidance and clarity, coupled with her leniency in letting me follow my own path, has been the foundation of my PhD. A huge thank you also goes out to my supervisory committee, Dr. Tim

O’Connor, Dr. Liz Conibear, and Dr. Lynn Raymond., who provided numerous insights, support, and their invaluable time throughout my graduate career.

Words cannot express the gratitude I have for my colleagues at UBC, and there simply is not enough time to list everyone who deserves to be here. To my classmates in Neuroscience, we have shared many difficulties and successes, and sharing our collective knowledge and experiences throughout grad school has made the journey all the easier. To all the members of the Neuroscience Graduate Student Association, our time together was amazing, both for scientific discussion and for all the good times that I will remember throughout my life.

To all Bamji lab members, past and present, it was a pleasure to work with you and you made lab a place that I looked forward to going to day in and day out. A special thank you to

Stefano Brigidi, Bhavin Shah, Riki Dingwall, Tashana Poblete, and Angie Wild, who at various times have all provided guidance, been a buffer for my rants and raves, and were inspirations, both scientifically and as a model of how to be a good person. “Quae est quisquiliarum.”

I would also like to take time to thank the entire 2C wing of the LSI, both students and professors, but notably Andrew Wong, Leslie Chan, and Anthony Berndt who were always there to help me troubleshoot or learn a new technique.

I would be remiss if I did not thank the hardworking members of the Centre for Disease

Management for their assistance in all things animal related. Your hard work and assistance behind the scenes helped keep everything running smoothly.

To my family, I would not have been able to embark on this amazing journey without your continued love and support. I can never repay you for all of the time and sacrifices you have made for me throughout the years. I consider myself truly blessed to have two families, and to the Creighton Clan, I have always felt at home with you. To my brother Darin, and his beautiful wife Devon, you are both an inspiration and your generosity and understanding have been invaluable. To my mom and dad, your advice and guidance through my academic career, and indeed my life, has led me to this place. I simply would not be the person I am without you.

To my friends outside of academia, and you all know who you are; it has been a great ride. I know that I neglected our relationships at times due to the demands of graduate school, but that never fazed any of you. You all mean more to me than words can acknowledge, but know that you share in this achievement with me.

And lastly, to my beautiful wife Elizabeth. Your endless and tireless support and love has always given me the motivation to succeed. I could not have done this without you. You have always believed in me, been a pillar of understanding, held my hand through the hardest times, and celebrated all of my accomplishments, big or small. Thank for your tireless encouragement and reminder to always be true to myself and pursue whatever makes me happy. I love you.

xxi

Chapter 1: Introduction

This thesis focuses on the regulation of dendritic arborization and synapse maintenance and/or formation by the palmitoyl acyltransferases, zDHHC15 and zDHHC9. The first project

(described in Chapter 2) determines the role of zDHHC15 in the development of neuronal connections including neuron morphology, dendritic outgrowth and synapse formation. The second project (described in Chapter 3) examines the role of zDHHC9 in dendritic outgrowth, synaptic formation and synaptic maintenance. In this Chapter, I will describe the processes of dendritic growth and synapse formation in cultured hippocampal neurons, detail the specific synaptic proteins relevant to this study, and introduce the fundamentals of palmitoylation and the zDHHC family of palmitoyl acyltransferase enzymes.

1.1 The Hippocampus as a Model System

The hippocampus is a distinct anatomical structure within the medial temporal lobe of the human brain known to play a critical role in the processes of learning and memory. The hippocampus is studied utilizing behavioral models to understand hippocampal function, and cellular and molecular methods to describe specific neural circuits, enzymes, or proteins. Two main lines of evidence support the necessity of the hippocampus in learning and memory. Firstly, lesions to the hippocampus selectively impair the formation of new memories (Scoville &

Milner, 1957). Secondly, activity induced synaptic plasticity is a fundamental feature of hippocampal neurons (Neves, Cook, & Bliss, 2008), and this plasticity is induced during memory formation and is both necessary and sufficient for the encoding and trace storage of memory (reviewed in Neves, Cook, & Bliss, 2008). Additionally, the regular and relatively simple cytoarchitecture of the hippocampus allows it to serve as a model for general neural

function (Shen et al., 1994). Therefore, specific and detailed examinations of the circuitry and synapses of the hippocampus can potentially shed light on the cellular and molecular mechanisms underlying learning and memory, and can reveal principles of neuronal signaling that give insight into how information processing functions in other brain regions.

Hippocampal cultures have been used in cellular and molecular studies of the hippocampus for nearly half a century. While, in principle, neuronal cell cultures can be made from any brain area, hippocampal cultures have proven to be popular due to the relatively simple architecture of the nerve cell population in the hippocampus (Kaech & Banker, 2006). The circuitry of the rodent hippocampus is typically depicted as a tri-synaptic loop with three major excitatory neuronal subtypes, including the granule cells of the dentate gyrus (DG) which project axons to the pyramidal cells of the Cornu Ammonis (CA) area 3, which in turn project to the pyramidal cells of CA1 along the Schaffer collateral tract (Figure 1.1).

Hippocampal neurons are an ideal in vitro model as dissociated cell cultures maintain the key properties of central nervous system (CNS) synapses (Dichter, 1975). Additionally, they are easily manipulated with acute chemical stimulation and molecular techniques and can be maintained for weeks to months. These properties allow for the analysis of subcellular protein localization and trafficking, and the examination of genetic manipulations on neuronal morphology and function (Neves, Cook, & Bliss, 2008; Banker & Cowan, 1977; Fletcher &

Banker, 1989). Neuroscientists typically use hippocampal cultures made from late-stage embryonic tissue, due to the relative ease of dissociation and lower density of glial cells when compared with mature brain tissue (Banker & Cowan, 1977). This also reduces dissociation

Figure 1.1 The Hippocampus and Tri-Synaptic Loop

The hippocampus is a distinct anatomical structure within the rodent brain that extends caudally between the neocortex and diencephalon before curving ventrally towards the temporal lobe. The hippocampus has a stereotyped circuitry involving the CA1, CA3, and DG. The DG receives information from the performant pathway of the entorhinal cortex. The DG then sends excitatory axonal projections, known as mossy fibers, to apical dendrites of the CA3 pyramidal neurons. The CA3 neurons then project to the CA1 using the excitatory Schaffer collateral axonal tract, one of the most characteristically studied synapses in neurobiology. induced shearing damage to axons and dendrites due to the presence of fewer adhesion contacts

(Brewer, 1997). Once dissociated, cultured hippocampal neurons establish axonal and dendritic polarity and form synaptic connections over a well-defined developmental timeframe that is functionally and structurally similar to that seen in vivo (Fletcher & Banker, 1989; Dotti,

Sullivan, & Banker, 1988; Bartlett & Banker, 1984a, 1984b). Additionally, decades of work examining synapse formation and maintenance has been done in hippocampal cultures allowing researchers to build upon previous knowledge and protocols. Hippocampal cultures are comprised of approximately 90% pyramidal cells, and an additional 6-15% are inhibitory interneurons (Benson et al., 1994; Pelkey et al., 2017), presenting a homogenous population of neurons with properties typical of CNS neurons (Fletcher & Banker, 1989). Together, this demonstrates that dissociated hippocampal neuronal cultures are ideal for the study of the formation, development, and maintenance of neuron morphology and synapses. In this dissertation, primary cultured hippocampal neurons are the model system used to elucidate the role of the palmitoylation, a post-translational modification mediated by the palmitoyl-acyl transferase enzymes, zDHHC15 and zDHHC9, in dendritic arborization and synaptic formation and/or development. 3

1.2 Post-Translational Modification of Proteins: Lipidation

Proteins undergo a variety of modifications that can have dramatic effects on subcellular localization, function, and trafficking. While there are numerous post-translational modifications that can occur, one of the most dramatic for protein localization is covalent lipid modification

(Fig 1.2) (Resh, 2013; Chamberlain & Shipston, 2015). Lipid modifications can include fatty acids, isoprenoids, sterols, phospholipids, and glycosylphosphatidylinositol (GPI) anchors.

Figure 1.2 Major Lipid Modifications of Proteins

Distinct enzyme families facilitate these lipid modifications. Lipid modifications generally result in the formation of a stable bond between either the NH2-terminal amino acid or cysteine/serine amino acid side chains in the protein. S-acylation, commonly referred to as palmitoylation, is reversible due to the labile thioester bond between the lipid (typically, but no exclusively, palmitate) and a cysteine amino acid. This is mediated by a family of enzymes known as zDHHC palmitoyl acyltransferases. Adapted with permission from Chamberlain and Shipston, 2015.

Proteins can be, and often are, modified by more than type of lipid (including myristate, farnesyl, and palmitate). The most common outcome from lipid modification is an increased affinity for membranes due to increased protein hydrophobicity, but attachment can also facilitate intra- or inter-molecular protein-protein interactions (Resh, 2013). There are three main types of covalent

protein-lipid modifications known for intracellular proteins. Myristoylation involves the addition of a 14 carbon saturated fatty acid, myristate, to N-terminal glycine residues through an amide linkage (Magee & Courtneidge, 1985), while prenylation (which consists of both farnesylation and geranylgeranylation) involves the transfer of a farnesyl (15 carbon) or geranylgeranyl (20 carbon) group to C-terminal residues through a thioether bond (Zhang & Casey, 1996). Protein palmitoylation involves the addition of a 16-carbon palmitate to cysteine residues through a unique thioester linkage (Magee & Courtneidge, 1985), and while discovered decades ago it has become the recent focus of neurobiologists due to its prevalent role in neuronal processes

(Fukata & Fukata, 2010).

1.3 Protein Palmitoylation

S-acylation refers to the attachment of fatty acids to cysteine residues. This term includes palmitoylation, but may also refer to the addition of both saturated and unsaturated fatty acids of various lengths. However, the addition of the palmitic acid is the most common, and S-acylation is often referred to simply as palmitoylation (Fukata & Fukata, 2010). Throughout this thesis, the term palmitoylation is used in this regard. While all lipid modifications enhance protein interactions, palmitoylation has distinct properties that make it advantageous and distinguish it from other protein-lipid modifications. Firstly, it is a larger lipid than prenyl and myristoyl groups, allowing it to have a higher propensity for stable membrane attachment (Peitzsch &

McLaughlin, 1993). Indeed, some proteins such as Ras, which is modified by myristate or farnesyl groups first, tend to sample membranes by cycling on and off until subsequent palmitoylation, which promotes stable membrane attachment (Goodwin et al., 2005; Rocks et al.,

2005). Secondly, palmitate addition is not restricted to a specific region or domain and can instead modify cysteine residues anywhere in the protein sequence (Nadolski & Linder, 2007). 5

Thirdly, palmitoylation is not coupled with translation in anyway, while myristoylation and prenylation often occur co-translationally (Zhang & Casey, 1996), allowing palmitic acid addition to be more precisely and dynamically regulated at specific subcellular locations and/or in response to specific stimuli. Lastly, the thioester bond that links palmitic acid to the protein is highly reactive and labile, allowing palmitate to be attached and removed in a cyclic process

(Figure 1.3) (Fukata & Fukata, 2010), adding another layer of dynamic regulation. Together, this gives palmitoylation a far greater degree of control in a protein’s localization and function compared to other lipid modifications.

Figure 1.3 Protein Palmitoylation

Palmitoylation, due to its reversible nature, allows for the dynamic, cyclic regulation of proteins through the regulated addition or removal of palmitate. Palmitoylation is Palmitoylation is mediated by a family of 23 palmitoyl acyltransferase enzymes known as zDHHC proteins. There is little known about the depalmitoylating palmitoyl protein thioesteraseses.

1.3.1 Palmitoylation Enzymes

While instances of protein palmitoylation can be observed in the literature as early as

1979 (Schmidt & Schlesinger), there was controversy about whether this was an enzymatic or spontaneous process (reviewed in Dietrich & Ungermann, 2004). For example, spontaneous, autocatalytic palmitoylation had been observed when some proteins, such as SNAP-25, were incubated with acyl-coenzyme A in vitro (Veit, 2000). This controversy was resolved with the identification of palmitoyl acyltransferase (PAT) enzymes in yeast, which confirmed that palmitoylation was an enzymatic process (Bartels et al., 1999; Lobo et al., 2002). The first PAT identified was the yeast Ras PAT complex Erf2-Erf4 (Bartels et al., 1999) and subsequent biochemical analysis confirmed that Erf2-Erf4 was a bonafide PAT for Ras2 (Lobo et al., 2002).

These same studies demonstrated that PATs were a family of multi-pass transmembrane proteins containing a conserved zinc-finger aspartate-histidine-histidine-cysteine (zDHHC) motif required for enzymatic activity both in vitro and in vivo (Mitchell et al., 2006). Mutation of the catalytic cysteine severely reduces palmitoylation of the enzyme, and often of the substrate as well (Lobo, 2002; Mitchell et al., 2010; Roth, 2002). There are 23 mammalian zDHHC proteins

(Figure 1.4) with the majority validated as having PAT function in either yeast (Ohno et al.,

2012) or mammalian cells (Fukata et al., 2004).

The catalytic core of PATs actually extends over ~51 amino acid cysteine-rich domain

(CRD) and the zDHHC domain extends on the cytosolic face of the membrane (Mitchell et al.,

2006; Politis et al., 2005). There is also a conserved aspartic acid-proline-glycine (DPG) motif and a conserved threonine-threonine-x-glutamine (TTxE) motif that help facilitate enzymatic activity, and some zDHHC proteins also contain SH3 domains, Ankyrin repeats, or PDZ-binding

Figure 1.4 Mammalian zDHHC Family

Phylogenetic organization of zDHHC proteins into subfamilies based upon sequence similarity. Adapted with permission from Fukata and Fukata, 2010.

motifs (Figure 1.5), and an additional 16 amino acid palmitoyltransferase conserved C-terminal

(PaCCT) motif has been described that is 70% conserved in all eukaryotic organisms.

The majority of zDHHC proteins display endoplasmic reticulum and/or Golgi localization, although a few such as zDHHC5 and zDHHC20 localize to the plasma membrane (Ohno et al.,

2006). It is believed that these distinct subcellular localizations help define substrate specificity, and indeed the palmitoylation of some substrates appears remarkably dependent on a specific

PAT (Roth et al., 2006). However, other proteins appear to undergo palmitoylation by multiple

PATs (Fukata et al., 2004; Huang et al., 2009; Salaun et al., 2010), and some PATs display promiscuous activity (Greaves & Chamberlain, 2011a). A complete characterization of the

enzyme-substrate interaction of each zDHHC will be necessary to completely understand this complex process and determine how, and why, these redundancies exist.

Figure 1.5 Schematic Structures of Mammalian zDHHC Enzymes . zDHHC proteins generally have 4 transmembrane domains, although zDHHC13 and zDHHC17 consist of six transmembrane domains. The conserved cysteine-rich domain containing the zDHHC motif is in the cytoplasmic loop proximal to transmembrane domain 3 or 5 respectively. zDHHC proteins also have a conserved TTxE domain after the C-terminus of final transmembrane domain. zDHHC6 is unique in that it contains an SH3 domain, while zDHHC3, zDHHC5, and zDHHC8 all have PDZ- binding domains at the C-terminus. zDHHC13 and zDHHHC17 have Ankyrin repeats on the N- terminal cytosolic region. Adapted with permission from Cho and Park, 2016

While much progress has been made in the characterization of palmitoylating enzymes over the past decade, research into the enzymes that catalyze the depalmitoylation of substrates has been limited, despite the fact that the removal of palmitate from proteins appears to be strictly an enzymatic process. Additionally, the depalmitoylation of many substrates is as critical for proper function, and factors heavily in the ability of palmitoylation to be dynamically regulated. Indeed, for substrates that undergo autoacylation, the enzymatic removal of the palmitate group serves as the only regulation (Zeidman et al., 2009). While beyond the scope of this thesis, a number of depalmitoylation enzymes have also been characterized: the acyl protein thioesterases (APT) APT1, APT2, and APTL1 (Lin & Conibear, 2015b) and the protein-

palmitoyl thioesterase (PPT) PPT1 (Zeidman et al., 2009). The APTs and PPTs belong to the metabolic serine hydrolase superfamily of proteins based on sequence homology and structural studies, and a number of additional members of this family exist, although investigations into their depalmitoylase activity is limited (Lin & Conibear, 2015a,b). For detailed descriptions of the enzymatic family and plausible mechanisms of action, see Long & Cravatt (2011). However,

PPT1 has been demonstrated to catalyze palmitate removal from H-Ras in vitro (Camp &

Hofmann, 1993) and APT2 knockdown alters H-Ras localization (Tomatis et al., 2010) and can depalmitoylate N-Ras in vitro (Rusch et al., 2011).

Two additional novel potential depalmitoylating enzymes, ABHD17 (Lin & Conibear,

2015a) and FKBP12 (Ahearn et al., 2011), have also been identified. FKBP12 binds H-Ras in a palmitoylation-dependent manner and then promotes its depalmitoylation, allowing Ras to return to the Golgi for additional palmitoylation cycling (Ahearn et al., 2011). Recent work has also demonstrated that ABHD17 catalytic activity is required for N-Ras depalmitoylation and re- localization to endomembranes (Lin & Conibear, 2015a). However, the research into depalmitoylation is ongoing and it remains to be seen if depalmitoylation enzymes are substrate- specific.

1.3.2 Quantitative Assessment of Protein Palmitoylation

Traditionally, protein palmitoylation was probed using metabolic labeling involving the radioactive palmitate analogue [3H]-palmitate in cultured cells (Berthiaume et al., 1995; Resh,

2006). This allowed the ability to utilize pulse-chase studies to determine the palmitate cycling time on proteins, which has been performed on a handful of substrates including Ras (Baker et al., 2003). Although this provides a direct measure of palmitoylation, this technique lacks sensitivity and requires the handling of radioactive substances, which can also be toxic to the 10

cells. A more sensitive method exploits probes that were designed for bio-orthogonal reactions, namely copper(I)-catalyzed azide-alkyne cycloaddition (CLICK) chemistry (Sletten & Bertozzi,

2011). In this method, bio-orthogonal alkyne- or azido-containing palmitate analogues are used to label proteins in cultured cells (Hang et al., 2007; Kostiuk et al., 2008; Charron et al., 2009), which are then conjugated to reporters such as biotin or fluorophores allowing for the assessment of protein palmitoylation using common techniques such as Western blotting or mass spectrometry. This technique is advantageous in that it allows a measure of protein palmitate turnover, but is limited in that it requires live cells and can be confounded by the rate of protein synthesis and/or palmitate turnover (i.e. proteins with slower palmitate turnover require extended labelling).

To investigate the total pool of palmitoylated proteins, additional methodology has been developed that is amenable to use in cultured cells and whole tissue extracts. The acyl-biotin exchange (ABE) assay (Drisdel & Green, 2004) and acyl-resin assisted capture (acyl-RAC) assay (Forrester et al., 2011) involve the same basic 3 steps: (1) free thiols are blocked, (2) thioester bonds are cleaved leading to the release of palmitate from palmitoylated cysteines, and

(3) the newly exposed thiols are bound, allowing for quantification. The difference in the assays is that in ABE, the exposed thiols are bound to a biotin-based, sulfhydryl-specific label and then a streptavidin-sepharose binding step, whereas the acyl-RAC uses thiopropyl-sepharose beads that directly react with the free cysteines (Figure 1.6). Although beneficial as this allows the quantification of the relative palmitoylated protein at a given time point, these assays do not provide any information on the kinetics of palmitoylation. Additionally, both assays have the potential for false positives due to incomplete blocking or other thioester modifications, such as

Figure 1.6 ABE and Acyl-RAC Assays

The ABE and Acyl-Rac assays use cysteine blockers, such as N-ethylmaleimide (NEM) to block free cysteines and cleavage reagents, such as hydroxylamine (HAM) to remove palmitate. This leaves a free cysteine which is either bound to thiopropyl-sepharose resin (Acyl-RAC) or to biotin beads which then react with streptavidin-sepharose beads (ABE). Reproduced with permission from Sanders et al., 2015. nitrosylation. It is also possible to utilize samples prepared with ABE or acyl-RAC in mass spectroscopy to directly detect palmitoylated proteins, and the techniques provide similar results

(Edmonds et al., 2017). This has led to many large-scale “palmitoylome” studies, generating useful data when comparing wildtype and specific zDHHC knockout animals to create a list of all potential enzyme-substrate pairs. This, coupled with novel in silico palmitoylation prediction software such as Swisspalm (Blanc et al., 2015) provides researchers with the ability to determine if their protein of interest is palmitoylated.

Recently, another method for detecting and measuring palmitoylation has been developed that uses the biochemical exchange of acyl groups on cysteines with defined mass-tags. Two main assays are included in this category, namely the Acyl-PEG Exchange assay or the Acyl

Pegyl Gel Shift (Percher et al., 2016; Yokoi et al., 2016). These assays not only measure the differential palmitoylation state or isoforms of proteins, but can also determine site-specific palmitoylation and determine protein palmitoylation stoichiometry. This is accomplished through analysis of the mass-tagged proteins by Western blotting revealing mobility shifted bands which represent mass tagged cysteine residues, such that each band represents a unique palmitoylation site. These assays will prove useful for investigation into the roles of palmitoylation in the regulation of palmitoylation enzymes themselves, which are known to be palmitoylated, as well as interrogating how multiple palmitoylation sites on proteins contribute to function. Indeed, these assays have been recently used in studies of the regulation of neuronal protein palmitoylation during synaptic plasticity (Woolfrey et al., 2018; Purkey et al., 2018), demonstrating their utility in investigating protein palmitoylation.

In the case of palmitoylation assays, it can often be useful to inhibit palmitoylation.

Currently, this is accomplished by using a global palmitoylation inhibitor known as a 2- bromopalmitate (2-BP). 2-BP is a non-metabolizable palmitate analogue that blocks palmitate incorporation into proteins (Webb et al., 2000) and is highly toxic to cells (Mikic et al., 2006). 2-

BP is also not specific to a given substrate, and as depalmitoylation enzymes are also believed to be palmitoylated, 2-BP addition represents a complete loss of palmitoylation from the sample

(Pedro et al., 2013) and suggests that conclusions drawn from 2-BP may be difficult to assess.

1.3.3 Functional Consequences of Palmitoylation

It is now understood that a diverse array of proteins undergo palmitoylation, including signaling proteins, scaffold proteins, membrane associated proteins, synaptic proteins, mitochondrial proteins, and even viral proteins (Charollais and Van Der Goot, 2009; Corvi et al.,

2001; Fukata and Fukata, 2010; Kostiuk et al., 2008; Mitchell et al., 2006; Veit and Schmidt, 13

2006). Consequently, palmitoylation plays a key role in regulating many cellular processes, including tethering of signal proteins, protein trafficking, protein stability, protein-protein interactions, membrane association, and protein localization, including segregation of proteins into particular protein subdomains (Fukata & Fukata, 2010; Linder & Deschenes, 2007; Salaun et al., 2010). Often, palmitoylation in conjunction with another lipid modification stabilizes the protein interaction with the membrane, and leads to their segregation into lipid rafts, where they are able to interact with other raft proteins to facilitate signaling events (Brown, 2006; Levental et al., 2010). For proteins that are already membrane localized, palmitoylation may further alter hydrophobicity or induce a tilt of a transmembrane domain and/or conformational change that facilitates and/or inhibits protein-protein interactions (Charollais & Van Der Groot, 2009). While still in its infancy, studies on palmitoylation in the last decade have advanced our understanding of the mechanism and function of protein palmitoylation, and begun to reveal the role of palmitoylation enzymes in human health.

1.4 Palmitoylation in the Nervous System

A critical role for palmitoylation in the nervous system has emerged in the last decade

(reviewed in Fukata & Fukata, 2010; Globa & Bamji, 2017). Palmitoylation has been demonstrated to be critical for neuronal development, affecting processes including neurite outgrowth, axon pathfinding, filopodial formation, and spine development, maintenance, pruning and plasticity (Arstikaitis et al., 2008; Gauthier-Campbell et al., 2004; El-Husseini et al., 2002;

Kato et al., 2000; Kutzleb et al., 1998; Laux et al., 2000; Ueno, 2000; Brigidi et al., 2015,

Thomas et al., 2012; Hayashi et al., 2005; Greaves & Chamberlain, 2011b; Kang, 2004; Kang et al., 2008), deficits in which can manifest as neural dysfunction. Indeed, over 41% of all synaptic

proteins are substrates for palmitoylation (Sanders et al., 2015). Palmitoylation serves to regulate the assembly and compartmentalization of many neuronal proteins at both the presynaptic and postsynaptic terminals including PSD-95, Gephyrin, GAD65, and SNARE proteins (El-Husseini

& Bredt, 2002; Fukata & Fukata, 2010; Huang & El-Husseini, 2005).

1.4.1 Involvement of Palmitoylation in Neurodevelopmental and Neurological Disease and Disorder

As our knowledge of palmitoylation and palmitoylating enzymes has grown, it has become clear that palmitoylation is a critical component of many biological processes. It comes as no surprise that impaired or disturbed palmitoylation often leads to diseases or disorders. Of the 23 zDHHC enzymes, several have been specifically linked to diseases or disorders of the nervous system (Table 1.1), demonstrating a clear need to understand the role of these enzymes in the brain. These include Alzheimer’s disease, Huntington’s disease, schizophrenia, and intellectual disability (Bhattacharyya et al., 2013; Hornemann, 2015; Huang et al., 2004; Korycka et al.,

2012; Mizumaru et al., 2009; Mukai et al., 2008; Yanai et al., 2006; Fromer et al., 2014; Sanders

& Hayden, 2015; Raymond et al., 2009; Linhares et al., 2016; Martinez et al., 2014; Mansouri et al., 2005; Masurel-Paulet et al., 2014; Mitchell et al, 2014; Baker et al., 2015; Tzschach et al.,

2015; Han et al., 2017). These have been extensively reviewed elsewhere (Cho & Park, 2016;

Chavda et al., 2014), and the focus of this thesis will be on the role of palmitoylation in intellectual disability.

1.5 Intellectual Disability

Intellectual disability (ID) affects approximately 1% of the population (Maulik et al.,

2011) and presents with substantial limitations in cognitive function and adaptive behavior with a variable profile of severity ranging from mild to profound. The majority of individuals with ID 15

Table 1.1: PAT and Thioesterase Involvement in Disease Processes Affecting the Nervous System.

Human Experimental Associated Features Enzyme Reference Disease Relevance Model

Huntington’s disease In vitro Weaker HTT-HIP14 interaction and reduced HIP14 Singaraja et al. 2002, palmitoylation in presence of disease Huang et al. 2004 causing mutation in HTT.

Animal model Reduced HTT palmitoylation in YAC128 HIP14 Yanai et al. 2006 mouse model of HD.

Animal model Phenotype in mice lacking murine Hip14 HIP14 Singaraja et al. 2011 resembles mouse model of HD

Animal model Strain 129S6/SvEv: Hair, skin, and bone HIP14L Saleem et al. 2010 abnormalities with impaired survival and global amyloidosis. Reduced HTT palmitoylation in vitro

Alzheimer disease In vitro DHHC12 modifies APP metabolism ZDHHC12 Mizumaru et al. 2009

Amyotrophic lateral In vitro Increased palmitylation and decreased -- Antinone et al. 2013 sclerosis processing of familial ALS SOD1 mutants

Schizophrenia Human Increased transmission of Schizophrenia- ZDHHC8 Liu et al. 2002a associated SNP in females with 22q11 microdeletions

Animal model In females: ZDHHC8 Mukai et al. 2004 Deficit in prepulse inhibition ZDHHC5 Mukai et al. 2008 Abnormal fear-related exploratory behaviour Decreased sensitivity to NMDAR blocker Decreased density of dendritic spines Rescue by WT ZDHHC8

X-linked mental Human Severe non-syndromic XLMR, epileptic ZDHHC15 Mansouri et al. 2005 retardation (XLMR) seizures, dysmorphic facial appearance

Human Moderate XLMR in males, developmental ZDHHC9 Raymond et al. 2007 delay

Infantile Neuronal Human Infantile onset blindness, PPT1 Vesa et al. 1995 Ceroid neurodegeneration, auto fluorescent Lipofuscinosis lipopigments in neurons

Learning & memory Animal model Impaired cognition and impaired survival in ZDHHC5 Yi et al. 2010 hypomorphic Zdhhc5 mouse

Ischemic stroke Animal model Enhanced interaction of HIP14 with JNK3 in HIP14 Yang and Cynader, 2011 brains from a rat model of transient ischemic stroke, treatment before or after the ischemic insult with a peptide that inhibits this interaction reduced the infarct size by 80%

**Adapted with permission from Shaun Sanders, 2007

are identified in early childhood because of delays in development, and ID is a prominent feature in a number of neurodevelopmental disorders. Indeed, ID often presents as comorbid with a number of other neurodevelopmental or neurological features such as epilepsy, sensory impairment, and autism spectrum disorders (Vissers et al., 2016). While the etiology of ID can be variable, common factors include exogenous factors such as malnutrition, oxygen deprivation, and maternal alcohol abuse. Genetic factors are also known to play a key role in the manifestation of ID (reviewed in Chiurazzi & Pirozzi, 2016). Mutation, deletion, or other deficits in key genes regulating the formation, maturation, or function of processes involved in brain development can have severe consequences. Several ID mutations have been mapped to two members of the DHHC family, zDHHC15 and zDHHC9. However, the association between zDHHC15 and zDHHC9 and clinical neurological phenotypes has only just begun to emerge. zDHHC9 and zDHHC15 are among a group of >110 genes on the X-chromosome that have been associated with X-linked intellectual disabilities (XLID), which includes over 200 syndromes and accounts for approximately 16% of intellectual deficiencies in males (reviewed in Stevenson

& Schwartz, 2009). The most common presentations accompanying ID are dendritic abnormalities and alterations in dendritic spines and the balance of excitatory and inhibitory synapses (Purpura, 1974; Kaufman & Moser, 2000; Valnergi et al., 2012; Verpelli et al., 2011;

Calfa et al., 2015; Martin et al., 2015). This dissertation aims to determine the function of zDHHC15 and zDHHC9 in dendritic outgrowth and synapse formation, and determine a plausible mechanism for how changes in these processes lead to intellectual disability.

1.6 zDHHC15

zDHHC15 is one of a number of genes on the X chromosome, and has been reported as a strong candidate for nonsyndromic X-linked intellectual disability (XLID) based on the 17

observation of a woman with ID due to a balanced reciprocal translocation between chromosome

X and 15 that caused the absence of zDHHC15 transcripts (Mansouri et al., 2005). zDHHC15 is expressed in many neuronal cell types, including astrocytes, neurons, microglia, and oligodendrocytes (Zhang et al., 2014). Interestingly, zebrafish zDHHC15b, which is homologous to human zDHHC15, has been shown to be important for the development of the diencephalon and specifically the differentiation of dopaminergic neurons, the loss of which results in learning and memory deficits (Wang et al., 2015). In Chapter 2, zDHHC15 will be further discussed, and the role of zDHHC15 will be examined in dendrite growth and arborization, synapse formation and maintenance, and the trafficking of one of its confirmed substrates, PSD-95.

1.7 zDHHC9

zDHHC9 was one of the first palmitoyl acyltransferases identified as a two protein complex in yeast, consisting of Erf2 (zDHHC9) and Erf4 (GCP16 or GOLGA7 in mammals)

(Lobo et al., 2002). zDHHC9 was deemed critically important as it is the PAT for Ras, a protein involved in numerous cellular processes. Despite being among the few validated enzyme- substrate pairs using purified proteins (Swarthout et al., 2005), knockdown of zDHHC9 in heterologous cells does not significantly affect Ras localization (Rocks et al., 2010).

Investigation into zDHHC9 has been limited, and traditional overexpression screens often fail to take into account the cofactor GCP16, which is required for proper enzymatic function

(Swarthout et al., 2005; Mitchell et al., 2012; Zhao et al., 2002), and experimentally validated as needing to be in a 1:1 ratio (Mitchell et al., 2014). It is believed that this cofactor increases zDHHC9 stability and also stabilizes the palmitoyl-intermediate. The most likely explanation for this is that the conserved C-terminus domain is a likely site for GCP16 binding, as the C- terminus domain is required for enzymatic function. The end of a specific PaCCT motif in the C- 18

terminus region contains two conserved prolines, which likely affect the length and secondary structure of the region. Interestingly, deletion of this C-terminus region significantly reduces the palmitoylation activity of the PAT, despite not affecting the conserved DHHC domain, and additionally significantly reduces protein levels (Mitchell et al., 2014). There are three additional cysteines in this region, with one found on the hydrophobic face, and it is possible that palmitoylation of these cysteines could help facilitate zDHHC9 function by anchoring the zDHHC9-GCP16 complex to the membrane, or enhancing protein-protein interactions. Indeed, this has been observed with other PATs as a “palmitoylation cascade”, such as the case of zDHHC16-zDHHC6 (Abrami et al., 2017), and at least one of the cysteines is predicted to be palmitoylated due to similar homology with known palmitoylated residues in zDHHC5 and zDHHC8 (Brigidi et al., 2015).

Genetic and biochemical evidence strongly suggests a role for zDHHC9 in XLID

(Raymond et al., 2007; Mitchell et al., 2014; Masurel-Paulet et al., 2014; Baker et al., 2015; Han et al., 2017; Tzschach et al., 2017). A number of mutations have been reported either directly from genetic sequencing of affected families, or through large genetic screens (Fig 1.7). The two most characterized mutations, R148W and P150S, result in a decrease in the steady-state palmitoyl intermediate of zDHHC9 (Mitchell et al., 2014), which has led many to hypothesize that reduced palmitoylation of zDHHC9 substrates may be causative or implicated in XLID.

However, the role of zDHHC9 in neurons and the effects of palmitoylation of its substrates has not been investigated or established. In Chapter 3, zDHHC9 will be further discussed and we investigate the role of zDHHC9 in regulating neuronal connectivity, and propose a plausible mechanism for how zDHHC9 may contribute to XLID.

Figure 1.7 Chromosomal Location and Known Mutations in zDHHC9

zDHHC9 is located on the X-chromosome at position q26.1 (Xq26.1), and to date 9 mutations have been observed that may contribute to the manifestation of intellectual disability.

1.8 zDHHC15 and zDHHC9 in Neuronal Connectivity

As the principal deficits observed in intellectual disability are dendritic abnormalities and alterations in dendritic spines and the balance of excitatory and inhibitory synapses (Purpura,

1974; Kaufman & Moser, 2000; Valnergi et al., 2012; Verpelli et al., 2011; Calfa et al., 2015;

Martin et al., 2015), this dissertation will examine the role of these two proteins in neuronal connectivity. Neuronal connectivity is influenced both by dendritic arborization and by synaptic alteration amongst many other processes. The following sections will describe the fundamentals of dendrite development, growth, and regulation, as well as excitatory and inhibitory synapses.

1.9 Dendrite Development

The nervous system is comprised of a vast number of polarized neurons, with distinct subcellular compartments, including one or multiple dendritic processes arising from the cell body. Neurons establish their polarity early on in development as neurons differentiate and

extend processes, or neurites, which must be actively maintained through the lifespan of the neuron. Both the establishment and maintenance of this dendritic arbor involves the coordinated and widespread regulation of cytoskeleton and molecular trafficking machinery (Foletti et al.,

1999; da Silva & Dotti, 2002). In hippocampal cultures, the establishment of polarity follows a stereotyped sequence of developmental events from plating until maturation, divided into five distinct stages (Dotti, Sullivan, & Banker, 1988) (Figure 1.8).

Figure 1.8 Establishment of Polarity and Stages of Neuronal Development in Hippocampal Neurons

Culture hippocampal neuron form multiple lamellipodia when they attach to the growth surface (Stage 1), and shortly after begin to form immature neurites (Stage 2). These neurites extend and retract without net elongation, until one breaks morphological symmetry and elongates rapidly to form the axon (Stage 3). Several days later, the remaining neurites grow and branch to form dendrites (Stage 4). Maturation of the neuron continues as the dendrites develop dendritic protrusions, or spines, which form synaptic contacts with other neurons (Stage 5). Used with permission from Govek et al., 2005.

In the first stage, shortly after plating, neurons form lamellipodia (0 – 6h, stage 1). These protrusions develop into several immature, morphologically indistinguishable neurites, which cycle between periods of growth and retraction (6 – 24h, stage 2). At one point, a single neurite will break the initial morphological symmetry, extend rapidly and become the axon (2 – 3 DIV, stage 3). Over the next several days, the remaining neurites branch and differentiate into 21

dendrites, forming the neuronal arbor (4 – 7 DIV, stage 4), before ultimately developing dendritic spines, forming synaptic contacts, and undergoing further maturation to establish a neuronal network (7 – 18 DIV, stage 5).

The structural development of dendrites is mediated by both intrinsic and extrinsic factors, which can be innately expressed and/or activity dependent (Horch et al., 1999; Hua &

Smith, 2004; Konur & Ghost, 2005; reviewed in Dong, Shen & Bülow, 2015). Once a neuron has matured, it consists of a soma, an axon with presynaptic boutons, and dendrites with postsynaptic spines, all of which are specialized compartments and have complex requirements and morphology. In fact, without the knowledge of ion channels or any of the biophysical properties of neurons, careful observation of the establishment of the formation of axons and dendrites, and the morphological characteristics of each, led to the first ideas of how neurons were able to transmit signals (Golgi, 1873; Ramon y Cajal, 1888, cited in Finger, 2000). Both the dendritic shafts and dendritic protrusions (often referred to as spines) will serve as the sites of postsynaptic contact in the mature neuron. Prior to spine development and synaptic maturation, the dendrites begin to form branch points in a process called dendritic arborization.

1.9.1 Dendritic Arborization

Dendritic arborization has three different, but overlapping stages: (i) neurite initiation, outgrowth, and guidance; (ii) branching and synapse formation, and (iii) stabilization. It is important to note that dendrites themselves are dynamic and continually exhibit changes in growth and stability (reviewed in Chen & Haas, 2011). Current research suggests that the maintenance of dendritic arborization is a balance between the metabolic costs of elaborating the dendrites and the need of the neuron to cover its receptive field (Wen & Chklovskii, 2008).

Indeed, arborization is not a static process, but rather dynamic with periods of rapid growth and 22

retraction. This dynamic branching, when combined with neuronal activity and synapse formation, leads to the establishment of the mature arbor. Further stabilization occurs over a long period of time, with the mature dendritic arbor having a very low branch turnover under basal conditions (Wu et al., 1999). It is believed that the stability of these branches is linked to the development and maturation of synapses (reviewed in Cline & Haas, 2008; Chen & Haas, 2011), and further activity or signaling can lead to further growth or retraction in “fine-tuning” of the dendritic arbor (Bjorkbolm et al., 2005). However, despite the importance of dendritic structure and arborization in neuronal functioning, the mechanisms that regulate these processes remain relatively unknown (Miller and Kaplan, 2003). The following sections will explain some of the known factors regulating dendritic development.

1.9.1.1 Transcription Factors in Dendritic Development

The fact that cultured neurons form reliably distinct and recognizable dendritic trees suggest that at least part of morphogenesis is regulated by intrinsic factors (Dong et al., 2015).

This is best exemplified in the Drosophila model where the transcription factor hamlet acts as a binary switch (in external sensory neurons) between simple dendrite morphology and more complex arborization (Moore et al., 2002). However, transcription factor regulation of dendritic field complexity in Drosophila neurons demonstrated that morphology can be controlled by the combinatorial expression of three other transcription factors: Cut, Abrupt, and Spineless

(Grueber et al., 2003; Li et al., 2004; Sugimura et al., 2004; Kim et al., 2006). This is perhaps still simplified, as RNAi screens for transcription factors that influence dendritic arborization identified over 70 regulators of class I sensory neurons in Drosophila neurons (Parrish et al.,

2006).

At least some of these transcriptional factors are conserved throughout species, as the vertebrate homologs of Drosophila Cut, Cux1 and Cux2, are implicated in the development of cortical dendrites (Li et al., 2010; Cubelos et al., 2010). Other transcription factors, such as

Neurogenin 2, appear to regulate dendrite morphology in mammalian systems (Hand et al.,

2005). Loss of Neurogenin 2 results in the loss of pyramidal neuron morphology, while overexpression induces pyramidal-like morphology in cultured neurons. While this data provides promising results on the understanding of transcriptional regulation, it is apparent that there is a complicated molecular framework that is controlled by transcriptional regulation to regulate dendritic arborization. Indeed, kinases such as JNK and ERK that regulate transcription factors play a role in dendritic arbor formation and maintenance (Bjorkblom et al., 2005). Transcription factor regulation is beyond the scope of this thesis, and has been well described elsewhere

(reviewed in Cort et al., 2016).

1.9.1.2 Extracellular Signals in Control of Dendritic Arborization

In contrast with the intrinsic regulators of dendritic arborization, a large body of evidence for the importance of extracellular guidance and regulation of dendritic arborization exists. Three main factors exist for extracellular signaling, which depending on developmental stage, consist of a combination of (1) diffusible cues, (2) cell contacts, and (3) neuronal activity (reviewed in

McAllister, 2000; Wong & Ghosh, 2002; Jan & Jan, 2003). Diffusible cues are numerous, but some of the most classically studied include BDNF (McAllister et al., 1995; Horch & Katz,

2002), semaphorins (Polleux et al., 2000; Ng et al., 2013), and Slits/Robos (Whitford et al.,

2002), which are reviewed elsewhere (Dong et al., 2015; Urbanska et al., 2008). Cell contact molecules, or adhesion molecules, have also been shown to regulate dendritic arborization, as overexpression of N-Cadherin, αN-catenin, or β-catenin has been shown to increase dendritic 24

arborization in hippocampal cultures (Yu & Malenka, 2003). Lastly, while the arbor itself largely dictates the connections formed, the inputs received within the circuit also influence arborization or differential stabilization. This is best demonstrated with in vivo imaging in the optic tectum of

Xenopus, which showed that neuronal activity via visual stimulation increased arborization via a process that required NMDA receptor activity and Rho GTPase activation (Sin et al., 2002).

Indeed, Rho GTPases are important regulators of cytoskeletal dynamics required for dendritic morphogenesis, with Rac and CDC42 promoting arborization and Rho acting as a negative regulator, and many processes regulating dendritic complexity converge on regulation of the cytoskeleton (reviewed in Govek et al., 2005). Importantly, activity driving dendritic complexity has also been demonstrated in vivo in mammalian systems, where mice reared in “enriched” environments demonstrate an increase in arbor complexity, consistent with the idea that increased neuronal activity positively influences dendritic branching (Faherty et al., 2003).

Dendritic morphogenesis in hippocampal cultures is also known to be regulated by the Ras- mTOR and Ras-MAPK signaling pathways (Kumar et al., 2005), underscoring the importance of

GTPase signaling in dendrite growth and maintenance. These same studies demonstrated that increased Ras activity (which has been linked to increased neural activity (Qin et al., 2005)) led to an increase in both the total number and the overall complexity of hippocampal dendritic arbors. Further evidence for the role of Ras in dendritic development was observed in vivo, as mice that had a constitutively active form of Ras, RasV12, were shown to have more complex arbors (Alpar et al., 2003). Interestingly, Ras is also known to play a role in inhibiting arborization, as overexpression of the Ras guanine nucleotide exchange factor (GEF) v-KIND reduced arbor complexity. The role of Ras and its regulation of dendritic arborization is further discussed in Chapter 3. GTPase signaling will be discussed further in Section 1.12. 25

1.9.1.3 Intracellular Mechanisms Regulating Dendritic Arborization

Extracellular factors typically lead to cascades that need intracellular messengers to exert cellular effects. Notably, extracellular signals regulating dendritic arborization activate a variety of signaling pathways, including GTPases, protein kinases, and protein phosphatases (reviewed in Urbanska et al., 2008; Dong et al., 2015). As a typical example, BDNF stimulates the simultaneous activation of signaling cascades involving PI3K-Akt and Ras-MAPK (McAllister et al., 1999), known regulators of dendritic branching as evidence through gain-of-function and loss-of-function experiments (Jaworski et al., 2005; Kumar et al., 2005). Another important intracellular molecular mediator of dendritic growth is the Calcium/Calmodulin-dependent protein kinase II (CamKII), which is proposed to translate calcium influx into long-lasting structural change in dendrites as altering CamKII levels in Xenopus alters the stability of dendritic arbors, such that CamKII inhibition increases dendritic growth (Zou & Cline, 1999).

Dendritic growth is also dependent on targeted changes in the cytoskeleton (reviewed in

McAllister, 2000), which in dendrites is primarily composed of microfilaments and microtubules which are dynamic polymers of actin or tubulin. Some of the most powerful regulators of intracellular regulation to signaling pathways involving the actin cytoskeleton are GTPase proteins (Luo et al., 1997), and it has been postulated that small GTPases work together to regulate dynamic dendritic growth due to their positioning to translate and/or transduce cellular activity in response to synaptic activity (McAllister, 2000). One mechanism for this regulation is palmitoylation, which may position GTPases in specific subcellular domains to interact with downstream regulators.

1.9.2 Palmitoylation in Dendritic Arborization

Despite the clear role of palmitoylation in nervous system function, there is little known about the role of palmitoylation in the process of dendrite morphogenesis. Recent work has shown that palmitoylation of the 190 kDA Ankyrin G variant stabilizes pyramidal neuron basal, but not apical, dendrites by stabilizing it in dendritic spines (Piguel et al., 2019). However, many of the proteins that are involved in the formation of neuronal processes and spines, and that regulate neuronal activity, are known to be palmitoylated, including NCAM, CDC42, Ras, and paralemmmin (for review see Fukata & Fukata, 2010). It has been suggested that palmitoylation of LIM Kinase 1 (which regulates actin dynamics by phosphorylating and inactivating cofilin

(Meng et al., 2002) and cdc42 (which regulates actin dynamics and filopodial formation) may regulate early stages of dendritic development (Montersino & Thomas, 2016), based on the observation that they are detectable in cultured neurons prior to spine formation (George et al.,

2015, Kang et al., 2008). This is further substantiated by the finding that Calcium/calmodulin- dependent protein kinase-1 gamma (CaMKI-γ), another palmitoylated protein, is involved in regulation of the Rac/CdC42-PAK-LIMK1 pathway (Takemoto-Kimura et al., 2008) and has previously been shown to play a role in dendritic arborization (Redmond et al., 2002). The effects of CaMKI-γ knockdown on dendritic growth are unable to be rescued by a palmitoylation-deficient variant, suggesting that palmitoylation is critical for the dendritic outgrowth (Takemoto-Kimura et al., 2008). Previous work has also shown that the palmitoylation of GAP-43 and paralemmin can induce filopodia formation and enhance dendritic and axonal branching (Gauthier-Campbell et al. 2004). A number of GTPases are also known to be palmitoylated, such as Ras, and it has been suggested that Ras palmitoylation is necessary for activation and signal transduction (Song et al., 2013). 27

1.10 Protein Trafficking in Dendrites

All cells require the trafficking to place cellular constituents, such as membrane proteins and secreted factors, at appropriate subcellular domain and localization to ensure proper functioning (reviewed in Ehlers, 2013). The complexity of neurons, arising from dendritic and axonal arborization, gives rise to highly compartmentalized function that necessitates both local synthesis and long range trafficking of proteins. Indeed, it has been suspected for decades that protein synthesis and post-translational modification of proteins required for dendritic growth is facilitated by polyribosomes localized at or near dendritic spines (Steward et al., 1988; Eberwine,

1999). Ribosomes localize to the endoplasmic reticulum (ER), and as with other cell types, the

ER is distributed throughout the cytoplasm and extends throughout the dendrites and soma, where it can function as a component of Ca2+ signaling (Rose & Konnerth, 2001) and as a site for protein and lipid synthesis (Borgese et al., 2006).

Progression of neuronal protein “cargo” requires the proper folding of proteins in the ER, accompanied by a number of post-translational modifications. Indeed, palmitoylation can occur at the ER, and subset of the aforementioned zDHHC proteins localize to the ER (Ohno et al.,

2006). Once proteins are properly synthesized, they accumulate at ER exit sites (ERES) present throughout the soma and dendrites (Horton & Ehlers, 2003; Aridor et al., 2004). ER “cargo” exists in the form of vesicles, which then fuse with the ER-Golgi intermediate compartment

(ERGIC) are thought to be responsible for shuttling proteins between the ER and Golgi (Krijnse-

Locker et al., 1995; Hanus et al., 2014). Once proteins arrive at the Golgi they are processed through the cis, medial, and/or trans compartments where further modifications can occur. As a typical example, Ras proteins are first targeted to the ER and Golgi by prenylation, but further modifications such as palmitoylation lead to plasma membrane associations (Choy et al., 1999; 28

Goodwin et al., 2005). In a typical mammalian cell, the ER and ERGIC are distributed throughout the cell, while the Golgi is perinuclear (Lee et al., 2004). Neuronal dendrites require their own trafficking and synthesis machinery distinct from the soma. As such, ERGIC is present in dendrites (Bowen et al., 2017) and a discrete compartment in dendrites termed Golgi outposts or Golgi satellites has been postulated to exist (Horton & Ehlers, 2003; Horton et al., 2005;

Pierce et al., 2001). These outposts are thought to allow for both the “long-range” trafficking of proteins from the soma to ER exit sites and a more “local” transport method utilizing Golgi outposts and ER exit sites (Figure 1.9). These local Golgi outposts contribute to the observation

Figure 1.9 Dual Modes of ER-to-Golgi Transport

In all neurons, cargo exits the ER (which is contiguous within the neuron) at specialized ER exit sites which are located in both the soma and dendrites, and traffics to or from the Golgi (1). In some neurons, Golgi is only present in the soma, necessitating long-range trafficking (2). In other neurons, dendritic Golgi outposts function in local secretory trafficking (3). Reproduced from Horton & Ehlers, 2000 (© 2003 Society for Neuroscience)

that secretory trafficking is required for dendritic growth and maintenance as disrupting Golgi function with drugs such as brefeldin A, or changing protein expression to block Golgi trafficking, blocks dendritic growth (Horton et al., 2005). These changes also lead to a reduction/retraction of existing dendrites, which suggests that there is a need for Golgi trafficking to maintain dendritic morphology. Interestingly, the organelles of the secretory pathway tend to concentrate at dendritic branch points and in cultured mammalian hippocampal neurons Golgi outposts predominantly localize to branch points (Horton et al., 2005; Ori-

McKenny et al. 2012). Indeed, it has been proposed that these outposts provide membrane for the nucleation and growth of dendritic branch points (Ye et al., 2007; Ori-McKenny et al., 2012).

However, some have suggested that mammalian neurons, and hippocampal neurons specifically, lack Golgi outposts or satellites (Krijnse-Locker et al., 1995; Torre and Steward,

1996; Gardiol et al., 1999; Hanus et al., 2016), and evidence suggests that a number of membrane proteins (including GluA1 and Nlg1) bypass the Golgi on their way from dendritic

ER to the cell surface, and many AMPARs on the cell surface have immature glycosylation

(Bowen et al., 2017). However, a novel probe/Golgi tracker known as pGolt has been used to identify and determine that a Golgi-related organelle exists in all dendrites of pyramidal neurons in close proximity to the ERGIC and retromer, containing glycosylation machinery but lacking protein components for sorting and organization of Golgi cisternae (Mikhaylova et al., 2015).

Together, these studies suggest that there is controversy over the presence or identity of a Golgi- like satellite, outpost, apparatus, or compartment for modification in neuronal dendrites, and further studies are warranted.

1.11 Synapse Structure and Function

As neurons grow and develop the spatial location of dendrites and the arborization pattern determines how neurons connect within a network, allowing the communication and exchange of chemical and electrical information. This occurs at specialized structures called synapses, which allow communication between discontinuous neurons via regulated secretion and exocytosis of chemical intermediate signals, or neurotransmitters. While electrical and chemical synapses (Figure 1.10) exist, the focus of this dissertation will be chemical synapses.

Figure 1.10 Electrical vs Chemical Synapses

Electrical synaptic transmission is mediated by clusters of intercellular channels (gap junctions) that directly connect the interior of two adjacent cells, enabling the bidirectional passage of electrical current carried by ions and/or intracellular messengers such as small metabolites. Chemical synapses, which evolutionarily predate electrical synapses, require more sophisticated molecular machinery to regulate neurotransmitter release and uptake. Depolarization of the presynaptic compartment via the arrival of an action potential leads to synaptic vesicle fusion and release. Neurotransmitters flood the synaptic cleft and a combination of ionotropic and metabotropic receptors detect and translate the neurotransmitter release into postsynaptic events.

Chemical synapses are polarized junctions that, for simplistic purposes, allow the flow of information in a single direction (although, at a more realistic level, allow some level of backpropagation (Williams & Stuart, 2000)). Initial electron microscopic studies of rat synapses led to the distinction of two distinct subtypes: Type I synapses are found mostly on dendrites and

dendritic protrusions called spines, and are characterized by a thickening especially at the postsynaptic membrane, while Type II synapses predominate on axosomatic areas and do not show an asymmetric thickening of the membranes (Gray, 1959). It was later determined that

Type I synapses were excitatory, making the postsynaptic cell more likely to fire, and Type II synapses were inhibitory, making the postsynaptic cell less likely to fire. As the postsynaptic cell receives inputs, the temporal and spatial summation of these inputs determines whether the cell will engage in an all-or-nothing action potential. It has been estimated that each neuron in the mammalian brain connects to other neurons by as many as 10,000 synapses (Sheng et al., 2012).

Neurons must therefore integrate and regulate these extensive synaptic inputs in order to generate an appropriate output.

1.11.1 Synaptogenesis

The organization, development, and maintenance of synapses play a critical role in directing circuit connectivity and establishing a mature network. This process is the culmination of multiple developmental events, including cell migration, axon guidance, and dendritic growth, as well as the processes of synapse formation and regulation. Synaptogenesis is therefore the culmination of two distinct steps: synaptic specificity and synaptic assembly (Colon-Ramos,

2009). Synaptic specificity refers to the process of finding the right place to form a synapse, such as finding an appropriate partner and forming a synapse at the right subcellular compartment.

Synaptic assembly refers to the generation of presynaptic active zones and postsynaptic scaffolds to facilitate neurotransmission. Synaptic specificity is believed to be mediated by cell adhesion molecules, which serve to identify correct pre- and post-synaptic partners (Waites et al., 2005;

Bamji, 2005). While these have been discussed extensively elsewhere (Dalva et al., 2007), briefly the adhesion molecules are believed to fall into four functional categories: (1) stabilizing 32

a nascent synapse by linking synaptic partners, (2) directing recognition of targets through adhesion, (3) regulating differentiation of pre- and/or post-synaptic specializations, or (4) modulating synaptic structure and function. Specifics of some cell adhesion molecules in synapse formation will be discussed with regards to excitatory synapses (1.11.2) and inhibitory synapses (1.11.3).

It was previously thought that tissue culture systems can only truly offer insights into assembly, but not specificity events, as the neuronal dissociation disrupts previously held architecture, removing much of the positional information mediating synaptic specificity. True developmental investigation into synapse specificity requires alternative preparations, and synapse specificity is reviewed elsewhere (Margeta & Shen, 2010). However, dissociated neurons still form synapses (Vaughn, 1989), and indeed it has been demonstrated that even in dissociated hippocampal culture systems DG neurons preferentially synapse onto CA3 neurons and CA3 neurons preferentially synapse onto CA1 neurons (Williams et al., 2011). Together, this demonstrates that investigation into the mechanisms underlying synaptic assembly, formation, maintenance, and elimination are amenable to tissue culture.

1.11.2 Excitatory Synapses

The major classes of excitatory synapses in the mammalian brain use glutamate as a neurotransmitter, and are thus glutamatergic synapses. Glutamatergic synapses promote neuronal depolarization, increasing the likelihood that the neuron will fire. At a simplistic level, the presynaptic compartment of an excitatory synapse is characterized by the presence of hundreds of neurotransmitter filled synapse vesicles (SVs), dedicated to releasing glutamate via fusion with the cell membrane. Glutamate then passes the synaptic cleft and binds with the postsynaptic compartment, localized primarily to dendritic membranous protrusions known as spines. These 33

spines contain glutamate receptors dedicated to receiving and transducing the signals from presynaptic glutamate release (Figure 1.11). The following sections provide more details on this process.

Figure 1.11 Schematic Represenation of an Excitatory Synapse

A simplistic overeview of excitatory synapses. At the most basic level (left), presynaptic boutons release glutamate into the synaptic cleft, and the post synaptic spine responds by causing depolarization with influx of ions. Some major protein classes are depicted (right). The pre- and post-synaptic terminals are tethered by cell adhesion molecules. At the presynaptic termina, vesciles contain glutamate and glutamate transporters. Glutamate is released into the synaptic cleft and detected by gluatamate receptors, predominantly AMPAR and NMDAR, are localized to the postsynaptic membrane and stabilized by scaffold proteins such as PSD-95. Activity leads to protein signaling cascades, such as Ras signaling.

1.11.2.1 Composition of the Excitatory Presynaptic Compartment

The presynaptic compartment of excitatory synapses, as mentioned above, is dedicated to releasing the neurotransmitter glutamate via the fusion of synaptic vesicles. Synthesis of glutamate from glutamine occurs directly in the cytoplasm of presynaptic terminals and is catalyzed by the enzyme phosphate-activate glutaminase (Takamori, 2006). The uptake of the glutamate into the compartmentalized vesicles is driven by a proton-dependent electrochemical gradient across the membrane of the vesicles, which is mediated by a family of presynaptic,

excitatory synapse-specific proteins known as VGlut (vesicular glutamate transporter, Takamori et al., 2000) which includes VGlut1, VGlut2, and VGlut3. Structurally, the presynaptic compartment is comprised of hundreds of these neurotransmitter-filled vesicles and by a region termed the “active zone”, a specialized region of the plasma membrane where vesicles dock, fuse, and release their contents into the synaptic cleft (Ziv & Garner, 2004). A subdomain of this active zone exists, known as the cytoskeletal active zone, which has an electron-dense matrix of proteins arranged in a lattice-like structure and tightly associated with the actin cytoskeleton

(Phillips et al., 2001). This is proposed to function as a platform for the regulation of synaptic vesicle movement to the plasma membrane and subsequent vesicle fusion and neurotransmitter release (Ziv & Garner, 2004; Phillips et al., 2001). The cytoskeletal active zone exists as a type of “scaffold” for the presynaptic compartment, and a number of proteins have been shown to localize or associate with this area, including cell adhesion molecules, structural and organizing molecules, and the protein and cellular machinery involved in exo- and endo-cytosis of synaptic vesicles (Phillips et al., 2001; Shupliakov & Brodin, 2010).

Within the presynaptic terminal there are hundreds to thousands of synaptic vesicles, however these vesicles are not functionally identical. According to their recruitment by differing electrical or chemical signals, synaptic vesicles can be functionally divided into three distinct pools: the readily releasable pool, the recycling pool, and the resting pool (sometimes referred to as the reserve pool) (Rizzoli & Betz, 2005; Denker & Rizzoli, 2010; Alabi & Tsien, 2012). The readily releasable pool contains approximately 2% of the total synaptic vesicles, and these are docked at the synaptic membrane and immediately available for exocytotic release. The recycling pool is comprised of approximately 20% of the available synaptic vesicles, and is recruited after the readily releasable pool has been depleted by moderate stimulation. The 35

remaining majority (almost 80%) belong to the resting pool, which is only recruited upon intense, high frequency stimulation or prolonged, sustained low frequency stimulation

(Fernandez-Alfonso & Ryan, 2008; Ikeda & Bekkers, 2009). Interestingly, the recycling pool and resting pool appear to be largely intermixed within the terminal (Denker & Rizzoli, 2010), suggesting that the vesicles may be partitioned by molecular identity or protein-protein associations.

A number of presynaptic proteins are also substrates for palmitoylation, including the

SNARE proteins VAMP2, SNAP-25, and syntaxin I, the Ca2+ sensor synaptotagmin, and the molecular chaperone cysteine-string protein (CSP) (reviewed in Prescott et al., 2009). Together this suggests that palmitoylation is an important regulator of presynaptic function and compartmentalization.

1.11.2.2 Composition of the Excitatory Postsynaptic Compartment

The postsynaptic compartment is generally thought to be localized to the tip of spines, protrusions of membranes from dendrites that are dedicated to receiving and transducing the signals from the release of presynaptic vesicles. In the case of excitatory synapses, this means that the postsynaptic membrane has glutamate receptors positioned on the membrane to bind glutamate from the synaptic cleft. The dendritic protrusions are often referred to as simply

“spines”, and will be referred to as such through this dissertation. The morphology of spines can be highly variable, but are commonly classified into three main categories: thin, mushroom, and stubby (Figure 1.12) (Peters & Kaiserman-Abramof, 1970). Thin spines, as the name suggests, have a long, thin neck with a small bulbous head. Mushroom spines have a shorter, distinct spine neck and a larger head that gives them a “mushroom”-like shape. Stubby spines are protrusions that lack a neck, but have a distinct bulbous head. The size and composition of spines varies 36

among brain areas, an individual dendrite can have multiple spine types, and spine morphology can change rapidly through activity-dependent and –independent mechanisms (reviewed in

Rochefort & Konnerth, 2012). An additional category of spine formation exists, termed

Figure 1.12 Classification of the Most Common Spine Morphologies

Representative confocal image of a dendrite with spines, illustrating the diverse morphology of spines on a given dendrite. Examples of the major classes of spines (thin, mushroom, and stubby) are shown. filopodia, although filopodia are often described as a completely separate entity. Filopodia have no bulbous head, are normally found on developing neurons, and are to be a transient, dynamic structure searching for appropriate synaptic connection so they may receive input and develop into spines (Fiala et al., 1998). The spine contains a 30-40 nm thick specialized domain known as the postsynaptic density (PSD), an electron-dense matrix that contains over 1000 different proteins (Collins et al., 2006). Additionally, spines are rich in actin filaments that exhibit dynamic activity during the development of spines and in maintenance. Indeed, actin is required for synapse formation, as young neuronal cultures treated with Latrunculin A (a molecule that binds and sequesters actin) disrupted synapse development and assembly, while mature neurons treated with Latrunculin A showed no significant impact to synapse number (Zhang & Benson, 37

2001). This suggests that once matured, synapses transition to more stable structures. The spine is therefore an actin-rich domain with a distinct post-synaptic density (PSD), which coordinate the structure and function of the synapse. Here I will focus on some key proteins of the excitatory postsynaptic PSD that are substrates for palmitoylation, including PSD-95, A-kinase anchoring proteins (AKAPs), α-amino-3-hydroxy-5-methyl-4-isoxazoleproprioninc acid type receptors (AMPAR), N-methyl-D-aspartate type receptors (NMDAR), and cell adhesion molecules (Figure 1.13).

Figure 1.13 Molecular Architecture of the Excitatory Synapse

Glutamate is released from the presynaptic bouton and diffuses through the synaptic cleft where it can bind to postsynaptic receptors. NMDARs and AMPARs are anchored by scaffolding proteins such as A-kinase anchoring protein (AKAP) and guanylate-kinase associated protein (GKAP) and post-synaptic protein 95 (PSD-95) (discussed below). Cell-cell contact is mediated by cell adhesion molecules, such as Ephrins, neuroligins, and cadherins. Only major protein classes are shown, and others important for synaptic regulation are discussed in the following section. Used with permission from Kennedy, 2013.

1.11.2.2.1 PSD-95

PSD-95 (Post Synaptic Density Protein – 95 kDa) is the most abundant scaffolding protein in the PSD with a typical PSD containing as many as 200-300 PSD-95 molecules (~25% content), a number that far exceeds the number of glutamate receptors in the PSD (Cheng et al.,

2006). PSD-95 binds many other key post-synaptic proteins, including NMDARs (Kornau et al.,

1995; Niethammer et al., 1996) and AMPARs (Nicoll et al., 2006), the glutamate receptors responsible for transducing glutamate released into the synaptic cleft. PSD-95 has three type I

PDZ domains localized to its N-terminus which facilitate the binding to short PDZ-binding motifs on the C-terminus of many other proteins, and PDZ domains regulate multiple biological processes including transport, ion channel signaling, and signal transduction (reviewed in Lee &

Zheng., 2010). The C-terminus of PSD-95 has a Src homology 3 (SH3) domain, known to bind proline rich domains of other proteins, and SH3 domains are involved in substrate recognition, membrane localization, and kinase activity regulation (reviewed in Kurochkina & Guha, 2013).

The variety of domains through which PSD-95 can associate with and bind to other molecules in the PSD highlights its key role in PSD organization. Indeed, PSD-95 plays an important role in synapse function, as overexpression of PSD-95 increases glutamate receptor clustering and increases the number of postsynaptic spines (El-Husseini et al., 2000), while knockdown reduces glutamate receptor clustering at the postsynaptic compartment (Elias et al., 2006).

PSD-95 is palmitoylated at Cys3 and Cys5 (Topinka & Bredt, 1998), two sites that are essential for PSD-95 to multimerize and bind multiple membrane proteins and PSD-95 clustering

(Hsueh & Sheng, 1999; Craven et al., 1999). Indeed, while prenylation of PSD-95 can form clusters, these clusters tend to be misformed and mislocalized (El-Husseini & Bredt, 2002), suggesting that the palmitoylation of PSD-95 was important for proper cellular trafficking and 39

localization. Subsequent work determined that two subfamilies of DHHC proteins, namely zDHHC3/7 and zDHHC2/15, palmitoylated PSD-95, with zDHHC2 responsible for local palmitate cycling within dendritic spines and zDHHC3 responsible for PSD-95 palmitoylation related to its trafficking from the soma (Noritake et al., 2009). This work is further expanded in

Chapter 2, where we demonstrate that PSD-95 trafficking by zDHHC15 palmitoylation is important for excitatory synaptic formation and/or maintenance. It is interesting to note that research has suggested that PSD-95 requires newly synthesized protein palmitoylation mediated perhaps by one group of zDHHC proteins, and another group regulates continuous palmitoylation/depalmitoylation cycles within a spine (Fukata et al., 2013). Together, this work suggests that local palmitoylation may be critical for PSD reorganization, allowing synapses to respond structurally and functionally to cellular activity and demand. For a more thorough review of PSD-95 palmitoylation, please refer to Matt et al (2019).

1.11.2.2.2 AMPA Receptors

One of the key functions of the PSD and PSD-95 is the clustering and transport of membrane-bound glutamate receptors. AMPA receptors mediate the majority of fast excitatory neurotransmission in the brain, and at hippocampal synapses these receptors are largely Ca2+- impermeable with fast signaling kinetics allowing them to mediate rapid basal synaptic signaling

(Lu et al., 2009). AMAPRs consist of four subunits (GluA1-GluA4), which are homologous to each other, and the responsiveness of the AMPAR is dependent upon subunit composition

(Greger & Esteban, 2007). The GluA2 subunit exerts the most impact on the biophysical properties of AMPAR, as the GluA2 subunit confers Ca2+-impermeability, leading to GluA2- containing AMPARs having a linear current-voltage relationship while GluA2-lacking AMPARs have an inwardly rectifying current-voltage relationship (Anggono & Huganir, 2012). In the 40

mature hippocampus, the majority of receptors are comprised of GluA1 and GluA2 heteromers, suggesting they are not a major source of calcium during postsynaptic activity; however,

AMPARs are highly sensitive and responsive to activity and can modulate synaptic strength and plasticity through their trafficking to and from the membrane (Greger & Esteban, 2007).

All four subunits feature two palmitoylation sites, one in the second transmembrane domain and the other in the C-terminal tail (Hayashi et al., 2005). For GluA1 and GluA2 increasing the palmitoylation of the second transmembrane domain cysteine lead to accumulation and retention in the Golgi, while C-terminal palmitoylation did not affect their steady state expression (Hayashi et al., 2005). However, rendering the C-terminal site palmitoylation deficient through a serine mutation affected GluA1 and GluA2 internalization/endocytosis following glutamate or NMDA treatment (Hayashi et al., 2005).

Numerous other reports have demonstrated that palmitoylation of AMPARs or AMPAR associated proteins play a key role in regulating the synaptic expression and localization of

AMPARs (Lin et al., 2009; Thomas et al., 2012, 2013; Thomas & Hayashi, 2013) and that

AMPAR palmitoylation cycles are themselves regulated in an activity-dependent manner (Kang et al., 2008). Together, this suggests that palmitoylation is essential for dynamic control of synaptic expression, trafficking, and regulation of AMPARs.

1.11.2.2.3 NMDA Receptors

A second class of receptors that mediate excitatory synaptic transmission are NMDARs.

In contrast to the regulation of fast synaptic transmission of AMPARs, NMDARs play a more subtle role as they have been termed “coincidence detectors” for their role in detecting the close temporal occurrence of glutamate release and postsynaptic depolarization (Hunt & Castillo,

2012). NMDARs have a voltage-sensitive block regulated by extracellular Mg2+, a high 41

permeability to Ca2+, and relatively slow activation kinetics (Hunt & Castillo, 2012).

Depolarization leads to Mg2+ removal and subsequent binding of glutamate and glycine, both of which are required for opening, while the calcium permeability makes NMDARs more likely to contribute to calcium flux to the postsynaptic cell when active (Huganir & Nicoll, 2013). Similar to AMPARs, NMDARs are tetramers, however the latter are composed predominantly of two

GluN1 and two GluN2 subunits (Traynelis et al., 2010; Gray et al., 2011). The GluN1 subunit contains the binding site for glycine, while the GluN2 subunits confer differences in receptor kinetics and contain the glutamate-binding site (reviewed in Dingledine et al., 1999). Within the hippocampus, NMDAR also contribute to a phenomenon known as “silent synapses”, synapses in which an excitatory postsynaptic current (EPSC) is absent at resting membrane potential but becomes active upon depolarization, which are believed to be critical for synaptic plasticity (for extensive review, see Kerchner & Nicoll, 2008)

Both GluN2A and GluN2B contains palmitoylated cysteines, with one cluster located in the membrane-proximal region of the C-terminus and another cluster located in the more distal

C-terminus (Hayashi et al., 2009). This palmitoylation appears to be activity-dependent, with sustained activity leading to the reduced palmitoylation of both substrates and decreasing synaptic activity with TTX leading to increased palmitoylation (Hayashi et al., 2005). Mutation of the cysteines in the membrane-proximal region in GluN2 prevents NMDAR internalization

(due to increased phosphorylation of the receptor) as well as reduces synaptic NMDAR currents

(Mattison et al., 2012). Subunits of NMDAR show significant age-related declines of protein expression in the hippocampus (Magnusson et al., 2002, 2010), and recent work has also established that an age-related increase in palmitoylated GluN2A and GluN2B (as well as Fyn,

PSD-95, and APT1) is associated with poorer reference memory and/or executive functions. 42

Together, this research demonstrates that palmitoylation of NMDARs has diverse implications for proper neural functioning.

1.11.2.3 Ras

An important regulator of excitatory synapses is the G-protein Ras (further discussed in

1.12). Ras is downstream of the CamKII pathway (see 1.9). One of the interactors for Ras is the protein synGAP (synaptic Ras GTPase-activating protein), which associates with PSD-95 and is phosphorylated by CamKII which decreases synGAP activity (Chen et al., 1998; Kim et al.,

1998). Ras has multiple downstream signaling pathways, and its position in the synapse and regulatory function positions it to respond in an activity-dependent manner. Indeed, Ras-ERK signaling is required for LTP in hippocampal synapses (English & Sweatt, 1997) and constitutively active Ras increases excitatory currents while dominant negative Ras decreases excitatory currents (Zhu et al., 2002). Further experiments determined that Ras controls synaptic delivery of AMPARs during LTP (Zhu et al., 2002; Qin et al., 2005; Hu et al., 2008), as Ras-

MAPK-ERK signaling stimulates phosphorylation of GluR1 and Glur2.

1.11.3 Inhibitory Synapses

While the vast majority (~80-90%) of synapse are excitatory (Heller et al., 2012), inhibitory synaptic transmission is critical for ensuring proper brain function. This has led to some researchers suggesting that gephyrin, the major scaffold protein at inhibitory synapses, is a master regulator of neuronal function (Tyagarajan & Fritschy, 2014). The principal neurotransmitters at inhibitory synapses are γ-aminobutyric acid (GABA) and glycine, which are released into the synaptic cleft, bound by the postsynaptic ligand-gated Cl- channels GABA and glycine receptors, respectively, leading to influx of Cl- to the cell and eliciting a hyperpolarizing current (Figure 1.14). 43

Figure 1.14 Schematic Representation of an Inhibitory Synapse

A simplistic overeview of inhibitory synapses. At the most basic level (left), presynaptic boutons release GABA into the synaptic cleft, and the post synaptic spine responds by causing hyperpolarization with influx of ions. Some major protein classes are depicted (right). The pre- and post-synaptic terminals are tethered by cell adhesion molecules. At the presynaptic termina, vesciles contain GABA and GABA transporters (VGAT; vesicular GABA transporter). GABA is detected on the postsynaptic membrane by GABA receptors, which are anchored through the scaffolding molecule gephyrin. TC10 and Collybistin are important for gephyrin clustering and will be dicussed below and in Chapter3. Note that inhibitory synapses are not spines, but most are located on dendrite shafts and the soma.

Intriguingly, GABAergic and glycinergic inhibitory synapses can, under the right circumstances, also be excitatory. This is developmentally regulated by the expression of the importer of Na+ /K+ /Cl− (NKCC1), which when upregulated this leads to an increase in cytosolic

Cl- concentrations. Hence, channel opening results in depolarization instead of hyperpolarization

(Ben-Ari, 2002). Later, upregulation of the K+ /Cl− exporter (KCC2) coupled with a downregulation of NKCC1 renders GABA- and glycinergic synapses inhibitory (Birke &

Draguhn, 2010, Blaesse et al., 2009). Inhibitory synapses are generally focused on as

counterbalances to excitatory neurons, and indeed the balance of excitation and inhibition (see

1.8.4) is required for proper functioning of the nervous system.

1.11.3.1 Interneurons

Interneurons are a class of inhibitory neurons that project to various sites on a target neuron, regulating synaptic transmission throughout the dendritic tree and at the soma (Draugh et al., 2008; Freund & Katona, 2007; Kullman et al., 2005). Inhibition at dendrites seems to be a local process that serves to regulate excitatory input, while perisomatic inhibition is a primary regulator of postsynaptic output (Freund & Katona, 2007). A number of different classes of inhibitory interneurons have been identified (Kepecs & Fishell, 2014), but the two major subtypes are parvalbumin (PV) and somatostatin (SST) expressing interneurons, accounting for nearly 40% and 30% of GABAergic interneurons respectively (reviewed in Rudy et al., 2011).

PV interneurons primarily target the perisomatic domain of pyramidal cells and mediate fast and powerful inhibition, while SST interneurons mainly target distal dendrites of pyramidal cells and exert a more focal inhibition of local synaptic inputs (Figure 1.15) (Hangya et al., 2014; Hu et al., 2014). For a more extensive review of inhibitory interneurons refer to Rudy et al. (2011), and the remainder of this section will be devoted to discussing the typical inhibitory synapse composition of a pyramidal cell.

1.11.3.2 Composition of the Inhibitory Presynaptic Compartment

Presynaptically, inhibitory synapses are functionally and structurally similar to excitatory synapses. The molecular machinery involving SV release are essentially the same, but release neurotransmitters (GABA and glycine) that serve to promote hyperpolarization (or reduce depolarization). GABAergic terminals synthesize GABA through the decarboxylation of

Figure 1.15 Interneurons

Schematic depication of the majory inhibitory interneuron subtypes and their functions. Parvalbuminterneurons project onto the perisomatic region of pyramidal neurons (1) where they mediate fast and powerful inhibition, preventing cells from firing. Somatostatin inter neurons project onto distal dendrites, where they can affect local synaptic summation and firing (2). Pyramidal neurons also project back onto the interneurons to regulate their function. Adapted with permission from Sudhof, 2008.

gluatamate via L-glutamic acid decarboxylase (GAD), and this is mediated through two enzymes, GAD65 and GAD67 (Bu et al., 1992). Once synthesized, GABA is then packaged into

SVs by the presynaptic transporter protein VGAT (vesicular GABA transporter; McIntire et al.,

1997) for release into the synaptic cleft.

1.11.3.3 Composition of the Inhibitory Postsynaptic Compartment

The postsynaptic compartment of inhibitory synapses also shares similar molecular machinery, which functions in a similar manner (e.g. scaffolding proteins help to anchor and transport receptors, cell adhesion molecules maintain pre- and post-synaptic approximations

(Figure 1.16)). The focus for the inhibitory postsynaptic compartment in this section will be

inhibitory specific proteins, including scaffolding protein gephyrin, the receptors GABAR and

GlyR, the cell adhesion molecule neuroligin, as well as a GTPase known to be involved in gephyrin cluster formation, TC10, and its effector, collybistin.

Figure 1.16 Molecular Archtitecture of the Inhibitory Synapse Some of the key proteins in inhibitory synapse assembly. Gephyrin creates a scaffold for GABA receptors, and collybistin (Cb) is involved in this process. Cb also interacts with neuroligin 2 (NL2). Mechanistic details are explained in the text following. Adapted from Brose (2013).

1.11.3.3.1 Gephyrin

The major inhibitory scaffolding protein is gephyrin, which helps to organize the inhibitory architecture and is critical to GABA receptor (GABAR) clustering (Luscher et al.,

2011). The central role of gephyrin in inhibitory synaptic transmission has been elucidated through experiments involving gephyrin KO mice or reduced gephyrin expression, which has demonstrated that loss of gephyrin results in a decrease in GABARs at inhibitory synapses

(Kneussel et al., 1999; Levi et al., 2004; Essrich et al., 1998). Additionally, the loss of GABA 47

subunits can also result in the loss of gephyrin from synaptic sites, suggesting there is an interdependence between gephyrin and GABAR clustering (Essrich et al., 1998). Gephyrin has been shown to be a target for post-translational modification, including phosphorylation

(reviewed in Tyagarajan & Fritschy, 2014) and palmitoylation (Dejanovic et al., 2014). The phosphorylation of gephyrin suggests a role for specific kinase networks in regulating gephyrin function. Ser270 and Ser268 are targeted by glycogen synthase kinase 3β and extracellular signal regulated kinases 1 and 2 (ERK1/2) respectively, and these can synergistically influence

GABAergic currents through changes in the clustering of gephyrin (reviewed in Tyagarajan &

Fritschy, 2014). Palmitoylation has also been demonstrated to be critical to regulating

GABAergic strength (Dejanovic et al., 2014). It is believed that palmitoylation may anchor gephyrin to the membrane or facilitate protein-protein interactions and aid in the recruitment of other inhibitory synapse specific proteins such as neuroligin and collybistin and facilitate

GABAR and GlyR insertion and recycling.

1.11.3.3.2 GABAR & GlyR

Both GABAR and GlyR belong to the superfamily of Cys-loop ligand-gated ion channels

(Barnard et al., 1998). There are five GlyR subunits, α1-4 and β, and these form either homomeric (only α subunits) or heteromeric assemblies. GlyRs predominantly assemble into pentamers of GlyRα12-GlyRβ3 (Dutertre et al., 2012; Grudzinska et al., 2005). Both alternative splicing and mRNA editing influence GlyR ligand binding and receptor localization (Betz &

Laube, 2006; Meier et al., 2005; Melzer et al., 2010). Surprisingly, while glycine, taurine, and β- alanine are all able to act as GlyR agonists (Legendre, 2001), glutamate can also potentiate GlyR responses, possibly providing a link between excitatory and inhibitory transmission (Liu et al.,

2010). GlyR is heavily colocalized with gephyrin, indicating a distinct post-synaptic localization. 48

Indeed, the affinity of this interaction is such that gephyrin is highly enriched in native GlyR preparations, and gephyrin was initially regarded as a GlyR subunit (Betz et al., 1991).

GABA receptors (GABARs) fall into two major classes, GABAAR and GABABR.

GABABR are metabotropic, G-protein coupled GABA receptors and mediate slow effects of inhibitory transmission, and are well described elsewhere as they are not immediately relevant to this work (Ulrich & Bettler, 2007). Throughout this dissertation, GABAR will be used to refer to

GABAARs. GABARs have a number of different subunits that, similar to AMPARs, confer different functional properties: α, β, γ, δ, ε, π, θ, and ρ subunits exist, although the classic

GABAR consists of two α subunits, two β subunits, and one γ subunit (reviewed in Tyagarajan

& Fritschy, 2014). The γ2 subunit regulates the subcellular localization of the entire receptor complex, since it is necessary for post-synaptic GABAR clustering despite being dispensable for the transport of GABARs to the plasma membrane (Essrich et al., 1998). Interestingly, the γ2 subunit is palmitoylated and palmitoylation-deficient mutants affect GABAR clustering

(Rathenberg et al., 2004; Keller et al., 2004). Compared to GlyRs, GABARs have lower affinity interactions with gephyrin (Tyagarajan et al., 2011), leading to the observation GABARs are not restricted to synaptic sites, and can be observed at extrasynaptic sites, where they display a subunit composition different from that of synaptic receptors (Brickley & Mody, 2012). This confers a higher transmitter affinity so that they can detect comparatively small ambient concentrations of neurotransmitters of glial origin (Lee et al., 2010) or resulting from release or transmitter spillover after synaptic transmission (Brickley & Mody, 2012, Takazawa &

MacDermott, 2010).

1.11.3.3.3 Collybistin

The protein that would be named collybistin (Cb) was first identified in ascidian embryos

(Satou & Satoh, 1997) and the human homolog (hPEM-2) was found on the X chromosome through a homology search to identify interactors for Rho GTPases (Reid et al., 1999). hPEM-2 was determined to have an N-terminal src homology (SH3) domain, a tandem of Dbl homology domains, and a pleckstrin homology (PH) domain. Kins et al. (2000) identified this same protein in a yeast two-hybrid screen aimed at detecting gephyrin binding partners, and named it collybistin. The same research demonstrated that expression of Cb with gephyrin and GlyR resulted in the clustering and localization of all three components in submembrane microclusters.

Further evidence for the role of Cb in the brain comes from the demonstration that Cb is expressed in postmitotic neurons at the time of neuronal differentiation and synaptogenesis

(Kneussel et al., 2001) and localized to inhibitory synaptic sites (Harvey et al., 2004; Chiou et al., 2011). Additionally, Cb knockout mice display loss of synaptic gephyrin and γ2-subunit

GABARs in the hippocampus with a concomitant reduction in dendritic inhibition and synaptic plasticity (Papadopoulos et al., 2007) and these mice also display an increase in network excitability and plasticity through a decrease of dendritic inhibitory inputs (Jedlicka et al., 2009).

Collybistin has also been shown to interact with NL2, where NL2 is able to relieve the auto- inhibition of the SH3 domain (Poulopoulos et al., 2009). The small GTPase TC10 is another interactor for collybistin, and TC10 stimulates Cb-dependent gephyrin clustering by binding to the pleckstrin homology domain of Cb (Mayer et al., 2013), demonstrating its importance in inhibitory synapse formation and/or maintenance. Further discussion of TC10 can be found in

1.12 and Chapter 3.

Collybistin mutations are also identified in patients with a variety of neurodevelopmental and neuropsychiatric disorders, including aggressive behavior, anxiety, epilepsy, and X-linked intellectual disability (Harvey et al., 2004; Marco et al., 2009; Kalcheuer et al., 2009; Lesca et al., 2011; Shimojima et al., 2011; Lemke et al., 2012; de Ligt et al., 2012). This dissertation will be focused primarily on the latter two, epilepsy and intellectual disability, two disorders that are associated with changes in the excitatory and inhibitory ratio (reviewed in Ramamoorthi & Lin,

2011; Gatto & Broadie, 2010).

1.11.4 Excitatory/Inhibitory Ratio

Together, excitatory and inhibitory synapses interact dynamically to maintain neural networks in a balanced state (reviewed in Haider & McCormick, 2009). The excitatory:inhibitory ratio (E:I ratio) is determined by both the structural (relative densities) and functional (synaptic strength) of each respective synapse type. This is dependent on the synapses themselves, but also on the integration of synaptic signals along a given dendrite, demonstrating that both dendrites and synapses are critically important for determining how and when neurons fire and integrate signals across the nervous system. Indeed, dendritic size and dendritic topology contribute to neuronal firing in pyramidal cells (van Elburg & Ooyen, 2010). Together, this suggests that either dysfunction of excitatory synapses, inhibitory synapses, or dendritic arborization, or some combination of all of these, during development or synaptic plasticity, can contribute to pathological alterations in E:I ratio. It is important for neurons to have a mechanism then to regulate and maintain proper E:I balance.

Stable, long-term network activity is proposed to be governed and maintained within a given dynamic range that prevents hyper- or hypo-signaling, a concept termed “synaptic homeostasis” (Turrigiano & Nelson, 2004; Davis, 2006). Indeed, pharmacological manipulations 51

forcibly altering neuronal activity lead to the cell employing compensatory mechanisms to restore the initial circuit setpoint (Turrigiano et al., 1998). These compensatory mechanisms can occur through adaptation of synaptic strength and/or efficacy, intrinsic membrane excitability, and/or synapse numbers. For example, diminished activity in hippocampal cultures can increase the prevalence of synaptic pairing to enhance activity (Nakayama et al., 2005). Interestingly, recent work has demonstrated that excitatory synaptic scaling and intrinsic excitability are coordinated by the same signaling pathway, whereas inhibitory synaptic scaling is mediated through an independent, potential circuit-level mechanism (Joseph & Turrigiano, 2017).

Many conditions exist that permanently alter E:I balance, indicating failure of these mechanisms or prevention of these mechanism from becoming engaged or maintained, ultimately leading to comprised neural function (reviewed in Sohal & Rubenstein, 2019). Recent work suggests that many genes help to regulate E:I balance, and deficits can occur at the transcriptional level, the translational level, and/or at the level of synaptic protein interactions, including scaffolding proteins, cell adhesion molecules, and receptors (reviewed in Gatto &

Broadie, 2010). Indeed, reduced PSD-95 expression has been shown to cause a redistribution of adhesion proteins from excitatory to inhibitory synapses, shifting E:I balance (Gerrow et al.,

2006; Graf et al., 2004; Levinson & El-Husseini, 2005; Levinson et al., 2010; Prange et al.,

2004). Similarly, manipulations that reduce gephyrin yield depression of GABAAR clustering

(Yu et al., 2007; Yu & De Blas, 2008), demonstrating that E:I ratio imbalance can be maintained or corrected at the level of individual synapses.

Two of the most consistent anatomical correlates of intellectual disability are disruptions to dendrites or spines (Purpura, 1974; Kaufman & Moser, 2000; Valnergi et al., 2012; Verpelli et al., 2011; Calfa et al., 2015; Martin et al., 2015), culminating in a dysfunction resulting from E:I 52

imbalance. While a number of proteins are implicated in dendritic growth, synapse formation, and intellectual disability, for this dissertation we will be focusing on small GTPases and palmitoylation enzymes. Small GTPases, such as the Rho family and the Ras family have been extensively linked to intellectual disability, and are known to play essential roles in neural development and synaptic plasticity (Ras, reviewed in Ye & Carew, 2011; Rauen, 2013; Rho, reviewed in Zamboni et al., 2018).

1.12 Small G Proteins (GTPases)

Small G proteins, also known as small GTPases, are monomeric G proteins and a family of hydrolase enzymes that can bind and hydrolyze guanosine triphosphate (GTP) to guanosine diphosphate (GDP). These proteins are conserved and exist in eukaryotes from yeast to human

(Takai et al., 2001). A typical G-protein is active when bound to GTP, and inactive when bound to GDP, allowing the G-protein to act as a molecular “switch” by cycling between “on” (active) and “off” (inactive) states. GTP-hydrolysis can be accelerated through GTPase activating proteins (GAPs), while GTP exchange is enhanced by guanine nucleotide exchange factors

(GEFs) (Figure 1.17). In this manner, small GTPases are able to regulate a wide variety of cell functions by initiating, terminating, and determining the periods of time for a variety of specific cellular functions. Indeed, they have been referred to as “biotimers” due to their key role in not only spatial but also temporal determination of cell functions, which include gene expression, cytoskeletal reorganization, vesicle trafficking, and cytoplasmic transport (Takai et al., 2001).

While there are more than 100 small G proteins identified that comprise a superfamily (Bourne et al., 1990), they have been structurally classified into at least five distinct subfamilies: Ras,

Rho, Rab, Sar1/Arf, and Ran. We will focus on the Ras and Rho families for the purposes of this thesis. 53

Figure 1.17 Schematic Representation of a Typical Small GTPase

GTPases cycle between an active (GTP-bound) state and an inactive (GDP-bound) state. Guanine exchange factors (GEFs) lead to activation of GTPases and downstream signaling, while intrinsic GTPase activity is stimulated by GTPase-activating proteins (GAPs), which results in the termination of signaling events.

1.12.1 Structural Similarities of Small G Proteins

Comparing the amino acid sequences of the G proteins demonstrates that they are relatively conserved in primary structures and range in sequence homology across species from

30-55%. Among the Ras subfamily, proteins share relatively high amino acid identity (~55%), while the Rho family share a lower amino acid identity (~30%) with Ras proteins (Hall, 1990).

However, they all share consensus amino acid sequences responsible for their specific interaction with GDP and GTP, and for intrinsic GTPase activity (Bourne et al., 1990). The Ras and Rho family also have similar sequences at their C-termini that undergo lipid posttranslational modification, including prenylation and palmitoylation. Furthermore, these lipid modifications are necessary for binding to membranes and regulators, and for their activation of downstream effectors (Takai et al., 2001), suggesting that the control of GTPases might be fine-tuned by

GEF/GAP interactions as a consequence of localization or trafficking accomplished by lipidation.

1.12.2 Ras Proteins

The Ras subfamily of small G proteins have been shown to have roles in diverse cellular processes, including proliferation, cell migration, adhesion, apoptosis, differentiation, gene expression, and cell function (Cox & Der, 2010; Simanshu et al., 2017). In the yeast S. cerevisiae, two membranes of Ras proteins, Ras1 and Ras2, are essential for cell viability.

Mammals have three main Ras proteins, H-Ras, N-Ras, and K-Ras, which can functionally replace Ras1 and Ras2 in yeast (Powers et al., 1984). As only H-Ras and N-Ras are palmitoylated, they will be the focus of this dissertation. A number of follow-up studies using genetic, biochemical, and cell biological techniques in a variety of model organisms and mammalian cell lines elucidated that Ras directly binds to and activates Raf kinase. Raf kinase then induces gene expression through the mitogen-activated protein kinase (MAPK) cascade in response to a variety of extracellular signaling molecules, notably signals activating receptors with intrinsic or associated tyrosine kinase activity (Cox & Der, 2010). However, the first demonstration of classical Ras proteins being involved in neuronal development came from

PC12 cells, in which overexpression led to a neuronal differentiation of these cells (Bar-Sagi &

Feramisco, 1985; Noda et al., 1985). The Ras proteins became of great interest to neurobiologists when they were found to be involved in the molecular cascade governing long-term potentiation

(LTP), as mice lacking a neuron specific GEF for Ras showed impairment of LTP in the amygdala despite normal hippocampal LTP (Brambilla et al., 1997). Additionally, inhibitors of

MAPK cascades, which are activated by Ras, inhibit hippocampal LTP induction (English &

Sweatt, 1997). This lead to further investigation into the role of Ras in neurons, determining that

Ras played a role in numerous cellular processes including dendritic outgrowth (Heumann et al.,

2000; Gartner et al., 2004; Kumar et al., 2005), axonal growth (Fivaz et al., 2008) synapse 55

formation and/or maintenance (Seeger et al., 2005; Xe & Carew, 2010), synaptic delivery of

AMPARs during LTP (Zhu et al., 2002; Qin et al., 2005; McCormack et al., 2006), and presynaptic vesicle release (Cui et al., 2008; Kushner et al., 2005), underscoring the importance of Ras in the nervous system. Indeed, based on clinical observations and behavioral analysis of mouse models, it has been proposed that learning and memory require a “happy-medium” regulation of Ras signaling (Thomas & Huganir, 2004).

Ras palmitoylation in mammalian cells is known to be regulated by zDHHC9 and GCP16

(Swarthout et al., 2015) and the palmitoylation status of Ras alters its trafficking and shuttling between the Golgi and plasma membrane (Rocks et al., 2005). Indeed, Ras cycling may be the most-studied example of palmitoylation sorting and trafficking. Both H- and N-Ras isoforms are first prenylated, allowing transient association with endomembranes (Conibear & Davis, 2010;

Salaun et al., 2010). Subsequently, palmitoylation of Ras occurs on the Golgi, further increasing membrane affinity (Rocks et al., 2010) and targeting Ras to the plasma membrane. The dynamic palmitoylation of Ras is of great importance as Ras has distinct signaling and activity profiles depending on its localization (reviewed in Omerovic & Prior, 2009 and Prior & Hancock, 2012). zDHHC9 and Ras palmitoylation will be further discussed and investigated in Chapter 3.

1.12.3 Rho Proteins

Members of the Rho family of small GTPases have been linked to actin remodeling and

Rho- and Ras-mediated signaling exhibit substantial cross-talk that has important implications for dendrite growth and spine morphological and functional plasticity (Woolfrey & Srivastava,

2016). Indeed, there have been extensive reviews on the Rho subfamily and its regulation of synaptic actin structure and dynamics (Saneyoshi & Hayashi, 2012). However, this thesis will focus on one member of this family, TC10, which has not been studied as extensively as other 56

Rho proteins, such as CDC42. TC10 was first classified based on the basis of sequence homology to other Rho GTPases, and was characterized as regulating cellular signaling to the actin cytoskeleton, activation of gene expression, and other processes associated with cell growth

(Murphy et al., 1999). Further work demonstrated a role for TC10 in vesicle fusion (Kawase et al., 2006), and in the promotion of neurite elongation following injury (Tanabe et al., 2000).

TC10 downstream effectors are known to be PAK, Par6, and N-WASP (Joberty et al., 2000;

Kanzaki et al., 2004) and it is believed to play a role in membrane trafficking and neurite outgrowth, as it leads to neurite growth in PC12 cells and cultured sensory neurons (Abe et al.,

2003; Michaelson et al., 2001; Tanabe et al., 2000). More recently, TC10 was identified as stimulating collybistin dependent gephyrin clustering and regulating GABAergic signaling

(Mayer et al., 2013) and axonal formation and membrane expansion (Dupraz et al., 2009).

Similar to Ras, TC10 is first modified by farnesylation and then palmitoylated, which leads to its trafficking and localization towards the plasma membrane (Michaelson et al., 2001;

Roberts et al., 2008). However, the identity of the PAT that palmitoylates TC10 is unknown.

TC10 palmitoylation will be further discussed in Chapter 3.

1.12.4 Ras and TC10 in Neurodevelopmental Disorders

While there is not a wealth of information available about the role of TC10 in neurodevelopment, there is a strong link with its binding partner collybistin and intellectual disability, epilepsy, and autism (Machado et al., 2016; Papadopoulous et al., 2015; Shimojima et al., 2011; Kelscheuer et al., 2009). Conversely, as Ras has been more extensively studied there is a whole subset of clinical manifestations related to Ras impairments known as RASopathies

(Figure 1.18), a clinically defined group of medical genetic syndromes caused by germline mutations that encode components or regulators of the Ras pathway that affects 1 in 1000 57

individuals (Rauen, 2013). These disorders include neurofibromatosis type I, Noonan syndrome,

Costello syndrome, LEOPARD syndrome, and others, which are extensively reviewed elsewhere

(Rauen, 2013). While each RASopathy exhibits a unique phenotype, they all have a hallmark of neurocognitive impairments, and animal models of RASopathies from Drosophila, zebrafish, and mouse have shown learning and cognitive defects (reviewed in Jindal et al., 2015).

Figure 1.18 The Major Ras Pathways and Implications in RASopathies

A signal, such as the binding of a growth factor to its receptor tyrosine kinase, causes phosphorylation events and binding of an adaptor protein such as SHS/GRB2. These adaptor proteins interact with Ras activators such as SOS1, inducing the dissociation of GDP and the binding of GTP. Ras-GTP activates a cascade of downstream signaling kinases, such as the MAPK pathway. GAP proteins such as neurofibromin I inhibit Ras signaling. RASopathies are developmental pathologies corresponding with mutations in genes at various steps of this pathway.

1.13 Thesis Objective and Hypothesis

Understanding how proteins regulate their localization and signaling is critical to understanding their function, and protein palmitoylation provides a way for proteins to regulate the subcellular trafficking, localization, and function. Palmitoylation plays a critical role in nervous system function, and specifically at synapses through palmitoylation of common scaffolding proteins such as PSD-95. My first objective was to examine the role of zDHHC15, a protein associated with intellectual disability in order to elucidate its functional role in dendritic growth and synapse function. We examined the temporal course of zDHHC15 expression during development, as well as the effects of altering zDHHC15 protein levels using knockout and overexpression constructs in hippocampal cultures. This allowed determination of whether changes in zDHHC15 expression affected neuron morphology, synapse formation, maturation, and balance, and PSD-95 (a known substrate for zDHHC15 palmitoylation) trafficking. This work is of particular importance as zDHHC15 is one of a number of genes on the X chromosome that is implicated in patients with intellectual disability (ID) (Linhares et al., 2016; Martinez et al., 2014), and one report has identified the loss of a zDHHC15 transcript in a female patient with non-syndromic X-linked ID (Mansouri et al., 2005). Additionally, recent work has also implicated truncating variants of PSD-95 in intellectual disability (Moutton et al., 2018).

My second objective was to examine the role of zDHHC9 in dendritic growth and complexity, and in excitatory and inhibitory synapse formation, maintenance, and balance. zDHHC9 is an enzyme that has high expression throughout the brain and is known to palmitoylate the small GTPase, Ras. Previous work has shown that Ras proteins and their subcellular localization are critical for numerous cellular functions; however, few studies have attempted to investigate the role of zDHHC9 in controlling Ras localization to specific 59

compartments or microdomains. Additionally, there have been no confirmed additional substrates for zDHHC9, although it was listed in a screen along with a number of other zDHHCs as potentially playing a role in the palmitoylation of the large conductance calcium and voltage- activated calcium channels (Tian et al., 2010). Despite this, the role of zDHHC9 in neuronal development has not been determined. This work is important as loss of zDHHC9 function has been implicated in X-linked intellectual disability with epilepsy (Raymond et al., 2007; Masurel-

Paulet et al., 2014; Mitchell et al., 2014; Baker et al., 2015; Tzschach et al., 2015; Han et al.,

2017).

Chapter 2: Regulation of dendrite morphology and excitatory synapse formation by zDHHC15

Intellectual disability affects 1-3% of the population (Tzschach et al., 2015) and presents in a male-to-female ratio of nearly 1.5 to 1 (Gecz et al., 2009). As a result of this excess male prevalence, and owing to the identification of many families demonstrating clear X-linked segregation, researchers have focused on genes located on the X chromosome. A 29-year old woman with severe nonsyndromic intellectual disability due to the balanced reciprocal translocation between chromosome X and 15 revealed an absence of ZDHHC15 transcripts

(Mansouri et al., 2005). Here we demonstrate that knocking down zDHHC15 in primary rat hippocampal neurons reduces dendritic outgrowth and arborization, as well as spine maturation.

Moreover, knockdown of zDHHC15 reduces palmitoylation of PSD-95 and its trafficking into dendrites, resulting in an overall decrease in the excitatory synapses being formed onto mutant cells. Our findings present a plausible mechanism for understanding how loss of zDHHC15 and reduced palmitoylation of its substrates may contribute to the pathology associated with intellectual disability.

As stated in the preface, Jordan Shimell and Bhavin Shah, who are co-first authors on the paper publishing these results, performed experiments presented in this chapter.

Bhavin Shah performed the developmental timecourse Western blots, localization immunochemistry, biotinylation assays, dendritic imaging, spine analysis, and synaptic imaging.

Jordan Shimell performed colocalization experiments with PSD-95, gephyrin, and giantin, Western blots for protein expression, live imaging of dendritic growth dynamics,

palmitoylation assays, and fluorescence recovery after photobleaching (FRAP) experiments.

Jordan Shimell and Bhavin Shah performed all analysis, data curation, figure creation, and manuscript preparation equally.

2.1 Introduction

Post-translational modification of cellular proteins by S-acylation involves the reversible attachment of fatty acids to cysteine residues, and is important for the trafficking of proteins towards cell membranes and the regulation of cellular signaling (Fukata and Fukata, 2010; Ko and Dixon, 2018). Palmitoylation is the reversible attachment of the fatty acid palmitic acid and is the most common post-translational lipid modification in the brain (Fukata and Fukata, 2010).

Enzymes that mediate palmitoylation consist of a family of 24 proteins containing a conserved

Asp-His-His-Cys (DHHC) motif that is required for enzymatic activity (Fukata and Fukata,

2010). Nine of the 24 genes encoding zinc finger DHHC-type (zDHHC) enzymes have been associated with brain disorders and ~41% of all identified synaptic proteins are substrates for palmitoylation (Sanders et al., 2015), underscoring the importance of palmitoylation in the development and function of the brain and, specifically, in synapse biology. Despite their implied importance, little is known about the role of zDHHC enzymes in the brain and whether disrupting their enzymatic activity contributes to brain pathology (Charollais and Van Der Goot,

2009; Fukata and Fukata, 2010; Rocks et al., 2010).

Although zDHHC15 is expressed in many cell types in the brain (Fukata et al., 2004;

Mansouri et al., 2005; Wang et al., 2015; Zhang et al., 2014), little is known about its function.

Known substrates of zDHHC15 include PSD-95 (officially known as DLG4), GAP43,

SNAP25b, CSP (officially known as DNAJC5), GABAγ2, Fyn, BACE1, CD151, CIMPR 62

(officially known as IGF24) and SORT1 (Fang et al., 2006; Fukata et al., 2004; Greaves et al.,

2010, 2008; Mill et al., 2009; Sharma et al., 2008; Tsutsumi et al., 2009; Vetrivel et al., 2009;

Yokoi et al., 2016), proteins that have been shown to play an important role in the development and function of neuronal connections. ZDHHC15 is one of a number of genes on the X chromosome that is duplicated in patients with intellectual disability (Linhares et al., 2016;

Martinez et al., 2014), and one report has identified the loss of ZDHHC15 transcripts in a female patient with non-syndromic X-linked intellectual disability (Mansouri et al., 2005).

Our study characterizes the role of zDHHC15 in the development of neuronal connectivity. zDHHC15 is highly expressed at embryonic stages and reduced in adult brains.

Knockdown of zDHHC15 in hippocampal cultures reduces dendritic arborization and spine maturation. Notably, zDHHC15 knockdown results in a marked reduction in the density of excitatory synapses, resulting from reduced palmitoylation of PSD-95 and its trafficking into dendrites.

2.2 Materials and Methods

2.2.1 Antibodies

Primary antibodies used were: anti-zDHHC15 (Abcam, Cambridge, MA, #ab121203,

1:250, Western blot), anti-zDHHC15 (Thermo Scientific, Waltham, MA, #PA39327, immunofluorescence, 1:100), anti-FLAG (Sigma Aldrich, St Louis, MO, #F3165, 1:1000), anti-

Myc (Origene, Rockville, MD, #TA150121, 1:1000), anti-β-actin (Novus Biologicals,

Centennial, CO, #NB600-503, 1:1000), anti-PSD-95 (Abcam, Cambridge, MA, #ab2723, 1:500), anti-gephyrin (Synaptic Systems, Göttingen, Germany #147011, 1:500), anti-GAD65 (Synpatic

Systems, Göttingen, Germany #198104, 1:500), anti-GFAP (Synaptic Systems, Göttingen,

Germany #173004, 1:400), anti-MAP2 (Millipore, Burlington, MA, MB3418, 1:2000), anti- 63

giantin (Synaptic Systems, Göttingen, Germany, #263004, 1:500), anti-GM130 (BD Biosciences,

Franklin Lakes, NJ, #610822, 1:200). Secondary antibodies from Life Technologies (Carlsbad,

CA,) used: (1:500), anti-mouse AF568 IgG2a (#A21134), anti-mouse AF647 IgG1 (#21240), anti-rabbit AF488 (#A11008), anti-rabbit AF568 (#A11011), anti-guinea pig AF633 (#A21105).

HRP conjugated secondary antibodies were obtained from Bio-Rad (Hercules, CA,): goat anti- mouse #170-6516 and goat anti-rabbit #170-6515, 1:300). DAPI was used for nuclear staining

(Life Technologies, Carlsbad, CA, #D1306, 1:1000).

2.2.2 Plasmids and Primers

The rat Myc-DDK tagged zDHHC15 ORF construct was obtained from Origene

(Rockville, MD, #RR212342). The shRNA target sequence that gave maximum knockdown of efficiency of zDHHC15 was 5’-GGTTCAATCTTGGCTTCATCA-3’ and was cloned into the pLL3.7 vector (a gift from Luk Parijs; Addgene plasmid #11795) using the following sense and anti-sense sequences based on previous cloning experiments in pLL3.7 using XhoI and HpaI sites (Rubinson et al., 2003): Sense: 5’-TGGTTCAATCTTGGCTTCATCATTCAAGAGATGA-

ATGAAGCCAAGATTGAACCTTTTTTC-3’ and anti-sense: 5’-TCGAGAAAAAAGGTTCA-

ATCTTGGCTTCATCATCTCTTGAATGATGAAGCCAAGATTGAACAA-3’. This plasmid is referred to as shRNA in all figures. To generate a scrambled shRNA construct to control for potential off-target effects of the shRNA, we used the following sense and anti-sense sequences:

Sense: 5’-TGAAGTTCGTACTTATCTCCGTTTCAAGAGAACGGAGATAAGTACGAACT-

TCTTTTTTC-3’ and anti-sense: 5’-TCGAGAAAAAAGAAGTTCGTACTTATCTCCGTTCT-

CTTGAAACGGAGATAAGTACGAACTTCA-3’. To generate an shRNA resistant form of zDHHC15 (zDHHC15R), site-directed mutagenesis (Agilent Technologies, Santa Clara, CA:

QuikChange II XL Site-Directed Mutagenesis Kit, 200-521) of the zDHHC15 ORF sequence 64

GGTTCAATCTTGGCTTCATCA into GGTTTAATCTAGGATTCATAA was carried out by using the following forward and reverse primers: Forward 5′- CAGAGAAAAATGGGTTTAAT-

CTAGGATTCATAAAGAATATTCAG-3′ and Reverse 5′-CTGAATATTCTTTATGAATCCT-

AGATTAAACCCATTTTTCTCTG-3′ (where the underlined residues indicate the same amino acid codons with mutated nucleotide bases). zDHHS15R was made by using the zDHHC15R template with the following forward and reverse primers: Forward 5’-AAAAATGGACCATCA-

TTCCCCATGGGTTAAT-3′ and Reverse 5′-ATTAACCCATGGGGAATGATGGTCCATTT-

3′.

2.2.3 Animals

All experimental procedures and housing conditions were approved by the UBC Animal

Care Committee and were in accordance with the Canadian Council on Animal Care (CCAC) guidelines.

2.2.4 Primary Culture from Sprague-Dawley Rats

Hippocampi from embryonic day 18 (E18) Sprague-Dawley rats (Rattus norvegicus) of either sex were prepared and plated on 18 mm Marienfeld (Lauda-Königshofen, Germany) coverslips for 12-well plates or directly onto the plastic of 6-well plates at a density of 130 cells/mm2. Coverslips had been previously treated with concentrated nitric acid (Sigma Aldrich,

St Louis, MO) overnight, rinsed several times with milliQ water, then sterilized with high heat

(>250°C overnight). Coverslips were then placed in the 12-well plate, UV sterilized for 30 min, and coated with 0.5 mg/mL poly-L-lysine hydrobromide mixed with borate buffer. The plates were covered with aluminum foil and left in a biosafety cabinet overnight. The following day, coverslips were rinsed 3x with autoclaved milliQ water and plating medium (84.75 mL MEM with Earle’s BSS, 10 mL FBS, 2.25 mL of 20% glucose, 1 mL sodium pyruvate, 1 mL 100X 65

GlutaMax, 1 mL 100X Pen/Strep) was added (1 mL per 12 well, 2 mL per 6 well). Plates were then placed in the 5% CO2 incubator at 37°C overnight.

For dissection, timed-pregnant Sprague-Dawley rats (Rattus norvegicus, Jackson

Laboratories, Sacramento, CA) were euthanized using CO2 and cervical dislocation. Pups were harvested into 150-mm Petri dishes and decapitated, and the brains were then dissected out and placed in pre-chilled HBSS on ice. Hippocampi were dissected out and placed in a 15 mL conical tube with 10 mL of fresh pre-warmed (37°C) HBSS. Once hippocampi settled, supernatant was removed and replaced with 10 mL of fresh pre-warmed HBSS for a total of three rinses. After the final rinse, 4.5 mL of fresh pre-warmed HBSS was added to the hippocampi with 0.5 mL of 2.5% trypsin and incubated in a 37°C water bath for 20 min with gentle agitation every 5 min. In the final 3 min, 1% DNAse was added and gently agitated to ensure mixture. Hippocampi were then rinsed 3x with fresh, pre-warmed HBSS and, after the final wash, ~1 mL of HBSS was used to transfer hippocampi to a 60-mm dish for trituration. The total volume following trituration was brought up to 5 mL and cell density was determined using a hemocytometer to plate the desired density. At 3-4 h after plating, plating medium was aspirated and replaced with pre-warmed (37°C) maintenance media (97 mL Neurobasal medium,

2 mL 50X B-27 supplement, 1 mL 100X Glutamax, 1 mL 100X Pen/Strep). Maintenance medium was renewed with fresh maintenance medium 2-3 days following plating.

2.2.5 Transfections (Primary Hippocampal Cultures/HEK293T Cells)

Primary hippocampal cultures were transfected at 9-11 DIV by using Lipofectamine 2000

(Invitrogen, Carlsbad, CA) according to the manufacturer’s recommendations. Briefly, for 12- well-plates, two aliquots of 25 µl Opti-Mem (Gibco, Thermo Fisher Scientific, Waltham, MA) were prepared. To one aliquot, 1-3 µg of total plasmid DNA was added. To the other aliquot, 1 66

µl of Lipofectamine 2000 was added and allowed to mix for 5 min. After this period, aliquots were combined and incubated for 20 min, then added drop by drop to the 12-well plates. Cells were then live imaged (10-13 DIV) or fixed (13 DIV) for subsequent experiments. HEK293T cells were transfected with Lipofectamine 2000 at 70% confluency at a ratio of 3:1 (shRNA or scrambled shRNA to target plasmid) and incubated for 48-72 h before harvesting for biochemical analysis. For HEK293T cells, 150 µ of Opti-Mem were used with 6 µl of

Lipofectamine 2000.

Nucelofections for 6-well plate biochemical assays were performed immediately before plating at 0 DIV by using an Amaxa Nucleofection Kit (Lonza, Basel, Switzerland, VPG-1003) according to the manufacturers optimized protocol (Number 101, program G-13), and were used for experiments at either 4 DIV (Figure 2A) or 7 DIV (Figure 4I). Briefly, hippocampi were dissociated as above and ~500,000 neurons isolated for nucleofection. Neurons were mixed with

100 µl Ingenio electroporation solution (Mirus Bio, Madison, WI) and cDNA constructs (~3 µg in total) were added to the solution. The solution was transferred to an Amaxa Nucleofection cuvette and inserted into the Amaxa Nucleofector with program G-013. Following electroporation, the solution was plated into a 6-well dish for future biochemical experiments.

2.2.6 Immunocytochemistry

Immunocytochemistry experiments were performed as previously reported (Sun and

Bamji, 2011). Briefly, cells were fixed for 10 min in a pre-warmed (37°C) 4% paraformaldehyde-sucrose solution. Cells were then washed in phosphate buffered saline (PBS) and treated with 0.1% Triton X-100 in PBS for 10 min at room temperature, washed in PBS, and blocked for 1 h at room temperature with 10% goat serum in PBS. Cells were then incubated in

1% goat serum in PBS with primary antibodies overnight at 4°C. Subsequently, cells were 67

washed 3X in PBS for 10 min each, incubated in secondary antibodies in 1% goat serum in PBS for 1 h at room temperature, washed again 3X in PBS and mounted on microscope slides by using Prolong Gold (Molecular Probes, Thermo Fisher Scientific, Waltham, MA). For DAPI staining, a 1:100 dilution from 5 mg/ml stock was applied after all other staining procedures for

5 min. Cells were then washed 3X in PBS and then mounted as above.

2.2.7 Biotinylation

For surface biotinylation assays, neurons in 6-well plates were washed with ice-cold PBS

(pH 8) supplemented with 0.1 mM CaCl2 and 1 mM MgCl2 (PBS-CM) and then incubated for 30 min with 0.5 mg/ml NHS-SS-Biotin (Thermo Fisher Scientific, Waltham, MA) in ice-cold PBS-

CM at 4°C with gentle rocking. After incubation, cells were washed once with PBS-CM; the unbound biotin was then quenched during two 8 min incubations with quenching buffer (20 mM glycine in PBS-CM). Lysis was performed by mechanical scraping in lysis buffer (1% IGEPAL-

CA630 and 1 mM PMSF with Roche complete protease inhibitor tablet) and followed by centrifugation at 500 g for 5 min at 4°C. Samples were then vortexed, run 3-4 times through a

26½ gauge syringe, and nutated at 4°C (VWR Nutating Mixer, 20° fixed tilt angle, mixing speed

24 rpm) for 30 minutes. After nutation, samples were centrifuged at 16,100 g for 30 min at 4°C to clear the lysate. Protein content in the cell lysate was then quantified using a BCA Assay Kit

(Thermo Fisher Scientific, Waltham, MA) as per the manufacturer’s instructions. Whole-cell lysate (10 µg) was then combined with SDS-sample buffer (50 mM Tris-HCl, 2% SDS, 10% glycerol, 14.5 mM EDTA and 0.02% Bromophenol Blue with 1% β-mercaptoethanol), boiled for

5 min at 95°C and stored at -20°C as the input sample. Of the remaining protein sample, 100-200

µg was added to 50 µl of a 50% slurry of NeurtrAvidin-conjugated agarose bears (Thermo Fisher

Scientific, Waltham, MA) that had been pre-washed three times in lysis buffer. Each sample was 68

then brought to a total volume of 500 µl with lysis buffer and nutated (VWR Nutating Mixer, 20° fixed tilt angle, 24 rpm) overnight at 4°C. The following day, beads were pelleted and washed seven times by centrifugation at 500 g for 3 min, each time discarding the supernatant. Beads were eluted using 40 µl of SDS-sample buffer supplemented with 100 mM DTT and samples were boiled at 90°C for 5 min followed by Western blotting with whole-cell lysates.

2.2.8 Acyl-RAC (palmitoylation) assay

For Acyl-RAC assays, we followed the manufacturer’s protocol from the CAPTUREome

S-palmitoylated protein kit (Badrilla, Leeds, UK), with the following minor changes. We optimized the protocol for our experiments by adding DNAse to the lysed-cell solution.

Additionally, we measured protein concentration (BCA Assay, ThermoFisher Scientific,

Waltham, MA) after dissolving the precipitated protein and prior to separating the lysate into experimental (Thioester Cleavage Reagent) and negative control (Acyl Preservation Reagent) samples to ensure accurate loading of equal protein concentrations.

2.2.9 Western blot analysis

Western blotting was performed as previously described (Sun and Bamji, 2011). Primary hippocampal neurons from Sprague Dawley rats, HEK293T cells, or brain regions from mouse

(i.e. cortex or cerebellum) were lysed at the time points indicated in figures (mouse brain regions

– E17, P10, P90; Hippocampal neurons – 13 DIV; Hippocampal neurons – 4 DIV; Hippocampal neurons, 7 DIV; HEK293T cells, 95-100% confluency) in ice-cold RIPA buffer of lysis buffer

(1% IGEPAL CA-630, 50 mM Tris-HCl-pH 7.5, 150 mM NaCl, 10% glycerol) supplemented with protease inhibitory cocktail solution at 4°C on a nutator (VWR Nutating Mixer, VWR,

Radnor, PA) for 2 h at 20° fixed tilt and 24 rpm. Lysates were cleared by benchtop centrifugation at 16,000 g for 30 min at 4°C and the supernatant was mixed with 5X SDS-dye for SDS-PAGE. 69

Proteins were separated on either a 10% resolving gel or a 4-20% gradient gel, transferred onto a

PVDF membrane, blocked in 3% BSA (in TBST) and were then immunoblotted for zDHHC15 or Myc prepared in 0.3% BSA in TBST. Blots were then visualized using enhanced chemiluminescence reagent (Merck Millipore, #WBKLS0500) on a Li-Cor C-Digit Blot Scanner

(LI-COR, Lincoln, NE). Blots were quantified using ImageJ software (NIH, Bethesda,

Maryland).

2.2.10 Imaging

Fixed and live neurons were imaged by using an inverted Olympus (Richmond Hill, ONT,

Canada) Fluoview 1000 (FV1000) confocal microscope. Imaging of synapses and colocalization studies utilized the 60X/1.42 Oil Plan-Apochromat objective while dendritic length measurements utilized the 20X/0.75 Oil Plan-Apochromat objective. MatTek glass-bottomed culture dishes (P35G-1.5-14-C, MatTek Corporation, Ashland, MA) were used to perform live imaging of dendrite growth of neurons in a chamber maintained at 37°C. eGFP-transfected cells with different experimental conditions were imaged in HEPES buffer (140 mM NaCl, 1.3 mM

CaCl2, 1.3 mM MgCl2, 2.4 mM K2HPO4, 25 mM HEPES pH 7.4) every 24 h post transfection up to 72 h to ensure cell viability. Levels and contrast of confocal images were moderately adjusted in Photoshop CS6 software (Adobe Systems, San Jose, CA).

2.2.11 Fluorescence Recovery after Photobleaching (FRAP)

Fluorescence recovery after photobleaching (FRAP) experiments were performed as previously described (Brigidi et al., 2014, 2015) but updated for use on an Olympus (Richmond

Hill, ONT, Canada) FV3000 microscope. Briefly, cells transfected with eGFP and PSD-95 were identified and up to four dendritic branches of the cell that were clear of other obstructions and relatively unbranched were photobleached from the soma to the end of the dendrite. Cell 70

positions were recorded using the motorized stage and ‘positions’ function of the FV3000, and re-imaged 9 h after bleaching. Fluorescence recovery was quantified using processing software within the Olympus FV3000 and then analyzed in Prism Software (Graphpad, La Jolla, CA).

2.2.12 Image Analysis and Quantification

2.2.12.1 Dendrite Length and Quantification

Neurons at 10-11 DIV were transfected and fixed at 13-14 DIV. To measure dendrite lengths, neurons were identified through eGFP fluorescence and imaged at a magnification of

20X. Images were processed using Adobe Photoshop to remove axons. Images were converted to

8-bit and then exported into the ImageJ tool ‘NeuronJ’ (Meijering et al., 2004) to trace the dendritic length. Sholl analysis (dendritic complexity analysis) was performed on the same 20X images that were first thresholded and then processed using the ‘Sholl Analysis’ plugin from

ImageJ (Ferreira et al., 2014) with settings to create a step size of 2 µm and an ending radius entered as ‘NaN’ (Not a Number) to enable sampling of the entire image to ensure the entire dendritic tree was captured. Data were imported into Microsoft Excel to create an average Sholl profile for each condition. For all images, the scale for ImageJ was set using the raw data from the Olympus FV1000/FV3000 dimensions to determine µm per pixel.

2.2.12.2 Spine Analysis

Dendritic spines were classified by using the default values and analysis from

NeuronStudio (version 0.9.92) adapted from Rodriguez et al. (2006), which uses a classification scheme based on the head to neck diameter ratio, length to head diameter ratio, and the head diameter. For data analysis, stubby and mushroom-type spines were classified as mature spines, whereas filopodial and/or thin protrusions were considered as immature spines.

2.2.12.3 Synapse Density, Puncta Size Analysis, and Colocalization

Transfected neurons were immunostained for PSD-95 (as a marker of excitatory synapses) and gephyrin (as a marker for inhibitory synapses). Synapses within eGFP transfected neurons were identified by making a neuronal ‘mask’ created from the eGFP-cell fill imaged at

60X magnification in Adobe Photoshop CS6 or CC2017 (Adobe Systems, San Jose, CA). The mask was devoid of cell soma and axons. We used the magic wand tool (non-contiguous, tolerance 30) to select a black background and this cell-fill GFP mask was applied onto PSD-95- positive and/or gephyrin-positive synapse marker channels to exclude synapses from other untransfected neurons and only examine synapses that had formed on mutant cells. These masks from synapse channels were then manually thresholded to make binary images and the total number of synaptic puncta was quantified using a macro in order to define synaptic puncta as having a size between 0.05 µm2 and 3 µm2. To obtain synapse density, the total number of puncta was divided by the total length of the masked dendrite. For the puncta size measurements, masked images were subjectively thresholded and PSD-95-positive puncta were measured for

Feret’s diameter, and then averaged for each image under each condition. Masked images were the thresholded and the ‘Colocalization’ plugin

(https://imagej.nih.gov/ij/plugins/colocalization.html) was used in a custom macro when applying the ‘Analyze Particles’ function of ImageJ (NIH, Bethesda, MA) to determine colocalized points from separate channels.

2.2.13 Statistical Analysis

For all experiments, data represent the mean±s.e.m. Statistical significance was measured using ANOVA with Tukey’s multiple comparison analysis in Graphpad Prism v6.01 (Graphpad,

La Jolla, CA). For all imaging experiments, the n value is the number of cells used per condition, 72

for at least three independent cultures. Statistical significance is assumed when α<0.05, unless mentioned otherwise (Figs. 3E,F and 4F,G, where the α-value is made more stringent to 0.001).

Significance is indicated when */#P<0.05, **/##P<0.01, or ***/###P<0.001 using one-way

ANOVA, Tukey’s multiple comparison test with mean±s.e.m. and determined by Prism software. All figures were generated by using Adobe Illustrator CS6 or CC2018 software in conjunction with Adobe Photoshop CS6 or CC2017 (Adobe Systems, San Jose, CA). Statistical significances (*) indicate differences compared with the control; significance within different rescue constructs is indicated by #.

2.3 Results and Discussion

2.3.1 zDHHC15 is Expressed During Early Stages of Brain Development, and in

Cultured Excitatory and Inhibitory Hippocampal Neurons

ZDHHC15 mRNA is abundantly expressed in the brain compared to other tissues (Fukata et al., 2004). After validating the specificity of our zDHHC antibody (Figure 2.1A-C), we determined that zDHHC15 is highly expressed during early stages of development – i.e. at embryonic day 17 (E17) and postnatal day 10 (P10) – and significantly reduced in the adult

(P90) brain (Figure 2.2A,B).

To further examine the cellular and subcellular distribution of zDHHC15, we immunostained primary hippocampal cultures at 13 days in vitro (DIV). zDHHC15 was observed in all neurons (identified by using anti-MAP2 antibody) – including GAD65-positive inhibitory neurons, demonstrating that zDHHC15 is expressed in both excitatory and inhibitory neurons. In contrast, in these young cultures, zDHHC15 was not expressed in GFAP-expressing glial cells (Figure 2.2C).

Figure 2.1 zDHHC15 Antibody Validation

(A) HEK293T cells were transfected with myc-tagged rat zDHHC15 (rzDHHC15), human zDHHC15

(hzDHHC15) or zDHHC2, the closest paralog of zDHHC15. Lysates were separated by SDS-Page and blots probed with the indicated antibodies, demonstrating the specificity of zDHHC15 antibody. Tagged- rzDHHC15 is calculated to run at 36-38kDa while -hzDHHC15 at 33-35kDa. (B) Representative confocal images of 13 DIV cultured hippocampal neurons transfected with control shRNA or zDHHC15 shRNA. Immunostaining with zDHHC15 antibody reveals decrease in immunofluorescence signal in neurons transfected with zDHHC15 shRNA, quantified in (C). Scale bar = 50µm. N = 8 neurons, 3 cultures.

zDHHC enzymes have been shown to localize to the plasma membrane and/or organelle membranes (Ohno et al., 2006). To determine whether zDHHC15 is localized to the plasma membrane, we performed a surface biotinylation assay using neurons at 13 DIV (Figure 2.2D).

Unlike zDHHC5 or zDHHC8, which have previously been shown to localize to plasma membranes (Brigidi et al. 2015; Ohno et al., 2006) zDHHC15 was not isolated in surface fractions, suggesting it is restricted to internal membranes. Coimmunostaining with zDHHC15 and the Golgi marker giantin demonstrated that zDHHC15 is highly expressed in the somatic

Golgi complex as well as in the dendrites – with 67±3% of all zDHHC15-positive puncta localizing with giantin (Figure 2.2E,F). The localization of zDHHC15 in the somatic Golgi of neurons is in accordance to a previous report demonstrating zDHHC15 localization to the Golgi of HEK293 cells (Greaves et al., 2008) and hippocampal neurons (Levy et al., 2011).

Figure 2.2 zDHHC15 is Expressed During Early Stages of Neocortical Development and in Golgi Compartments of Cultured Excitatory and Inhibitory Neurons

(A,B) Western blot and quantification of zDHHC15 protein levels in the hippocampus (Hc), cerebral cortex (C), hindbrain (Hb) or cerebellum (Cb) of mouse brain at E17, P10, and P90. Levels of zDHHC15 decreased from E17 to P90 in these regions of mouse brain. n=4 tissue samples per timepoint, *P<0.05, **P<0.001, one-way ANOVA with Tukey’s post-hoc test, means±s.e.m. (C) Representative confocal images of 13 DIV rat hippocampal cultures. All neurons, identified by using anti-MAP2 antibody (white arrows), expressed zDHHC15 (yellow arrows) – including inhibitory GAD65-positive neurons (white arrowheads). GFAP-positive glial cells (green arrows) exhibited weak to no zDHHC15 immunostaining. White arrows denote DAPI-stained nuclei that are positive for zDHHC15, whereas green arrows denote DAPI-stained nuclei that are negative for zDHHC15. (D) Western blot of lysates from biotinylated rat hippocampal neurons at 13 DIV probed for zDHHC5, zDHHC8, and zDHHC15; the latter is not localized at the membrane. n=3 cultures. (E) zDHHC15 colocalizes with cis-medial Golgi marker Giantin. (F-H) Microimages (F) showing 13 DIV rat hippocampal neurons stained for zDHHC15 and the synaptic markers PSD-95 and gephyrin. zDHHC15 colocalizes ith a small subset of PSD-95 or gephyrin puncta. zDHHC15 and PSD-95 images were masked as per Methods. n=10 neurons, three cultures. Scale bars: 100 µm (C), 20 µm (E, top) and 10 µm (E, bottom; F). 75

As zDHHC15 was distributed in a punctate manner within dendrites, we determined whether zDHHC15 localized to PSD-95 and gephyrin puncta that – as determined by us – are faithful excitatory and inhibitory markers, respectively (Figure 2.3A-D). The density of colocalized zDHHC15 and PSD-95, and zDHHC15 and gephyrin puncta, was quantified within eGFP ‘masks’ of neurons (Figure 2.3E). Coimmunostaining demonstrated minimal localization of zDHHC15 at synapses (Figure 2.2G). Only 18±2% of zDHHC15-positive puncta colocalized with PSD-95 and 10±2% colocalized with gephyrin (in turn 22±2% of PSD-95-positive and

23±3% of gephyrin-positive puncta colocalized with zDHHC15) (Figure 2.2G-H). Although the modest colocalization of PSD-95 and zDHHC15 may appear surprising – as PSD-95 is a known substrate of zDHHC15 (Fukata et al., 2004) – zDHHC15 function is thought to be dependent on synaptic activity (Noritake et al., 2009). It is, therefore, possible that the subcellular localization of zDHHC15 is altered following changes in synaptic activity. Together, these data suggest that zDHHC15 mediates its function during early stages of neuronal development, and that its primary site of action is at Golgi compartments that are distinct from Golgi satellites associated with excitatory or inhibitory synapses.

2.3.2 zDHHC15 Promotes Dendritic Outgrowth and Arborization

In humans, the ZDHHC15 gene is located on the X-chromosome and loss of ZDHHC15 transcripts has previously been shown to be associated with intellectual disabilities (Mansouri et al., 2005). Based on this and the high expression of zDHHC15 protein during development, we hypothesized that zDHHC15 is important for neuronal connectivity. To test this, we generated small hairpin RNA targeting zDHHC15 (shRNA; validated in Figure 2.4A) and examined whether zDHHC15 knockdown impacts neuronal morphology (Figure 2.4B-D shows processed images used for analysis as well as raw images, neuron masking, tracing, and Sholl analysis). 76

Figure 2.3 Validation of Synaptic Markers and zDHHC15 shRNA-mediated Changes in Excitatory Synaptic Density

(A-D) Confocal images of 14 DIV neurons fixed and immunostained with antibodies against PSD-95 and VGlut1 (A,B) or gephyrin and VGAT (C,D) demonstrate up to 75% colocalization indicating that gephyrin and PSD-95 are faithful markers of synapses at this point. The percentage of PSD-95 at excitatory syanpses was quantified by the number of PSD95/VGlut colocalized puncta divided by the total number of PSD-95 puncta (75±2%) or gephyrin/VGAT divided by total gephyrin (67±2%). (E) Illustration of masking procedure to identify and quantify immunostaining within eGFP transfected cells. Axons, background, and cell bodies were manually removed using Adobe Photoshop CS6 and the remaining eGFP masks overlaid on immunostained images (PSD-95 and gephyrin in this example. Raw images are shown on the left and immunostained proteins only within the eGFP mask are shown on the right. (F) Representative masked images of neurons expressing control or zDHHC15 shRNA demonstrate a reduction in excitatory synapse density (colocalization of PSD-95 and Vglut). Scale bars:20µm(A, C), 10 µm (E,F)

Neurons expressing shRNA targeting zDHHC15 exhibited significantly shorter dendrites

(Figure 2.4B,E), decreased branch count (Figure 2.4B,F), and less dendritic complexity (Figure

2.4B,G) compared to neurons transfected with a control scrambled shRNA. These dendritic

Distance from Soma (µm)

Figure 2.4 zDHHC15 Promotes Dendritic Growth and Arborization

(A) Western blot of hippocampal neurons nucleofected at 0 DIV, lysed at 4 DIV, and probed with the indicated constructs. n=3 cultures. (B) Rat hippocampal neurons were transfected with the indicated constructs, fixed, and imaged at 13 DIV. (C) Raw images of those used in (B). (D) An example neuron with an example of masking, tracing, and Sholl analysis. Scale bar: 100 µm. (E-G) Quantification of dendritic length and counts from 13 DIV rat hippocampal neurons. Knockdown of zDHHC15decreases dendritic length and complexity. Statistical significance was calculated for the entire Sholl profile. n=43 (control), 49 (shRNA), 34 (shRNA+zDHHC15), 40 (shRNA+zDHHS15), 36 (zDHHC15) neurons, three cultures. **P<0.01, ***P<0.001, one way ANOVA and Tukey’s post-hoc test; mean±s.e.m. 78

defects were rescued in neurons expressing an shRNA-resistant form of zDHHC15 (zDHHC15R;

Figure2.4B,E-G). To determine whether the palmitoylation function of zDHHC15 is important for dendritic outgrowth, neurons were co-transfected with shRNA and a palmitoylation-deficient form of zDHHC15, in which the catalytic cysteine residue of the DHHC motif had been changed to serine (zDHHS15R). Although expression of zDHHS15R was similar to that of zDHHC15R

(Figure 2.4A), it was unable to rescue the knockdown phenotype, indicating that the enzymatic activity of zDHHC15 is essential for zDHHC15-mediated regulation of dendritic arbor length and complexity (Figure 2.4E-G). Overexpression of zDHHC15 had no effects on dendritic length, dendrite number, or arbor complexity (Figure 2.4B, E-G).

Figure 2.5 zDHHC15 Knockdown Leads to Inhibition of Dendritic Outgrowth

(A) Representative, masked, time-lapse confocal images of neurons transfected at 10 DIV with eGFP plus the indicated constructs and imaged at 24, 48, and 72 h post transfection. Scale bar: 100 µm. (B) zDHHC15 knockdown inhibits dendrite growth. n=10 neurons, three cultures. 79

To further investigate whether zDHHC15 promotes dendritic outgrowth or stabilization, neurons were imaged every 24 h for 72 h post transfection using time-lapse imaging. At 72 h post transfection, neurons expressing zDHHC15 shRNA exhibited an increase of only 13±4.7% in total dendritic length, whereas the dendritic length of cells expressing control shRNA, shRNA+zDHHC15R, or zDHHC15 alone increased by 33±3.9%, 34±4.0%, or 39±1% respectively (Figure 2.5A,B). These data indicate that zDHHC15 is required for dendritic growth and not stability, as knockdown of zDHHC15 inhibited further dendritic outgrowth.

2.3.3 zDHHC15 Promotes Spine Maturation and the Formation of Excitatory Synapses

We next determined whether zDHHC15 regulates the formation and/or maturation of spiny protrusions – including filopodia – that exhibit dynamic motility and are associated with smaller, less mature synapses, as well as with stubby or mushroom spines, which are typically associated with large, mature synapses (Casanova et al., 2012; Marin-Padilla, 1972). Although knockdown of zDHHC15 did not impact the total number of spiny protrusions at 13-14 DIV

(Figure 2.6A), it resulted in a significant increase in the proportion of thin, immature protrusions, and a decrease in the proportion of mature, stubby, and mushroom spines (Figure 2.6B,C).

Although zDHHC15R rescued this phenotype, zDHHS15R did not, indicating that zDHHC15 regulates spine morphology and/or maturation through its palmitoylation function.

As the excitatory postsynaptic protein PSD-95 can be palmitoylated by zDHHC15 (as well as by zDHHC2, zDHHC3, and zDHHC7) (Fukata et al., 2004), and because knockdown of zDHHC15 decreases the proportion of mature spines (Figure 2.6C), we next investigated whether zDHHC15 regulates the formation and/or maintenance of synaptic connections. At 10

DIV, cultures were transfected with eGFP plus the indicated constructs, cells were fixed and immunostained for PSD-95 and gephyrin at 13 DIV and masks were generated as previously 80

Figure 2.6 zDHHC15 Knockdown Leads to Inhibition of Mature Spine Formation

(A) Hippocampal neurons were transfected at 10 DIV and fixed at 13 DIV to calculate the density (A) and type (B,C) of spiny protrusions. While there was no significant change in the density of overall protrusions (A), zDHHC15 knockdown significantly increased the proportion of filopodia (B) and reduced the proportion of mature, mushroom, and stubby spines (C). n=56 (control), 43 (shRNA), 38 (shRNA+zDHHC15), 33 (shRNA+zDHHS15), 31 (zDHHC15) neurons, three cultures. ***P<0.001, one-way ANOVA and Tukey’s post-hoc test (C-E), mean±s.e.m., ns-not significant. described (Brigidi et al., 2015). zDHHC15 knockdown led to a significant reduction in the density of puncta positive for PSD-95 but not gephyrin (Figure 2.7A,B), causing an overall decrease in the ratio of excitatory to inhibitory (E:I) synapses (Figure 2.7C). Of note, zDHHC15 knockdown also resulted in a decrease in the density of colocalized PSD-95 and VGlut1, demonstrating a reduction in excitatory synapse number and not just PSD-95 density (Figure

2.7D). While expression of wild-type zDHHC15R rescued the knockdown phenotype, zDHHS15R did not, indicating that the enzymatic activity of zDHHC15 is important for its regulation of excitatory synapse density (Figure 2.7A-C).

As PSD-95 is also palmitoylated by zDHHC2, zDHHC3, and zDHHC7 (Fukata et al.,

2004), we determined whether overexpression of these enzymes can rescue the zDHHC15 knockdown phenotype (we did not use zDHHC7 as its levels are low in the hippocampus

[https://www.proteinatlast.org/ENSG00000153786-ZDHHC7/tissue] and proved difficult to 81

express). The zDHHC15 shRNA-mediated decrease in PSD-95 density was rescued by zDHHC3 and zDHHC15, and to a lesser extent, zDHHC2 (Figure 2.7D). Moreover, while zDHHC15 knockdown did not affect PSD-95 puncta size, co-expression of zDHHC15 with zDHHC2 enhanced the area of PSD-95 clusters by approximately 10%, suggesting a different mechanism of action (Figure 2.7E). In accordance with zDHHC2 and zDHHC3’s ability to rescue PSD-95 density, expression of these proteins rescued spine maturation (Figure 2.7F,G). However, in contrast to the ability of zDHHC2 and zDHHC3 to rescue synapse density and spine maturation, neither zDHHC2 nor zDHHC3 rescued zDHHC15 knockdown-mediated changes regarding the dendritic length (Figure 2.7H). The results demonstrate that zDHHC15 knockdown decreases excitatory synapse density and outgrowth, and that the pathway regulating excitatory synapse density is independent of the pathway involved in dendritic length and complexity.

2.3.4 zDHHC15 Promotes the Trafficking of PSD-95 into Dendrites

As zDHHC15 is known to palmitoylate PSD-95, and because excitatory synapse density is reduced following zDHHC15 knockdown, we tested whether PSD-95 palmitoylation is decreased in zDHHC15 knockdown cells. Hippocampal neurons were nucleofected with zDHHC15 shRNA resulting in a 58±3.8% efficiency of transfection (investigated by counting the eGFP-positive transfected cells in each culture before lysis). By using the acyl-RAC assay, we demonstrated a 62.5%±7.5% reduction in PSD-95 palmitoylation in primary neuron cultures nucleofected with zDHHC15 shRNA compared to control shRNA, with no change in total PSD-

95 levels (Figure 2.7I,J).

Previous work has demonstrated that the N-terminal palmitoylation motif of PSD-95 is necessary for targeting of PSD-95 into dendrites (El-Husseini Ael et al., 2001), as well as its association with vesiculotubular structures that traffic PSD-95 to synapses (El-Husseini et al., 82

Figure 2.7 zDHHC15 Knockdown Decreases Excitatory Synapse Density in Hippocampal Neurons and 83

Disrupts PSD-95 Trafficking into Dendrites

(A) Representative confocal images of rat hippocampal neurons at 13 DIV expressing eGFP plus the indicated constructs. Control shRNA is a scrambled form of the shRNA targeting zDHHC15 (zDHHC15 shRNA, shortened to shRNA) to ensure there are no off-target effects of the shRNA. zDHHC15R is a form of zDHHC15 that has had mutations introduced to modify the sequence recognized by the targeting shRNA to make it shRNA resistant. zDHHS15R is a construct wherein the catalytic cysteine has been modified to serine to reduce palmitoylation activity and is also resistant to the targeting shRNA. Immunostaining was for the excitatory post-synaptic marker PSD-95, and the inhibitory post-synaptic marker gephyrin (masked as per Methods). Scale bar: 10 µm. (B) The density of PSD-95 positive puncta is significantly decreased in zDHHC15 knockdown neurons, while the density of gephyrin-positive puncta remains unchanged. (C) This alters the ratio of excitatory:inhibitory synapse density being formed onto zDHHC15 mutant neurons. n=51 (control), 43 (shRNA), 33 (shRNA+zDHHC15), 34 (shRNA+zDHHS15), 33 (zDHHC15) neurons, three cultures. (D) The zDHHC15 shRNA-mediated reduction in excitatory synapse density (density of colocalized PSD-95 and VGlut1 (an excitatory pre-synaptic marker) puncta) is rescued by coexpression of zDHHC15, zDHHC2 or zDHHC3. (E) The area of PSD-95 puncta is unchanged in cells expressing zDHHC15 shRNA but significantly increased compared to controls in cells expressing zDHHC15 shRNA plus zDHHC2. (F,G) zDHHC15, zDHHC2 and zDHHC3 rescue zDHCH15 shRNA mediated changes in the proportion of filopodia (F) and mature spines (G). (H) By contrast, zDHHC2 and zDHHC3 are unable to rescue zDHHC15-mediated decreases in dendritic length. n=25 neurons, three cultures. (I) Acyl-RAC assay from cultured neurons at 7 DIV that were nucleofected with control or zDHHC15 shRNA. Left panel (input levels): zDHHC15 was reduced by 66.25±6% in neurons that had been transfected with zDHHC15 shRNA, reflecting the efficiency of transfection. No changes in PSD-95 levels were observed. Right panel (Acyl-RAC fractions): Results from the same blot; PSD-95 palmitoylation is decreased in cultures transfected with zDHHC15 shRNA (cleaved bound fraction, cBF), low levels of non-palmitoylated PSD-95 (cUF, cleaved unbound fraction), and minimal non-specific binding to the resin (pBF, preserved bound fraction). The pUF (preserved unbound fraction) represents samples that have not been cleaved. (J) PSD-95 palmitoylation is decreased in zDHHC15 knockdown neurons. n=3 cultures. ***P<0.001, student t-test, mean±s.e.m. (K) Hippocampal neurons at 10 DIV were transfected with the indicated constructs plus PSD-95-RFP. Puncta of PSD-95-RFP within an entire primary dendrite branch (white outline) were photobleached at 13 DIV and fluorescence recovery was analysed 9 h after bleaching. Fluorescence recovery was decreased in cells expressing zDHHC15 shRNA (PSD-95 puncta pseudo-colored for fluorescence intensity). Magnified area of interest shown below. Scale bar: 10 μm. (L) Quantification of fluorescence recovery. n=10 neurons, three cultures. ***P<0.001, **/##P<0.01, *P<0.05, one-way ANOVA and Tukey’s post-hoc test, mean±s.e.m

2000). To determine whether zDHHC15 is specifically involved in the targeting and trafficking of PSD-95, hippocampal neurons were transfected with PSD-95-RFP plus the indicated constructs. All PSD-95 puncta within an entire dendrite were photobleached, and fluorescent recovery within the photobleached dendrite was measured 9 h after photobleaching (Figure

2.7K,L). While neurons transfected with control shRNA exhibited 73±7% recovery of PSD-95-

RFP fluorescence, neurons transfected with zDHHC15 shRNA exhibited only 35±6% recovery.

This knockdown phenotype was rescued by co-transfection with zDHHC15R (80±8% recovery)

(Figure 2.7K,L). In support of the fact that zDHHC15 regulates trafficking of PSD-95 into

dendrites, PSD-95-RFP levels were significantly higher in the soma of neurons expressing shRNA targeting zDHHC15 (Figure 2.7K).

Our data suggest that the decrease in excitatory synapse density within zDHHC15 knockdown neurons is due to decreased PSD-95 trafficking into dendrites (Figure 2.8). While the synapse density phenotype in knockdown cells could be rescued by zDHHC15, zDHHC3, and zDHHC2, all of which mediate PSD-95 palmitoylation (Fukata et al., 2004), rescue with either zDHHC3 or zDHHC15 was significantly more robust than the rescue with zDHHC2. As zDHHC3 and zDHHC15 are both localized to the somatic Golgi (Greaves et al., 2011; Noritake et al., 2009), it likely that these two enzymes regulate excitatory synapse density by palmitoylating PSD-95 and promoting the trafficking of PSD-95 into dendrites. In contrast, zDHHC2 is localized to dendritic compartments (Greaves et al., 2011; Noritake et al., 2009), and has been shown to mediate PSD-95 palmitoylation and to promote its clustering at synapses.

Together, our data demonstrates a role for zDHHC15 in the regulation of dendritic outgrowth, and the formation and maturation of glutamatergic synapses. This is of particular interest, given the association between zDHHC15 and X-linked intellectual disability (Mansouri et al., 2005;

Piton et al., 2013), and the fact that disruptions in dendrite outgrowth and synapse function are one of the most-consistent hallmarks of intellectual disability, observed in both postmortem brain sections and mouse models of intellectual disability (Casanova et al., 2012; Huttenlocher, 1974;

Jiang et al., 1998;Marin-Padilla, 1972; Purpura, 1974; Swann et al., 2000).

Figure 2.8 Model of zDHHC15 Function

In the wildtype neurons (top), zDHHC15 functions to palmitoylate PSD-95, facilitating its trafficking into dendrites where it can form stable PSD-95 puncta in dendrites and specifically at spines to stabilize and/or mature synapses to promote dendritic growth. However, in the zDHHC15 knockdown cells (bottom), PSD-95 palmitoylation is reduced, hindering its trafficking into dendrites and causing somatic accumulation, leading to the formation of more immature, filopodial-like spines and impairing further growth.

Chapter 3: The X-linked Disability Gene, zDHHC9, is Essential for Dendrite

Outgrowth and Inhibitory Synapse Formation

A number of genetic and biochemical studies strongly suggest a role of zDHHC9 in the etiology of X-linked intellectual disability (XLID) (Raymond et al., 2007; Masurel-Paulet et al.,

2014; Mitchell et al., 2014; Baker et al., 2015; Tzschach et al., 2015; Han et al., 2017). The most extensively studied mutations in zDHHC9 comprise two deletions in the protein coding region and two missense mutations (R148W and P150S) that result in a decrease in steady-state palmitoylation activity of zDHHC9 (Mitchell et al., 2014). These observations have led to the hypothesis that reduced palmitoylation of zDHHC9 substrate(s) could lead to alterations in neural function resulting in XLID. Recent work has also established that patients with loss-of- function mutations in zDHHC9 exhibit a higher than normal prevalence of epilepsy (Baker et al.,

2015). Here we report that zDHHC9 is essential for dendrite outgrowth and inhibitory synapse formation through the palmitoylation of two distinct substrates, Ras and TC10, respectively. The loss of zDHHC9 function in hippocampal cultures leads to shorter dendritic arbors and fewer inhibitory synapses, altering the ratio of excitatory-to-inhibitory inputs formed onto zDHHC9- deficient cells. While zDHHC9 promotes dendrite outgrowth through the palmitoylation of the

GTPase, Ras, it promotes inhibitory synapse formation through the palmitoylation of another

GTPase, TC10. zDHHC9 knockout mice exhibit seizure-like activity together with increased frequency and amplitude of both spontaneous and miniature excitatory and inhibitory postsynaptic currents, indicating an overall increase in network excitability. These findings present a plausible mechanism for how loss of zDHHC9 function may contribute to XLID and epilepsy.

As stated in the preface, the majority of work on this chapter was performed by

Jordan Shimell. Stefano Brigidi and Bhavin Shah provided help with HEK cell validation of zDHHC9 shRNA and TC10 palmitoylation, respectively. Blair Jovellar helped perform zDHHC9 localization experiments and quantification of synapse density for zDHHC9 knockdown and overexpression. Igor Tatarnikov, Naila Kuhlmann, and Austen

Milnerwood designed and performed the in vitro electrophysiolocial experiments and provided valuable input in writing and interpreting those results. Stuart Cain, Samrat

Thouta, Jennifer Kass, and Terrance Snutch designed and performed the in vivo electrophysiological experiments and provided valuable input writing and interpreting those results.

3.1 Introduction

Neurons exhibit distinct patterns of dendritic arborization that determine the number and type of synaptic connections and thereby the computational capabilities of the cell (Rall, 1995;

Hume & Purves, 1981; Purves & Hume, 1981; Purves et al., 1986; Hausser et al., 2000). The development of dendritic arbors and synaptic connections requires the correct targeting, trafficking, and function of neuronal proteins, and the post-translational lipidation of proteins is one process by which neurons generate and maintain protein distributions (Fukata & Fukata,

2010). The most common form of lipid modification in the brain is palmitoylation, the reversible addition of a palmitate fatty acid chain to cysteine residue(s) of substrate proteins. Palmitoylation is mediated by a family of palmitoyl acyl-transferases (PATs) consisting of 24 zDHHC enzymes that contain a zinc finger domain as well as a DHHC (aspartate-histidine-histidine-cysteine) motif (Fukata & Fukata, 2010). The core DHHC motif is highly conserved in eukaryotes

(Putilina et al., 1999; Mitchell et al., 2006; Hemsley, 2009) and is essential for catalytic function both in vitro and in vivo (Mitchell et al., 2006).

Previous work has demonstrated important roles for palmitoylation in the regulation of neurite outgrowth (Lievens et al., 2016; Ponimaskin et al., 2008), spine and synapse formation

(George et al., 2015; Kaur et al., 2016; Zhu et al., 2013), synaptic transmission (Ohyama et al.,

2007) and synaptic plasticity (Brigidi et al., 2015; Brigidi et al., 2014; Dejanovic et al., 2014;

Fukata et al., 2013; Thomas et al., 2012; Woolfrey et al., 2015). Moreover, proteomic analysis has demonstrated that 41% of all synaptic proteins are substrates for palmitoylation (Saunders et al., 2015), including SNARE proteins (Prescott et al., 2009; Veit et al., 1996; Veit et al., 2000;

Greaves & Chamberlain, 2011; Greaves & Chamberlain, 2006), GPCRs (Goddard & Watts,

2012; Qanbar & Bouvier, 2003; O’Dowd et al., 1989), AMPA and NMDAR subunits (Kang et al., 2008; Hayashi et al., 2005; Fang et al., 2006; Keller et al., 2004; Hayashi et al., 2009), ion channels and their components (Shipston, 2013; Schmidt & Catterall, 1987; Jeffries et al., 2010;

Tian et al., 2008; Bosmans et al., 2011; ), and scaffold proteins (el-Husseni et al., 2002; Topinka

& Bredt, 1998; Noritake et al., 2009; Dejanovic et al., 2014). )

Loss-of-function mutations in zDHHC enzymes and altered palmitoylation of substrates have been identified in patients with neurodegenerative and neurodevelopmental disorders

(reviewed in Cho & Park, 2016), with loss-of-function variants in zDHHC9 specifically identified in patients with X-linked intellectual disability (XLID) (Raymond et al., 2007;

Masurel-Paulet et al., 2014; Mitchell et al., 2014; Baker et al., 2015; Tzschach et al., 2015; Han et al., 2017). Indeed, ~2% of XLID patients have mutations in zDHHC9 (Raymond et al., 2007;

Tzschach et al., 2015) and a case of sporadic XLID with a de novo zDHHC9 variant was recently

reported (Han et al., 2017). Patients with XLID plus zDHHC9 mutations also have an elevated incidence of epilepsy compared to XLID patients with intact zDHHC9 (Baker et al., 2015).

Despite clinical data associating zDHHC9 with XLID and epilepsy, the roles of zDHHC9 in brain development and pathophysiology remain poorly understood. Here we demonstrate that zDHHC9 plays an important role in promoting dendritic outgrowth and the formation of inhibitory synapses. These effects are mediated through the direct palmitoylation of two distinct substrates, the small GTPases N-Ras and TC10. Loss of zDHHC9 function in primary hippocampal cultures results in shorter and less complex dendritic arbors and an increase in the ratio of excitatory-to-inhibitory (E:I) synapses formed onto zDHHC9 mutant cells. This is further supported in vivo wherein zDHHC9 knockout mice exhibit spontaneous high frequency spiking activity indicative of non-convulsive seizures and in vitro where CA1 hippocampal neurons in acute brain slices exhibit increased frequency and amplitude of both spontaneous and miniature excitatory and inhibitory postsynaptic currents (sEPSCs/sIPSCs and mEPSCs/mIPSCs).

Together, our findings identify an additional substrate (TC10) for zDHHC9 palmitoylation, demonstrate that zDHHC9 is a critical component of neural circuit formation by regulating dendritic outgrowth and E:I synaptic balance, and suggest that zDHHC9 loss-of-function mutations are causative for the epileptic comorbidities observed in XLID patients.

3.2 Methods and Materials

3.2.1 Key Resources Table

REAGENT or RESOURCE SOURCE IDENTIFIER Antibodies

Rabbit polyclonal anti-zDHHC9 Thermo Fisher Cat#PA5-26721;

Scientific RRID:AB_2544221;

LOT#RI2260461

Mouse monoclonal (IgG2a) anti-PSD95 [6G6-1C9] Abcam Cat#ab2723;

RRID:AB_303248;

LOT#GR299294-3

Guinea pig polyclonal anti-PSD95 Synaptic Systems Cat#124 014;

RRID:AB_2619800

Mouse monoclonal (IgG1) anti-Gephyrin Synaptic Systems Cat#147 011;

RRID:AB_887717

Guinea pig polyclonal anti-VGLUT1 Millipore Cat#AB5905;

RRID:AB_2301751

Guinea pig polyclonal anti-VGAT Synaptic Systems Cat#131 004;

RRID:AB_887873

Guinea pig polyclonal Anti-GAD65 Synaptic Systems Cat#198 104;

RRID:AB_10557995

Mouse monoclonal (IgG1) anti-NeuN [A60] Millipore Cat#MAB377;

(Chemicon) RRID:AB_2298772

Mouse monoclonal (IgG1) anti-Myc [9E10] Sigma-Aldrich Cat#M4439;

RRID:AB_439694

Rabbit monoclonal (Ig) anti-HA [C29F4] Cell Signaling Cat#5017S;

Technology RRID:AB_106933385

Rabbit polyclonal anti-FLAG Sigma-Aldrich Cat#F7425;

RRID:AB_439687

Mouse monoclonal (IgG1) anti-Giantin [9B6] Abcam Cat#ab37266;

RRID:AB_880195

Guinea pig polyclonal anti-Giantin Synaptic Systems Cat# 263 005;

RRID:AB_2619984

Mouse monoclonal (IgG1) anti-NRas [F155] Santa Cruz Cat#sc-31;

RRID:AB_628041

Mouse monoclonal (IgG1) anti-HRas [F235] Santa Cruz Cat#sc-29;

RRID:AB_627750

Rabbit monoclonal (IgG) anti-TC10 [Y304] Abcam Cat#ab32079;

RRID:AB_2179515

Rabbit polyclonal anti-TC10 Thermo Fisher Cat#PA1-1061;

Scientific RRID:AB_2179424

Mouse polyclonal anti-TC10 Abcam Cat#ab168645

Mouse monoclonal (IgG1) anti-βactin Sigma-Aldrich Cat#A1978;

RRID:AB_476692

Mouse monoclonal (IgG2b) anti-GAPDH [mAbcam 9484] Abcam Cat#ab9484;

RRID:AB_307274

Mouse monoclonal (IgG1K) anti-GFP [7.1, 13.1] Sigma-Aldrich Cat#11814460001;

(Roche) RRID_AB390913

Mouse monoclonal (IgG1) anti-p44/42 MAPK [L34F12] Cell Signaling Cat#4696;

Technology RRID:AB_390780

Rabbit monoclonal (IgG) anti-phospho44/42 MAPK Cell Signaling Cat#4370;

Technology RRID:AB_2315112

Rabbit polyclonal anti-SAPK/JNK Cell Signaling Cat#9252;

Technology RRID:AB_2250373

Mouse monoclonal (IgG1) anti-phosphoSAPK/JNK Cell Signaling Cat#9255;

Technology RRID:AB_2307321

Goat anti-rabbit Alexa Fluor 405/488/568/633 Thermo Fisher Cat#A-31556 / A-

Scientific 11008 / A-11011 / A-

21070;

RRID:AB_221605 /

AB_143165 /

AB_143157 /

AB_2535731

Goat anti-mouse Alexa Fluor 405/488/568/633 Thermo Fisher Cat#A-31553 / A-

Scientific 11001 / A-11019 / A-

21050; RRID:

AB_221604 /

AB_2534069 /

AB_143162 /

AB_141431

Goat anti-guinea pig Alexa Fluor 488/568/633 Thermo Fisher Cat#A-11073 / A-

Scientific 11075 / A-21105;

RRID:AB_142018 /

AB_141954 / 2535757

Goat anti-mouse IgG1 Alexa Fluor 568/647 Thermo Fisher Cat#A-21124 / A-

Scientific 21240;

RRID:AB_2535766 /

AB_2535809

Goat anti-mouse IgG2a Alexa Fluor 568/647 Thermo Fisher Cat#A-21134 / A-

Scientific 21241;

RRID:AB_2535773 /

AB_2535810

Goat anti-mouse IgG-HRP BioRad Cat#170-6516;

AB_11125547

Goat anti-rabbit IgG-HRP BioRad Cat#170-6515;

RRID:AB_11125142

Streptavidin HRP Thermo Fisher Cat#21126;

Scientific LOT#SD250815

Chemicals, Peptides, and Recombinant Proteins

Lipofectamine 2000 Invitrogen Cat#11668019

EZ-Link NHS-SS-Biotin Thermo Fisher Cat#21331

Scientific cOmplete mini, EDTA-free protease inhibitor cocktail Sigma-Aldrich Cat#11836170001; tablets LOT#30496700

UltraPure Glycine Invitrogen, Thermo Cat#15527-013

Fisher Scientific

Pierce High Capacity Neutravidin Agarose Thermo Fisher Cat#29204

Scientific

DNAse I Sigma-Aldrich Cat#DN25-100mg

GlutaMax Invitrogen, Thermo Cat#35050-161

Fisher Scientific

Pen/Strep Invitrogen, Thermo Cat#15140-148

Fisher Scientific

Trypsin Worthington Cat#LS003667

Prolong Gold Antifade Mountant Invitrogen, Thermo Cat#P36930

Fisher Scientific

N-ethylmaleimide Sigma-Aldrich Cat#E3876

EZ-link HPDP-Biotin Thermo Fisher Cat#21341

Scientific

Hydroxylamine Sigma-Aldrich Cat#255580

Dexmedetomidine hydrochloride Tocris Cat#2749

Fentanyl CDMV Cat#108594

Midazolam CDMV Cat#104388

Critical Commercial Assays

QuikChange II XL Site-Directed Mutagenesis Kit Agilent Technologies Cat#200522

CAPTUREome S-Palmitoylation Assay (Acyl-RAC) Badrilla Ltd Cat#K010-311

AMAXA Rat Neuron Nucleofector Kit (25 RCT) Lonza Cat#VPG-1003

Pierce BCA Assay Kit Thermo Fisher Cat#23227

Scientific

Pierce ECL Western Blotting Substrate Thermo Fisher Cat#32106

Scientific

Automated Genotyping Services Transnetyx, Inc www.transnetyx.com

Experimental Models: Cell Lines

Human: HEK293T Cells Laboratory of Fabio N/A

Rossi, UBC

Rat: E18 Primary Hippocampal Cultures Sprague Dawley N/A

Timed Pregnant

Rats, Jackson Labs

Experimental Models: Organisms/Strains

M.musculus (mouse), zDHHC9 knockout. Homologous MMRRC, University RRID:MMRRC_03271 recombination of coding exon 1 (NCBI ascension of California, Davis 4-UCD

NM_172465.1)

Oligonucleotides

Primers for zDHHC9 shRNA resistance construct This paper

Forward:GTGTTGGAAAGAGGAATTATCGATACTTCT

ACCTCTTCATCCT

Reverse:AGGATGAAGAGGTAGAAGTATCGATAATTC

CTCTTTCCAACAC

Primers for zDHHC9 R148W mutation This paper

Forward:GCGCGCGGCGGCCAAAAATTTTGCAGGTA

TAGCA

Reverse:TGCTATACCTGCAAAATTTTTTGGCCGCCG

CGCGC

Primers for zDHHC9 P150S mutation This paper

Forward:GGCTCGCGCGCGACGGGCGAAAAAT

Reverse:ATTTTTCGCCCGTCGCGCGCGAGCC

Primers for zDHHS9 shRNA resistance construct This paper

Forward:GAGCGCTTCGACCATCACTCCCCCTGG

Reverse:CAATTCCCCACCCAGGGGGAGTGATGGT

Primers for RasC181S construct This paper

Forward:GATGGGACTCAGGGTTCTATGGGATTGC

Reverse:CATGGCAATCCCATAGAACCCTGAGTCC

Primers for TC10C206,209S construct This paper

Forward:GAATAGGATCAAGAAGTATAAACAGTTGTTT

AATTACG

Reverse:CGTAATTAAACAACTGTTTATACTTCTTGAT

CCTATTC

Recombinant DNA

Plasmid: zDHHC9-myc Alaa El-Husseini, N/A

University of British

Columbia

Plasmid: N-Ras-GFP Liz Conibear, N/A

University of British

Columbia

Plasmid: GCP16-FLAG Maurine Linder, N/A

Washington

University School of

Medicine

Plasmid: zDHHC9 shRNA Santa Cruz SC-90991-SH

Biotechnology

Plasmid: zDHHC9 shRNA resistant construct This paper N/A

(zDHHC9R), zDHHS9 shRNA resistant construct

(zDHHS9R), zDHHC9 R148W shRNA resistant construct

(R148WR), zDHHC9 P150S shRNA resistant construct

(P150SR)

Plasmid: Pooled N-Ras / H-Ras shRNA OriGene TR712458A/C,

Technologies TR704050A/C

Plasmid: pCGN-N-Ras; NRasC181S Fiordalisi et al., Addgene Plasmid

2001; This paper #14723

Plasmid: GFP-TC10 Roberts et al., 2008 Addgene Plasmid

#23232

Plasmid: TC10 University of TC10000000

Missouri-Rolla cDNA

Resource Center

Plasmid: TC10 shRNA; TC10C203,206S Santa Cruz SC-41894-SH

Biotechnology; This

paper

Software and Algorithms

ImageJ National Institute of https://imagej.nih.gov/i

Health, Bethesda j/

Photoshop CS6 Adobe Systems, Inc

Olympus Fluoview FV10 ASW 4.2 Olympus

Corporation

Prism 6.01 Graphpad Software,

Inc

ImageJ “Sholl Analysis” Plugin Ferreria et al., 2014 http://fiji.sc/Sholl

ImageJ “NeuronJ” Plugin Meijering et al., 2004 http://www.imagescien

ce.org/meijering/softw

are/neuronj

Custom MatLab Script (EEG Analysis) LeDue, Bohnet, & https://ninc.centreforbrainhe

alth.ca/sites/default/files/eeg Cain .zip

Other

3.2.2 Experimental Model and Subject Details

3.2.2.1 Cell Lines (HEK293T cells)

HEK293T cells were aliquoted into a 10 cm dish with 15 mL prewarmed (37°C) DMEM

(Gibco, Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum

(FBS) (Gibco, Thermo Fisher Scientific, Waltham, MA) and 1% Pen/Strep(P/S) (Gibco, Thermo

Fisher Scientific, Waltham, MA). HEK293T cells were then placed in a 37°C incubator with 5%

CO2 and media was replaced every 2 days until confluency achieved.

3.2.2.2 Animal Ethics

All experimental procedures and housing conditions were approved by the UBC Animal

Care Committee and were in accordance with the Canadian Council on Animal Care (CCAC) guidelines.

3.2.2.3 Primary Culture from Sprague-Dawley Rats

Plate preparation: Hippocampal neurons from embryonic day 18 (E18) Sprague-Dawley rats (Charles River, Sherbrooke, Canada) of either sex were prepared as per below and plated either on 18 mm coverslips (Marienfeld, Lauda-Königshofen, Germany) placed in 12 well dishes, or directly on 6 well plates at a density of 130 cells/mm2. Coverslips were previously treated with concentrated nitric acid (Sigma-Aldrich, St. Louis, MO) overnight, rinsed several times with milliQ water, then sterilized with high heat (>250°C overnight). Coverslips were coated overnight with 0.5 mg/mL poly-L-lysine hydrobromide (Sigma-Aldrich, St. Louis, MO) in 0.1M borate buffer. Plates were covered and left in a biosafety cabinet overnight. Coverslips were rinsed 3X with autoclaved milliQ water and then stored in plating medium (84.75 mL

MEM with Earle’s BSS, 10 mL FBS, 2.25 mL of 20% glucose, 1 mL sodium pyruvate, 1 mL

100X GlutaMax (Gibco, Thermo Fisher Scientific, Waltham, MA), 1 mL 100X Pen/Strep

(Gibco, Thermo Fisher Scientific, Waltham, MA) (1 mL plating media per well of 12 well dish,

2 mL plating media per well of 6 well dish). Plates were placed in the 5% CO2 incubator at 37°C overnight.

Preparation of hippocampal neurons: Timed-pregnant Sprague-Dawley rats (Charles

River, Sherbrooke, Canada) were euthanized and E18 pups harvested and placed in pre-chilled 99

HBSS (Gibco, Thermo Fisher Scientific, Waltham, MA) on ice. Hippocampi were dissected out and placed in a 15 mL conical tube with 10 mL of fresh, prewarmed (37°C) HBSS (Gibco,

Thermo Fisher Scientific, Waltham, MA). Hippocampi were washed 3x with 10 mL prewarmed

HBSS (Gibco, Thermo Fisher Scientific, Waltham, MA). 4.5 mL of prewarmed HBSS (Gibco,

Thermo Fisher Scientific, Waltham, MA) was added to the hippocampi with 0.5 mL of 2.5%

Trypsin (Worthington Biochemical, Lakewood, NJ) and incubated in a 37°C waterbath for 20 minutes with gentle agitation every 5 minutes. In the final 3 minutes, 1% DNase (Sigma-Aldrich,

St. Louis, MO) was added and gently agitated to ensure mixture. Hippocampi were washed 3x with fresh, prewarmed HBSS (Gibco, Thermo Fisher Scientific, Waltham, MA). Hippocampi were resuspended in 1 mL of HBSS (Gibco, Thermo Fisher Scientific, Waltham, MA) and transferred to a 60 mm dish for trituration. Total volume was brought up to 5 mL and cell density determined using a hemocytometer. 3-4 hours after plating, plating medium was aspirated and replaced with prewarmed (37°C) maintenance media (97 mL Neurobasal medium (Gibco,

Thermo Fisher Scientific, Waltham, MA), 2 mL 50X B-27 supplement (Gibco, Thermo Fisher

Scientific, Waltham, MA), 1 mL 100X GlutaMax (Gibco, Thermo Fisher Scientific, Waltham

MA), 1 mL 100X Pen/Strep (Gibco, Thermo Fisher Scientific, Waltham, MA)). Maintenance medium was replaced with fresh medium 2-3 days following the plating.

3.2.2.4 zDHHC9 Knockout Mice

Male C57BL/6J mice (Jackson Laboratory, Sacramento, CA) were bred with female heterozygous zDHHC9 knockout mice (B6;129S5-Zdhhc9tm1Lex/Mmucd) obtained from the

Mutant Mouse Resource and Research Center (MMRRC) at the University of California, Davis.

These mice were made by homologous recombination targeting coding exon 1 (NCBI ascension

NM_172465.1). Knockout males were selected for further experiments with age-matched 100

littermate controls. All pups were weaned between 21 and 26 days following birth and ear- notched for identification. Ear notch tissue was sent to a commercial vendor for genotyping

(Transnetyx, Cordova, TN). For acute brain slice, electrophysiology litter matched knockout and wildtype males were used at P20-P30, while for in vivo local field potential recordings litter matched knockout and wildtype males were used at P50-P70.

3.2.3 Method Details

3.2.3.1 Transfection (Primary Hippocampal Cultures / HEK293T Cells)

Primary hippocampal cultures were transfected at 9-11 DIV using Lipofectamine 2000

(Invitrogen/Life Technologies, Carlsbad, CA) according to the manufacturer’s recommendations.

Briefly, for each well of a 12 well plate, 2 aliquots of 25 μL of Opti-Mem (Gibco, Thermo Fisher

Scientific, Waltham, MA) were prepared. To one aliquot, 1-3 μg of total plasmid DNA was added. To the other aliquot, 1 μL of Lipofectamine 2000 (Invitrogen/Life Technologies,

Carlsbad, CA) was added and allowed to mix for 5 minutes. Aliquots were combined and incubated for 20 minutes, then added dropwise to the well. Cells were then live imaged (10-13

DIV) or fixed (13 DIV) for subsequent experiments. HEK293T cells were transfected with

Lipofectamine 2000 (Invitrogen/Life Technologies, Carlsbad, CA) at 70% confluency in a 3:1 ratio with total plasmid DNA to be transfected and incubated for 48-72h before harvesting for biochemistry. For HEK293T cells 150 µL of Opti-Mem (Gibco, Thermo Fisher Scientific,

Waltham, MA) was used with 6 µL of Lipofectamine 2000 (Invitrogen/Life Technologies,

Carlsbad, CA).

Nucleofections were performed immediately prior to plating at 0 DIV using an Amaxa

Nucleofection Kit (VPG-1003; Lonza, Basel, Switzerland) and Amaxa Nucleofector 2b (Lonza,

Basel, Switzerland) according to the manufacturers optimized protocol (Number 101) and used 101

for experiments at 6-7 DIV. Briefly, hippocampi were dissociated as above and 500,000 neurons isolated for nucleofection. Neurons were mixed with 100 μL of Ingenio Electroporation solution

(Mirus Bio, Madison, WI) and then cDNA constructs (approximately 3 total μg) were added to the solution. The solution was then transferred to an Amaxa Nucleofection cuvette and inserted into the Amaxa Nucleofector with program G-013. Following electroporation, the solution was plated into a 6 well dish for future biochemical experiments.

3.2.3.2 Immunocytochemistry

Immunocytochemistry experiments were performed as previously reported (Sun &

Bamji, 2011). Briefly, cells were fixed for 10 min in a pre-warmed (37°C) 4% PFA/sucrose solution. Cells were washed 3X 10 min in phosphate buffered saline (PBS) and permeabilized using 0.1% Triton-X 100 (Sigma-Aldrich, St. Louis, MO)/PBS for 10 minutes at room temperature, washed 3X in PBS, and blocked for 1 hr at room temperature with 10% goat serum

(GS; Thermo Fisher Scientific, Waltham, MA) in PBS. Cells were then incubated in 1% GS/PBS with primary antibodies overnight at 4°C, washed 3X 10 min in PBS, incubated in secondary antibodies in 1% GS (Thermo Fisher Scientific, Waltham, MA)/PBS for 1 hr at RT, washed 3X

10 min in PBS, and mounted on microscope slides using Prolong Gold (Molecular Probes,

Thermo Fisher Scientific, Waltham, MA). For DAPI staining, a 1:100 dilution from a 5 mg/mL stock was applied after all other staining procedures, incubated for 5 min, washed 3X 10 min in

PBS and then mounted as above.

3.2.3.3 Biotinylation

For biotinylation experiments, neurons in 6 well plates were washed with ice-cold PBS-

CM (0.1 mM CaCl2 and 1 mM MgCl2) in 1X PBS, pH 8) and incubated for 30 min with 0.5 mg/mL NHS-SS-Biotin in ice cold PBS-CM at 4°C with gentle rocking. After incubation, cells 102

were washed once with PBS-CM and the unbound biotin quenched with 2X 8 min incubations with quenching buffer (20 mM glycine in PBS-CM). Lysis was performed using mechanical scraping in lysis buffer (1% IGEPAL-CA630 and 1 mM PMSF with Roche complete protease inhibitor tablet) and subsequently spun down at 500 x g for 5 minutes at 4°C. Samples were vortexed, run through a 26½ gauge syringe 3-4 times, and place on a nutator at 4°C for 30 min.

After nutation, samples were spun down at 16100 x g for 30 minutes at 4°C to clear the lysate.

The cell lysate was then quantified for protein using a BCA Assay Kit (Thermo Fisher Scientific) as per the manufacturer’s instructions. 10 µg of each whole cell lysate was then combined with

SDS-sample buffer (50 mM Tris-HCl, 2% SDS, 10% glycerol, 14.5 mM EDTA and 0.02% bromophenol blue with 1% β-mercaptoethanol), boiled for 5 min at 95°C and stored at -20°C as the input sample. 100-200 µg of the remaining protein sample was added to a 50 µL 50% slurry of Neutravidin-conjugated agarose beads (Thermo Fisher Scientific) that was pre-washed 3X in lysis buffer. Each sample was then brought to a total volume of 500 µL with lysis buffer and nutated at 4°C overnight. The following day beads were pelleted and washed 7X using centrifugation (500 x g for 3 min). Elution of the beads was performed using 40 µL of SDS- sample buffer with 100 mM DTT. Samples were boiled at 90°C for 5 min and then run on a

Western blot with the whole cell lysates.

3.2.3.4 Palmitoylation Assays (ABE and Acyl-RAC)

For the ABE assay, cells or hippocampi were lysed in 300-500 µl of lysis buffer (1%

Triton X-100, 100 mM NEM, 1 X Roche Complete Protease Inhibitor Tablet, 0.2% SDS, pH

7.4). The protein was left in lysis buffer overnight at 4°C to block free cysteines. The solution was the acetone precipitated and the protein pellet dissolved in 400 µL 4% SDS-Buffer (50 mM

Tris-Base, 5 mM EDTA, 4% SDS, pH 7.4). At this time, the sample was split for hydroxylamine 103

(HAM) treatments (+HAM; 50 mM Tris-Base, 150 mM NaCl, 5 mM EDTA, 1 M HAM, 1 mM

HPDP, 0.2% TX-100, Roche Complete Protease Inhibitor Tablet, pH 7.4 / -HAM same without

HAM). +HAM and –HAM samples are then incubated at room temperature for 1 hr with end- over-end rotation, then acetone precipitated. This pellet was dissolved in 110 µL of 2% SDS-

Buffer (50 mM Tris-Base, 5 mM EDTA, 2% SDS, pH 7.4) and protein concentration determined with a BCA Assay (Pierce). 150-250 µg of protein was then removed for experimental inputs, and 50 µg of protein saved for input. To both samples,1 mL of 0.2% TX-100 buffer (50 mM

Tris-Base, 150 mM NaCl, 5 mM EDTA, 0.2% TX-100, pH 7.4) with 50 µL of pre-washed agarose-streptavidin beads was added, and incubated overnight at 4°C. The following day, unbound proteins were removed by 3X 1 mL washes with 0.2% TX-100 buffer with 0.1% SDS.

Bound proteins were then eluted through reduction of the protein-biotin disulfide linkages with

30 µL of 2X loading buffer with 100 mM DTT, incubated for 10 min at 70°C. The samples were then loaded for SDS-PAGE and Western blotted for the protein of interest.

For the Acyl-RAC assay, the commercially available CAPTUREome S-palmitoylated protein kit (Badrilla, Leeds, UK) was used in accordance with the manufacturer’s guidelines with the following optimizations: During the cell lysis and free thiol blocking steps DNAse (Sigma-

Aldrich, St. Louis, MO) was added to the solution, and protein concentration was measured prior to the separation of experimental sample (Thioester Cleavage Reagent) and negative control sample (Acyl Preservation Reagent) using the BCA Assay (Pierce, Thermo Fisher Scientific,

Waltham, MA).

3.2.3.5 Western Blot Analysis

Western blotting was performed as previously described (Sun and Bamji, 2011). Primary hippocampal neurons, HEK293T cells, or whole hippocampal lysates were homogenized in an 104

ice-cold RIPA buffer (50 mM Tris-HCl, 150 mM NaCl, 1% Triton X-100 (Sigma-Aldrich, St.

Louis, MO), 1% sodium deoxycholate, 0.1% SDS, 1 mM EDTA) or lysis buffer (50 mM Tris-

HCl, 150 mM NaCl, 1% IGEPAL CA-630 (Sigma-Aldrich, St. Louis, MO), 0.5% Triton X-100

(Sigma-Aldrich, St. Louis, MO), 10% glycerol) supplemented with 1 mM PMSF and 1 Roche

Complete Protease Inhibitor Tablet. Whole cell lysates were passed through a 26½ gauge syringe

5-6 times for further mechanical disruption. Lysates were cleared by 16100 x g centrifugation for

30 min at 4°C and the solubilized protein faction was used for further biochemical experimentation. Proteins were separated by SDS-PAGE (10% resolving gels, 4% stacking gels), then immunoblotted with the indicated antibodies and visualized using enhanced chemiluminescence (Immobilon Western Chemiluminescent HRP Substrate, Millipore

Corporation, Billerica, MA / SuperSignal West Pico PLUS Chemilumescent Substrate, Themo

Fisher Scientific, Waltham, MA) on a BioRad VersaDoc 4000 (Bio-Rad Laboratories, Hercules,

CA) or a Li-Cor C-Digit Blot Scanner (LI-COR, Lincoln, NE). For reprobing blots were stripped with stripping buffer (Restore Western Blot Stripping Buffer, Themo Fisher Scientific, Waltham,

MA). Blots were then quantified using ImageJ (NIH, Bethesda, MA) software.

3.2.4 Confocal Imaging

Fixed neurons were imaged using either an inverted or upright Olympus Fluoview 1000

(FV1000) confocal microscope. Synapse imaging and analysis utilized the 60X/1.42 Oil Plan-

Apochromat objective while dendritic morphology utilized the 20X/0.75 Oil Plan-Apochromat objective. Identical acquisition parameters were used for all cells across all separate cultures within an experiment. For time-lapse live imaging we utilized MatTek glass bottom culture dishes (P35G-1.5-14-C, MatTek Corporation, Ashland, MA) and the inverted FV1000 with an environmental chamber maintained at 37°C. eGFP transfected cells were identified and imaged 105

in HEPES buffer (140 mM NaCl, 1.3 mM CaCl2, 1.3 mM MgCl2, 2.4 mM K2HPO4, 25 mM

HEPES, pH 7.4) every 24 hours post transfection up to 72 hours to ensure cell viability. Levels and contrast of confocal images were moderately adjusted in Photoshop CS6 software (Adobe

Systems, Inc., San Jose, CA).

FRAP experiments were conducted as described previously (Brigidi et al., 2014; 2015).

Briefly, dendritic spines were identified within 150 μm of the cell body, photobleached, and imaged every 5 seconds for 5 min. Spines were identified using DsRed and a circular ROI of approximately 1 μm in that spine was photobleached using the Tornado function of Olympus

Fluoview software. The fluorescence recovery of N-Ras- or TC10-GFP in the photobleached region was quantified over time using Fluoview software and normalized to an additional ROI

(to account for passive bleaching) in the same manner as previously reported (Brigidi et al.,

2014; 2015): R = (Ft – F0) / (Fi – F0), where Ft represents the intensity at a specific time, F0 represents fluorescence at the time of photobleaching and Fi designates the initial fluorescence prior to bleaching. The normalized fluorescence recovery data was subsequently analyzed in

Prism software (Graphpad, La Jolla, CA) and fit to a single exponential model, which generated plateau values for the mean R-value among each groups of cells. Plateau values and exponential models were statistically compared in Prism (Graphpad, La Jolla, CA,).

For live cell time-lapse imaging, neurons were imaged every 24 hrs post-transfection until 13-14 DIV. Neurons were identified using morphological markers as well as the gridded

MatTek dishes to ensure the same neuron was captured at every time point. Images were then overlaid in Adobe Photoshop CS6 (Adobe Systems, Inc., San Jose, CA) and moderately cropped to produce the same orientation and lengths observable over time.

106

For GABAγ2 analysis, 13 DIV neurons were imaged on a Zeiss LSM 880 with AiryScan

(Carl Zeiss Canada, North York, ON, CA) utilizing the oil 63X Objective Plan-Apochromat 1.4 with an M27 correction collar. Images were taken in a 7 slice Z-stack over a 3x3 grid, AiryScan processed, maximum intensity projected, and stitched using ZenBlack 2.3 SP1 software.

3.2.5 Quantification

3.2.5.1 Dendritic Complexity (Sholl Analysis)

To measure dendritic complexity, fixed neurons were imaged on the confocal microscope at 20X magnification and manually processed in Adobe Photoshop CS6 (Adobe Systems, Inc.,

San Jose, CA) to remove the axon and any dendrites/axons that were extraneous to the cell of interest (e.g. belonging to other neurons) to remove any background and isolate the neuron/dendrites of interest. Images were the binarized with subjective thresholding in ImageJ and run through the “Sholl Analysis” plugin (Ferreria et al., 2014; http://fiji.sc/Sholl). Sholl analysis was performed with a 5 µm starting radius, a step size of 2 µM, and an ending radius of

“NaN” (Not a Number) to ensure sampling of the entire dendritic arbor. Data was exported by the plugin and combined in a Microsoft Excel spreadsheet to create an average Sholl profile for each condition. The scale for distance was set in ImageJ in accordance with the Olympus

FV1000 dimensions for the 20X objective (0.621 µm per pixel).

3.2.5.2 Total Dendritic Length

To measure total dendritic length, fixed neurons were imaged on the confocal microscope at 20X magnification. Images were manually processed as above to isolate the cell of interest.

The image was then converted to 8-bit using ImageJ software and processed with semi-automatic tracing performed by the “NeuronJ” plugin for ImageJ (Meijering et al., 2004; http://www.imagescience.org/meijering/software/neuronj/). The scale for distance was set in 107

ImageJ in accordance with the Olympus FV1000 dimensions for the 20X objective (0.621 µm per pixel).

3.2.5.3 Synaptic Density and Total Synapses

To quantify synaptic density, fixed neurons immunostained with antibodies against

Gephyrin (to visualized inhibitory synapses) and PSD-95 (to visualize excitatory synapses) were imaged on the confocal microscope using the 60X objective with a 2X optical zoom. In order to examine only the inputs being formed onto the transfected neurons, Adobe Photoshop CS6

(Adobe Systems, Inc., San Jose, CA) was used to create “masks” of the dendrites from the 60X eGFP cell fill. In doing so, the axon and soma of the imaged cell is removed, as well as any dendrites or axons not originating from the cell of interest. The black background was then selected using the magic wand tool (non-contiguously, tolerance of 30) and the dendrite mask created was overlaid onto the synaptic marker images, limiting the visualized staining to only the dendrites from the cell of interest. The synaptic mask was then individually and subjectively thresholded to binarize the image (removing any discrepancies of minor fluctuations in staining intensities or focus between images) and synaptic puncta were evaluated. Synaptic puncta were defined as being between 0.05 µm and 3 µm, and synaptic density was calculated by obtaining the total number of puncta in the dendrite mask divided by the total length of the dendrite mask.

Puncta numbers were obtained using a custom macro that made use of the “Analyze Particles” function of ImageJ and the “Colocalization” plugin

(https://imagej.nih.gov/ij/plugins/colocalization.html), and total length was measured using

NeuronJ as above. E:I ratio was calculated by taking the density of the PSD-95 puncta and dividing by the density of the Gephyrin puncta. To evaluate the total number of synapses formed onto the transfected neurons, the same neuron was imaged at both 20X and 60X, allowing 108

determination of the average synaptic density as well as the total dendritic length. By summing these (synapses/µm x total µm) we were able to estimate the total number of inputs being formed onto the transfected cell. The scale for distance was set in ImageJ in accordance with the

Olympus FV1000 dimensions for the 60X objective with 2X optical zoom (0.103 µm per pixel).

For GABAγ2 synaptic analysis, fixed neurons immunostained with GABAγ2 and gephyrin antibodies were imaged on a LSM 880 microscope with 63X objective as described above. Final images were run through SynD (Schmitz et al., 2011) to garner Sholl analysis of puncta counts, and eGFP masked images were manually binned using the Sholl analysis plugin

(Ferreria et al., 2014). From the binned images, NeuronJ (Meijering et al., 2004) was used to manually measure the dendrite length of the neurites in each bin, and counts from SynD Sholl were used to generate a Sholl density profile.

3.2.6 Electrophysiological Methods

3.2.6.1 In Vitro Electrophysiological Recordings

Hippocampal neurons were transfected at 10DIV with eGFP and control or zDHHC9 shRNA and taken for analysis at 13-14 DIV. Recording conditions used were similar to previous publications from our lab (Brigidi et al., 2014). To obtain whole-cell patch recordings, cells were perfused at 2-4 mL per minute with an extracellular recording solution (ECS): 125 mM NaCl,

5mM KCl, 2mM CaCl2, 1 mM MgCl2, 5 mM HEPES, 33 mM D-glucose, at pH 7.3, 290 mOsm, room temperature. Tetrodotoxin (TTX; 1 µm) was added to block burst firing. Pipette resistance was 3-6 MΩ when filled with 130 mM Cs methanosulfonate, 5 mM CsCl, 4 mM NaCl, 1 mM

MgCl2, 5 mM EGTA, 10 mM HEPES, 5 mM QX-314, 0.5 mM GTP, 10 mM Na2- phosphocreatine, 5 mM MgATP and 0.1 mM spermine at pH 7.4, 290-295 mOsm. Series resistances (Rs < 30 MΩ) were uncompensated, and ΔRs tolerance was < 10%. 109

Following a 2 minute settle period, miniature excitatory post-synaptic currents (mEPSCs) were recorded at Vh = -65 mV. The voltage was then increased to +10 mV to record miniature inhibitory post-synaptic currents (mIPSCs), with CNQX (10 µm) washed in to block AMPA currents. mEPSCs and mIPSCs were analyzed with experimenter blind to genotype using

Clampfit10 (threshold 8pA); all events were checked by eye and monophasic events were used for amplitude and decay kinetics, while others were suppressed but included in frequency counts.

3.2.6.2 Brain Slice Preparation

Wildtype and zDHHC9 knockout littermate P20-P30 males were used for acute brain slice patch-clamp experiments in accordance with Canadian Council for Animal Care guidelines.

Mice were deeply anaesthetized using isoflurane (5% in O2), and then decapitated once anesthesia was confirmed. The brain was rapidly removed and transferred to ice-cold sucrose- artificial cerebral spinal fluid (sucrose-aCSF) containing in mM: 214 sucrose, 26 NaHCO3, 1.25

NaH2PO4,11 glucose, 2.5 KCl, 0.5 CaCl2, 6 MgCl2, and bubbled with 95%O2:5% CO2. The whole brain was then glued to a cutting chamber in a vibrating microtome (VT 1200, Leica,

USA) which was filled with ice-cold sucrose-aCSF. Horizontal brain slices (350 μm thick) of the ventral hippocampus were cut and incubated at 33-35°C in aCSF containing in mM: 126 NaCl,

2.5 KCl, 26 NaHCO3, 1.5 NaH2PO4, 2 CaCl2, 2 MgCl2, 10 glucose and bubbled with 95%

O2:5% CO2. Slices were maintained in a holding chamber with aCSF at 33-35°C until transferred to a recording chamber superfused with aCSF and maintained at 33-35°C. CA1 neurons were visualized using a DIC microscope and infrared camera (Slicescope 6000,

(Scientifica, UK) and/or Axioskop 2-FS Plus, (Carl Zeiss, USA) with IR 1000 (DAGE-MTI,

USA)) and visually identified by their location, morphology and orientation.

110

3.2.6.3 Acute Brain Slice Electrophysiology

Whole-cell patch-clamp recordings were performed to record neuronal excitability and synaptic properties. All recordings were undertaken using a Multiclamp 700B amplifier and pClamp software version 10 and/or 11 (Molecular devices). The recording chamber was grounded with an Ag/AgCl pellet. Recording pipettes (4-6 MΩ) were made with borosilicate glass (Sutter Instruments). For current-clamp recordings, the glass pipettes were filled with the intracellular solution containing in mM: 120 K-gluconate, 10 HEPES, 1 MgCl2, 1 CaCl2, 11

KCl, 11 EGTA, 4 MgATP, 0.5 NaGTP, with pH adjusted to 7.2 using KOH and osmolarity adjusted to 290 mOsm/kg using D-mannitol. The liquid junction potential for current-clamp solutions was calculated as +13.3 mV, which was corrected off-line in current clamp recordings.

To evaluate the current vs frequency relationship action potential firing was recorded in response to hyperpolarization and depolarization, DC current was injected from -110 pA to +200 pA in 10 pA increments for a duration of 1000 ms at the cells intrinsic resting membrane potential.

Membrane potential responses under current clamp conditions were sampled at 50 kHz and filtered at 10 kHz. Bridge balance was monitored during recordings and any neurons displaying bridge balance values greater than 20 MΩ were discarded. Capacitance neutralization was performed between 3.8 and 4.2 pF. To perform current-frequency analyses the firing frequency for each current injection was plotted individually for each cell and fitted with a linear curve, with the slope value used to perform statistical analyses t-test between knockout and wild-type mice.

Whole-cell voltage-clamp recordings were performed to record spontaneous and miniature excitatory and inhibitory postsynaptic currents (sEPSCs/sIPSCs and mEPSCs/mIPSCs) respectively. For sEPSCs, mEPSCs and sIPSCs experiments the internal pipette solution 111

contained 120 mM potassium-gluconate, 10 mM HEPES, 1 mM MgCl2, 1 mM CaCl2, 11 mM

KCl, 11 mM EGTA, 4 mM Mg-ATP, 0.5 mM Na-GTP, pH adjusted to 7.2 using KOH and osmolarity adjusted to 290 mOsm/kg using D-mannitol. For mIPSCs experiments, K-gluconate or KCl was substituted by Cs-methanesulphonate. To record sEPSCs/mEPSCs the GABA receptor antagonist picrotoxin (100 µM) was added to the aCSF and to record sIPSCs/mIPSCs the NMDA receptor antagonist D-APV (100 µM), and AMPA receptor antagonist CNQX (20

µM) were added to the aCSF. In addition to the GABAergic and glutamatergic receptor blockers, the Na+ channel blocker tetrodotoxin (TTX, 1 µM) was added to the aCSF for recordings of mEPSCs and mIPSCs to block action potential-dependent synaptic transmission. To evaluate sEPSCs/mEPSCs cells were held at a membrane potential of -50mV, and sIPSCs/mIPSCs held at a membrane potential of +20mV during a 60-second gap-free recording. Data acquisition was sampled at 20 kHz and filtered at 2.4 kHz. Voltage-clamp recordings were not liquid-junction potential corrected. Recordings with a series resistance of greater than 20 MΩ were discarded and series resistance was compensated to 70%. sEPSCs/mEPSCs and sIPSCs/mIPSCs amplitude and inter-event intervals were measured by creating a template for detection in Clampfit11.

Graphing and statistical analyses were performed using Origin 2018 (OriginLab 2018b version).

3.2.6.4 In Vivo Local Field Potential Recordings

Litter matched knockout and wildtype males (P50-P70) were used for in vivo seizure analysis. Both wild-type and DHHC9 null animals were anesthetized using a reversible anesthetic combination of dexmedetomidine/fentanyl/midazolam and stereotaxically implanted with stainless steel wire electrodes in the following brain regions: right dorsal CA1 hippocampus

(bregma -2.2 mm, midline +1.5 mm, depth = 1.5 mm), left S1 cortex (bregma +0.4 mm, midline

-2.0 mm, depth = 0.5 mm), left posterior dentate gyrus (bregma +3.6 mm, midline -2.5 mm, 112

depth = 3.0 mm), right visual cortex (bregma -3.6 mm, midline +2.0 mm, depth = 0.5 mm), a reference electrode in the left visual cortex (bregma +4.6 mm, midline - 2.5 mm, depth = 0.5 mm) and a ground electrode connected subcutaneously to the neck muscles. Wire electrodes were connected to a 4-channel electronic interface board (ALA Systems, USA) which was fitted to the skull with dental cement.

Animals were allowed to recover for 14 days and then the interface connected to a

W2100 W4 wireless headstage (Multichannel Systems, Germany) and free-moving recordings acquired at a frequency of 1 kHz over an 2 hour period. Recordings sessions were performed during the dark cycle. Analysis was performed on EEG data using a custom Matlab script (Cain and LeDue, UBC) designed to semi-automatically identify seizures as high power wavelets in the

2-15 Hz range with each seizure then confirmed by the study team.

3.2.7 Statistical Analysis

All data values are presented as a mean ± standard error of the mean (SEM). For all imaging experiments, the value of “n” refers to the number of cells used per condition, over at least 3 separate cultures, with the exception of FRAP analysis where “n” refers to the number of spines. No statistical analysis was used to pre-determine the sample sizes used for our experiments; however, the sample sizes are consistent with published literature from our lab and others (Thomas et al., 2012; Fukata et al., 2013; Brigidi et al., 2014; Brigidi et al., 2015).

Statistical analysis was performed in GraphPad Prism (La Jolla, CA, USA) and electrophysiological statistical analyses using Originlab v8.6 (Northampton, MA, USA) software. Statistical significance was determined by one-way ANOVA, repeated measures one- way ANOVA with application of Tukey’s post hoc test, or t-test where applicable. Statistical significance was assumed when p < 0.05. In all figures, * = p < 0.05, ** = p < 0.01, and *** = p 113

< 0.001, as a determined by Prism or Originlab software. All figures were generated using Adobe

Illustrator CS6 software in conjunction with Adobe Photoshop CS6 (Adobe Systems Inc., San

Jose, CA).

3.3 Results

3.3.1 zDHHC9 is Expressed in Neurons and Localized to Golgi Satellites Associated with

Excitatory and Inhibitory Synapses

zDHHC9 is highly enriched in brain tissue (Human Protein Atlas; Uhlén et al., 2015) where it is ubiquitously expressed in all brain regions (Allen Brain Atlas; Lein et al., 2007) and in all major neuronal cell types examined (Mancarci et al., 2017). To further examine the localization of zDHHC9 in cells of the hippocampus, 14 days in vitro (DIV) primary rat hippocampal cultures were immunostained for zDHHC9 together with cell- and organelle- specific markers. In our embryonic hippocampal culture system, zDHHC9 was expressed in both excitatory and inhibitory neurons, but not non-neuronal cells (Figure 3.1A). Unlike zDHHC5

(Bridigid et al., 2015), zDHHC9 was not localized to the plasma membrane (Figure 3.1B), but localized to the somatic Golgi (in agreement with previous work (Swarthout et al., 2005; Ohno et al., 2006; Figure 3.1C) and notably also to Golgi satellites associated with synapses (Figure

3.1D,E). Indeed, zDHHC9 was associated with 81.7% ± 4.4% of all excitatory synapses (defined by the colocalization of VGLUT-1 and PSD-95), 82.8% ± 3.1% of all inhibitory synapses

(defined by the colocalization of VGAT and Gephyrin), as well as 85.8% ± 2.8% of all Golgi satellites (defined by labeling with Giantin (Figure 3.1D,E)). Giantin is a faithful marker of

Golgi satellites within dendrites (Pierce, Mayer & McCarthy, 2001; see Figure 3.2 for colocalization with ER-Golgi intermediate compartment (ERGIC; Bowen et al., 2017)). 53.7% ±

1.8% of zDHHC9 puncta associated with Golgi satellites at excitatory synapses (colocalized 114

Figure 3.1 zDHHC9 is Localized to Golgi Membranes Associated with Excitatory and Inhibitory Synapses (a,c,d,e) Representative confocal images of 13 DIV rat hippocampal cultures. (a) zDHHC9 is expressed in excitatory neurons (NeuN positive, magenta arrows) and inhibitory neurons (GAD-65 positive, cyan arrow) but not non-neuronal cells (DAPI positive, NeuN negative, yellow arrows). White arrows denote zDHHC9 in neurons in merged image. Scale bar = 100 µm. (b) Biotinylation assays demonstrate that unlike zDHHC5, zDHHC9 is not localized at the plasma membrane, indicating its location at endomembranes. N=3 blots from 3 cultures. (c) zDHHC9 is colocalized with Giantin in the soma and along dendrites indicating its localization at the somatic Golgi as well as Golgi satellites. Scale bar = 20 µm. (d) zDHHC9 is localized to a subset of excitatory (VGLUT1/PSD95) and inhibitory (VGAT/Gephyrin) synapses. Scale bar = 5 µm. (e) zDHHC9 is localized to Golgi satellites (Giantin) at excitatory (PSD95) and inhibitory (Gephyrin) synapses. Scale bar = 5 µm. 115

giantin/PSD-95 puncta) and 32.6% ± 2.2% of zDHHC9 puncta associated with Golgi satellites at inhibitory synapses (colocalized giantin/gephyrin puncta). Together, this demonstrates that zDHHC9 is localized to the somatic Golgi as well as Golgi satellites associated with both excitatory and inhibitory synapses.

Figure 3.2 zDHHC9 Colocalizes with Giantin and ERGIC

(a,b) Confocal images of 13 DIV hippocampal neurons immunostained for zDHHC9, Giantin, and ER-to-Golgi Intermediate Complex (ERGIC). zDHHC9, Giantin, and ERGIC are localized to the soma and to discreate puncta within dendrites. Scale bar = 5 µm (a), 20 µm (b).

3.3.2 zDHHC9 Promotes Dendritic Outgrowth and Maintenance

Dendritic abnormalities, including reductions in the number, length, and/or complexity of dendritic branches, are some of the most consistent anatomical correlates of intellectual disability

(Huttenlocher, 1970, 1974; Purpura, 1974, 1975a, 1975b; reviewed in Kaufmann & Moser,

2000). To determine whether zDHHC9 contributes to dendritic morphology, hippocampal cells were transfected at 10 DIV with eGFP plus zDHHC9 shRNA to knockdown zDHHC9 (validated in Figure 3.3), or zDHHC9 plus its known cofactor, GCP16 (Swarthout et al., 2005; Mitchell et

116

al., 2012) to increase zDHHC9 activity. Representative, unaltered images of hippocampal

neurons transfected with eGFP and the indicated constructs, and the masking, tracing, and Sholl

process can be seen in Figure 3.4.

Figure 3.3 shRNA Validation of zDHHC9 in Hippocampal Neurons and HEK293T Cells

(a) Hippocampal neurons were nucleofected at time of plating with a control scrambled shRNA (shRNA-C) or zDHHC9 shRNA. Transfection efficiency of hippocampal neurons using nucleofection was ~50%, accounting for residual protein levels. N=3 separate blots from 3 cultures. (b) HEK293T cells were transfected at 70% confluency with a control scrambled shRNA (shRNA-C), zDHHC9 shRNA, and zDHHC9 shRNA plus shRNA resistant zDHHC9 (zDHHC9R). N=3 separate blos from separate passages of HEK293T cells. Values at the bottom of each blot represent percent protein remaining in shRNA transfected lysates compared to controls.

Figure 3.4 Representative Confocal Images, Neuron Masking, ImageJ Tracing, and Sholl Analysis

Unaltered representative 13DIV images ofhippocampal neurons transfected with eGFP and the indicated constructs. The Control neuron is shown on the right, with the masking process, NeuronJ tracing arbor, and Sholl mask represented. Refer to Methods for specifics of tracing and Sholl anaylsis. Scale bar = 100 µm. 117

zDHHC9 knockdown resulted in a significant reduction in dendritic length (Figure

3.5A,C,D) and complexity (Figure 3.5A,B), but not the total number of dendrites (Figure

3.5A,E). The knockdown phenotypes were rescued through co-expression of shRNA resistant zDHHC9 (zDHHC9R). While co-expression of zDHHC9 and GCP16 significantly increased dendrite length (Figure 3.5A,C,D) and complexity (Figure 3.5A,B), overexpression of either zDHHC9 or GCP16 alone did not impact neuron morphology compared to control cells (Fig 2A-

E), consistent with results that demonstrate that zDHHC9 and GCP16 are required in a 1:1 ratio for enzymatic active (Mitchell et al., 2014).

To determine whether dendritic length depends on the palmitoylation function of zDHHC9, we generated a catalytically inactive mutant in which the cysteine residue of the

DHHC domain was mutated to a serine (zDHHS9R) (Lobo et al., 2002; Mitchell et al., 2006;

2010). We also generated two point mutations in zDHHC9 previously identified in patients with

XLID and shown to decrease its enzymatic activity (zDHHC9 R148W and zDHHC9 P150S,

(Raymond et al., 2007; Mitchell et al., 2014)). While zDHHC9R rescued the shRNA-mediated decrease in dendrite arbor length, the palmitoylation-deficient mutants, DHHS9R, R148WR and

P150SR, did not, indicating that the palmitoylation activity of zDHHC9 is essential for zDHHC9- mediated regulation of the dendritic arbor (Figure 3.5F). To investigate whether zDHHC9 promotes dendritic outgrowth or stabilization, neurons were imaged every 24 hours for 72 hours post-transfection using time-lapse imaging. zDHHC9 knockdown resulted in the retraction of dendrites, whereas overexpression of zDHHC9 and GCP16 resulted in a significant increase in dendrite outgrowth (Figure 3.5G,H), demonstrating that zDHHC9 is required for both active dendritic growth and arbor maintenance.

118

Figure 3.5 zDHHC9 Promotes Dendrite Outgrowth and Maintenance

(a) Representative confocal images of 13 DIV cultured rat hippocampal neurons transfected with eGFP and the indicated constructs at 10 DIV. Neurons have been masked to remove axons and background as per Methods. Scale bar = 100 µm. (b-e) zDHHC9 knockdown (shRNA) decreases while zDHHC9 and GCP16 overexpression (zDHHC9+GCP16) increases dendritic arborization and complexity (b), total dendrite length (c), and average length of individual dendrites (d). The average number of dendrites per neuron was unchanged (e). N = 40 neurons per condition, >3 cultures. **P<0.01, ***P<0.001; one-way ANOVA (B:F=14.60, C:F=17.17, D:F=12.85, E:F=1.63); Tukey’s post-hoc; mean ± SEM. (f) zDHHC9 shRNA-mediated decrease in total dendritic length was rescued by shRNA resistant zDHHC9 (DHHC9R) but not by zDHHC9 palmitoylation- defective mutant (zDHHS9R) or two clinical mutations shown to inhibit zDHHC9 palmitoylation (R148WR, P150SR). N=40 neurons per condition, >3 cultures. ***P<0.001, one-way ANOVA (F=18.87), Tukey’s post- hoc; mean ± SEM. (g) Time lapse confocal images of neurons expressing eGFP plus the indicated constructs 24, 48, and 72 hours post-transfection. Red arrows denote areas of growth and blue arrows areas of retraction. Scale bar=100 µm. (h) Knockdown of zDHHC9 results in a retraction of dendrites whereas overexpression of zDHHC9+GCP16 enhances dendritic outgrowth. N=10 neurons per condition, >3 cultures. *P<0.05, **P<0.01, ***P<0.001; one-way ANOVA (48hr:F=25.01, 72hr:F=43.81); Tukey’s post-hoc; mean ± SEM. 119

3.3.3 zDHHC9-Mediated Palmitoylation of Ras is Essential for Dendritic Outgrowth

The small GTPases, H- and N-Ras, are the only confirmed substrates for zDHHC9

(Swarthout et al., 2005), with previous work demonstrating that Ras promotes dendritic growth and arborization (Heumann et al., 2000; Gartner et al., 2004; Kumar et al., 2005). We first validated expression of pooled Ras shRNA (which targets both H- and N-Ras, Figure 3.6).

Figure 3.6 Validation of Pooled Ras shRNA

Representative blots from nucleofected rat hippocampal neurons demonstrating efficacy and efficiency of pooled Ras shRNA constructs designed to target both H- and N-Ras. N=3 blots from 3 separate cultures. Values at the bottom of each blot represent percent protein remaining in shRNA transfected lysates compared to controls.

Here we demonstrate that zDHHC9 promotes dendrite outgrowth and maintenance by palmitoylating H- and N-Ras (Figure 3.7A,B). First, expression of pooled Ras shRNA attenuated total dendrite length similar to neurons expressing zDHHC9 shRNA and neurons expressing both

Ras shRNA and zDHHC9 shRNA, suggesting they may function in the same pathway (Figure

3.7A). While expression of N-RasR was able to rescue the knockdown phenotype, the palmitoylation deficient N-Ras (N-Ras C181SR) did not, indicating that the palmitoylation of N-

Ras is required for its ability to regulate dendrite length (Figure 3.7A). Second, cells overexpressing N-Ras or zDHHC9/GCP16, or co-expressing zDHHC9/GCP16 plus N-Ras, exhibited similar increases in dendritic length, suggesting they may function in the same pathway

(Figure 3.7B). Of note, overexpression of N-RasC181SR did not increase dendritic length, but

120

rather acted as a dominant-negative (Figure 3.7B), phenocopying zDHHC9 and Ras knockdown

(Figure 3.7A), further demonstrating that Ras palmitoylation is essential for its ability to regulate dendrite outgrowth. Third, Ras knockdown abolished the effects of zDHHC9/GCP16 overexpression, and overexpression of N-Ras was not able to rescue the zDHHC9 knockdown phenotype (Figure 3.7B). Together, this demonstrates that zDHHC9 regulates dendrite outgrowth through the palmitoylation of Ras.

Evidence suggests that Ras can activate distinct signaling pathways depending on its subcellular localization (Mor & Phillips, 2006; Omerovic & Prior, 2009; Aran & Prior, 2013;

Plowman & Hancock, 2005), and that palmitoylation of Ras is required for its localization to the plasma membrane (Eisenberg et al., 2013; Rocks et al., 2005; Goodwin et al., 2005; Song et al.,

2013; Lynch et al., 2015). However, none of these studies investigated the role of zDHHC9- mediated palmitoylation of Ras and the subsequent effects of distinct signaling effectors. In the following section, we show that zDHHC9 mediates dendrite outgrowth and maintenance by targeting Ras to the membrane where it has been shown to specifically activate ERK and suppresses JNK pathways.

Using fluorescence recovery after photobleaching (FRAP), we confirmed that zDHHC9 mediated palmitoylation of N-Ras targets it to the membrane. To delineate defined regions of interest (ROIs), N-Ras-GFP fluorescence in spines within 150 µm of the cell body was photobleached (Figure 3.7C). The fluorescence recovery in the ROI was then imaged every 5 s over a 5 min period and normalized to an ROI in an adjacent spine as previously described

(Figure 3.7D; Brigidi et al., 2014; 2015). Figure 3.7E and 3.7F show normalized FRAP recovery curves as a single exponential fit for each condition and plateau values (mobile fraction at 5 min). In neurons expressing control shRNA the fluorescence recovery of N-Ras-GFP plateaued 121

at 68.3% ± 1.59% (Figure 3.7E,F), consistent with previous observations of N-Ras mobility

(Goodwin et al., 2005; Goodwin & Kenworthy, 2005; Eisenberg et al., 2011). Neurons expressing zDHHC9 shRNA exhibited a significant increase in fluorescence recovery indicating an increase in the mobile fraction of N-Ras-GFP (Figure 3.7E,F), consistent with the demonstration that palmitoylation-deficient N-Ras accumulates in the cytosol (Choy et al.,

1999). This effect was rescued by expression of wildtype zDHHC9R but not the palmitoylation- defective zDHHS9R, or zDHHC9 R148W and P150S (Figure 3.7E,F).

Ras has been shown to traffic between the endoplasmic reticulum (ER), Golgi apparatus, and plasma membrane of various cell lines, and is capable of differential signaling from these locations (Choy et al., 1999; Chiu et al., 2002; Bivona & Phillips, 2003; Aran & Prior, 2013;

Pedro et al., 2017). Specifically, localization of Ras at the plasma membrane mediates sustained

ERK activation, while localization of Ras to ER/Golgi leads to activation of JNK (Chiu et al.,

2002; Matallanas et al., 2006). While studies indicate that JNK and ERK play important roles in many neuronal processes, including dendritic growth and arborization (Björkblom et al., 2005;

Rosso et al., 2005; Vaillant et al., 2002; Ha & Redmond, 2008), it is unclear whether in neurons these signaling pathways may be controlled by zDHHC9-mediated palmitoylation and Ras localization. As predicted, knockdown of zDHHC9 resulted in a reduction in N-Ras palmitoylation (Figure 3.7G) with a concomitant increase in phosphorylated (active) JNK and reduction in phosphorylated (active) ERK (Figure 3.7H,I). In contrast, overexpression of zDHHC9/GCP16 resulted in an increase in Ras palmitoylation (Figure 3.7G), and a concomitant increase in phosphorylated ERK and reduction in phosphorylated JNK (Figure 3.7H,I). Together, this suggests that zDHHC9 regulates dendrite arbor growth or retraction depending on a

122

Figure 3.7 zDHHC9 Enhances Dendritic Growth Through Palmitoylation of Ras-GTPase

(a,b) Cells were transfected with eGFP and the indicated constructs at 10 DIV and fixed and imaged at 13 DIV. N=40 neurons per condition, >3 cultures. ***P<0.001; one-way ANOVA (A:F=34.6, B:F=34.95); Tukey’s post-hoc; mean ± SEM. (c) Cells were identified with dsRed cell fill, and N-Ras-GFP fluorescence within spines photobleached at 0s within a 1 µm ROI (red circle). Images were cropped to a single spine, and pseudocolored as a heat map. Scale bar = 1 µm. (d) Fluorescence recovery within photobleached and non- photobleached ROIs were normalized for bleaching. (e-f) FRAP analysis demonstrates an increase mobility of Ras following zDHHC9 knockdown. Points with error bars represent the mean ± SEM, and solid lines represent a single exponential fit. (f) Statistical tests compare the plateau values from exponential fits ± SEM. N=20 neurons, >3 cultures. ***P<0.001; one-way ANOVA (F=22.64), Tukey’s post-hoc, mean ± SEM. (g) Acyl-biotin exchange (ABE) assay from 13 DIV hippocampal cultures nucleofected at the time of plating with the indicated constructs demonstrates a decrease in Ras palmitoylation upon zDHHC9 knockdown and an increase in Ras palmitoylation upon zDHHC9 overexpression. Omission of NH2OH is used as a control for nonspecific labeling with biotin. (h) Western blots from hippocampal neurons nucleofected at time of plating with the indicated constructs and assayed at 6-7 DIV. zDHHC9 knockdown decreases the levels of phosphorylated ERK and increases the levels of phosphorylated JNK, while zDHHC9 overexpression increases phosphorylated ERK and decreases phosphorylated JNK levels. Red boxes denote phosphorylated proteins. (i) Densitometric analysis of blots from (h). N=3 blots, 3 separate cultures. ***P<0.001; one-way ANOVA, **P<0.05; one-way ANOV; Tukey’s post-hoc; mean ± SEM. 123 combination of the level of Ras palmitoylation, its differential targeting to different subcellular domains, and the activation of distinct signal transduction pathways.

3.3.4 zDHHC9 is Required to mMaintain the Balance Between Excitatory and Inhibitory

Synapses

Post-mortem analysis of neurons from patients with IDs are often associated with dendritic spine dysfunction, and specifically, altered spine densities (Huttenlocher, 1970;1974;

Purpura, 1974, 1975a, 1975b; Cragg, 1975; reviewed in Kaufmann & Moser, 2000), with the PSD-95 VGlut1 Merge severity of these deficits often correlated with the severity of the ID (Purpura, 1974; Humeau et al., 2009). As excitatory hippocampal synapses form on or near spines, we next examined synaptic inputs being formed onto zDHHC9 mutant neurons. 10 DIV neurons were transfected with eGFP plus the indicated constructs and immunostained at 13 DIV for the excitatory post- Gephyrin VGAT Merge synaptic marker, PSD-95, and the inhibitory post-synaptic marker, gephyrin, which we established as faithful markers of synapses in control and shRNA conditions (Figure 3.8A-D, H-

K). To quantify the synapses being formed onto our transfected cells, eGFP masks were generated (Figure 3.8G) and the subsequent analysis is from the density of PSD-95 and gephyrin puncta quantified within the masks.

zDHHC9 knockdown resulted in a significant increase in the density of PSD-95 puncta

(Figure 3.9a,b; Figure 3.8e demonstrates comparable increase in excitatory synapse density), without affecting PSD-95 puncta size or integrated density (Figure 3.8h,i). The increase in the density of excitatory synapses may be attributable in part to the decrease in dendrite length

(Figure 3.5c). Indeed, the total number of excitatory synapses being formed onto zDHHC9 knockdown cells was not significantly different from control cells (Figure 3.10a). In contrast, zDHHC9 knockdown results in a decrease in gephyrin density (Figure 3.9a,b; Figure 3.8f 124

Figure 3.8 Pre- and Post-Synaptic Markers of Excitatory and Inhibitory Synapses in Hippocampal Neurons

(a-b) Confocal images of 13 DIV neurons immunostained for PSD-95 and VGlut1 (a) or Gephyrin and VGAT (b). Merged images show excitatory and inhibitory synapses respectively. Scale bar =20 µm. (c-d) Quantfication of colocalized puncta divided by the total number of PSD-95 or gephyrin puncta, showing that PSD-95 and gephyrin are faithful markers of synapses. (e-f) Quantification of synaptic puncta and colocalized synaptic puncta in cells expressing zDHHC shRNA. (g) Representative images demonstrating eGFP masking and quantification of excitatory (PSD-95) and inhibitory (Gephyrin) synapses formed onto transfected cells. The eGFP cell fill is used to create a mask over the Gephyrin and PSD-95 synaptic markers, removing puncta outside the cell being analyzed. Scale bar = 10 µm. (h-k) There is no significant difference in the size of PSD-95 (h) or Gephyrin puncta (j), and no significant changes in the integrated density of PSD-95 (i) or Gephyrin puncta (k) in cells expressing zDHHC9 shRNA. N = 10 neurons for a-d, 30 neurons over 3 separate cultures for e-f, j-k. ***P<0.001; t-test. demonstrates comparable decrease in inhibitory synapse density) and total number of inhibitory

125

demonstrates comparable decrease in inhibitory synapse density) and total number of inhibitory synapses (Figure 3.10a), with no significant change in gephyrin puncta size or integrated density

(Figure 3.8j,k). Together, this indicates that zDHHC9 plays an important role in the formation

Figure 3.9 zDHHC9 is Required to Maintain Excitatory/Inhibitory Synapse Balance

(a) Representative confocal images of rat hippocampal neurons transfected at 10 DIV with eGFP plus the indicated constructs and immunostained at 13 DIV for the excitatory post-synaptic marker, PSD-95, and the inhibitory post-synaptic marker, Gephyrin. Masks were created of the eGFP channel as per Supplementary Figure 4G, and applied to the PSD-95 and Gephyrin channels. Scale bar = 10 µm. (b-c) The density of excitatory and inhibitory synapses is differentially regulated by zDHHC9 (b), resulting in an alteration in the ratio of excitatory:inhibitory synapses (c). Asterisks denote significant changes in PSD-95 density and hashtags denote significant changes in Gephyrin density compared to control. N=40 neurons per condition, >3 cultures. **/## P<0.01, ***/### P<0.001; one-way ANOVA (B:PSD-95 F=9.794, Gephyrin F=24.74, C:F=6.575), Tukey’s post-hoc; mean ± SEM. (d) Ras knockdown decreases while N-Ras overexpression increases the density of PSD-95 puncta. This is in contrast to zDHHC9 knockdown which leads to increased PSD-95 puncta density. While zDHHC9 knockdown decreases gephyrin puncta, Ras knockdown has no effect. (e) Differential regulation of excitatory and inhibitory synapse density differentially alters the ratio of excitatory:inhibitory synapses. Asterisks denote significant changes in PSD-95 density and hashtags denote significant changes in Gephyrin density compared to control. N=40 neurons per condition, >3 separate cultures. */# P<0.05, **/## P<0.01, ***/### P<0.001; one-way ANOVA, D:PSD-95 F=34.09, Gephyrin F=10.35, E:F=41.68, Tukey’s post- hoc; mean ± SEM.

126

and/or maintenance of inhibitory synapses. While the zDHHC9 knockdown phenotype was rescued in cells expressing zDHHC9R, it was not rescued in cells expressing zDHHS9R, zDHHC9 R148WR or zDHHC9 P150SR, indicating that the enzymatic activity of zDHHC9 is essential for its ability to regulate inhibitory synapse number (Figure 3.9a,b). While control neurons exhibited an E:I synapse ratio of 1.4:1 in agreement with previous reports (Liu et al.,

2004), disrupting zDHHC9 function increased this ratio to 2.4:1 (Figure 3.9c, e).

Figure 3.10 Total Synapses for zDHHC9 and Ras Calculated Using Synapse Density and Total Dendritic Length

(a) zDHHC9 knockdown does not impact total number of excitatory synapses but significantly reduces the total number of inhibitory synapses. (b) Ras knockdown impacts both excitatory and inhibitory synapses, which may be an effect of the significant reduction in total length. N=40 neurons per condition, >3 cultures. */# P<0.05, **/## P<0.01, ***/### P<0.001; one-way ANOVA, A:PSD-95 F=4.043, Gephyrin F=40.37, B:PSD-95 F=54.23, Gephyrin F=63.76, Tukey’s post-hoc; mean ± SEM.

We next examined whether zDHHC9-mediated changes in synapse density is dependent on Ras palmitoylation. Ras knockdown decreased the density of PSD-95 puncta (Figure 3.9d) consistent with previous findings demonstrating that Ras mutants or inhibitors decrease spine density (Kumar et al., 2005), but opposite to the effects of zDHHC9 knockdown which increased

PSD-95 density (Figure 3.9b). The Ras knockdown-mediated decrease in PSD-95 density could be rescued by both N-RasR and the palmitoylation-deficient N-Ras C181SR (Figure 3.9d),

127

suggesting that the palmitoylation of Ras is not required for the promotion of excitatory synapse formation. This is supported by the fact that overexpression of both N-RasR and N-RasC181SR increased the density of PSD-95 puncta (Figure 3.9d), further uncoupling zDHHC9 and Ras in the regulation of excitatory synapse formation. Neither Ras knockdown nor N-Ras overexpression significantly altered the density of gephyrin puncta (Figure 3.9d) demonstrating that zDHHC9 does not promote the formation of inhibitory synapses through the palmitoylation of Ras.

To further investigate the inhibitory synaptic phenotype, we turned our attention to

GABA receptor subunits, as gephyrin is not always associated with GABA receptor subunits

(Tretter et al., 2012) and a gephyrin-independent GABA receptor clustering mechanism exists

(Knuesel et al., 1999; Fischer et al., 2000; Sassoe-Pognetto & Fritchy, 2000; Kneussel et al.,

2001; Levi et al., 2004). We therefore also examined the distribution of the GABAR γ2 subunit, which is important for GABA receptor localization, targeting, and anchoring (Essrich et al.,

1998; Alldred et al., 2005; Christie et al., 2006). Similar to gephyrin, there was a significant decrease in the total number of GABAR γ2 clusters on zDHHC9 knockdown cells (Figure

3.11a). However, we also observed a significant increase in GABAR γ2 density (Figure 3.11b)

Figure 3.11 Loss of zDHHC9 Affects GABA γ2 Clustering

(a) zDHHC9 knockdown results in a reduction in the total levels of GABA γ2. (b) The density of GABA γ2 is increased in zDHHC9 knockdown cells, specifically in the proximal stretch of dendrite 20-60 µm from the soma (c). N=30 neurons per condition, 3 cultures. ***P<0.001; T-test, t=4.396. 128

and in particular, an increase in the de nsity of GABAR γ2 in proximal (20-60 m) dendrites

(Figure 3.11c). Together, this indicates that zDHHC9 plays an important role in the formation and/or maintenance of inhibitory synapses.

We next determined if the observed shift in E:I synapse ratio leads to a functional difference in excitatory and inhibitory synaptic transmission in zDHHC9 knockdown cultures compared to controls, using whole-cell voltage-clamp recordings to record miniature excitatory and inhibitory post-synaptic currents (mEPSCs / mIPSCs). Cumulative probability analysis revealed an increase in the amplitude of AMPA-receptor mediated events at excitatory synapses

(mEPSC amplitude) following zDHHC9 knockdown (a significant effect of condition upon all amplitude bins) (Figure 3.12a) suggesting an increase in current flux at the majority of

Figure 3.12 In Vitro Electrophysiological Recordings

Cumulative probability plots of mEPSC and mIPSC recordings from cultured neurons. (a) Analysis revealed a significant effect of condition (two-way ANOVA, P<0.0002, F(1,63) = 15.1) and interaction (two-way ANOVA, P<0.0001, F(30,1890)=8.5) on mEPSC amplitude in zDHHC9 knockdown cells (multiple comparisons shown, * P<0.05, ** P<0.01, ***P<0.001 by Holm-Sidak post-test). (b) There was also a significant interaction between condition and mISPC amplitude in DHHC9 knockdown cells (two-way ANOVA, P<0.0001, F(30,900)=3.5). (h,i) There were no significant effect of zDHHC9 knockdown on mEPSC (c) or mISPC 129 frequency (d). glutamatergic synapses. The amplitude of GABA-mediated events at inhibitory synapses

(mIPSC amplitude) was also increased, but with the largest difference occurring within medium- sized events, rather than across all amplitude bins, suggesting an increase in current flux within a subset of synapses (significant condition x amplitude interaction effect) (Figure 3.12b).

In contrast to amplitude increases, we observed no change in the frequency of either mEPSCs or mIPSCs, as demonstrated by similar interevent interval cumulative probability plots

(Figure 3.12c,d). A lack of change in mEPSC frequency is consistent with the conservation of total excitatory synapse number in knockdown cells (Figure 3.10a). However, the lack of change in mIPSC frequency (Figure 3.11d) initially appeared to be in contrast to the decrease in inhibitory synapse number as calculated by decreased overall gephyrin and GABAR γ2 clusters

(Figure 3.10a, Figure 3.11a). As it has been shown that whole cell recordings may be biased toward somatic and proximally-localized synapses (Williams & Mitchell, 2008), a plausible explanation for the observed increase in amplitude and maintained frequency of events is the increased proportion of perisomatic/proximal inhibitory synapses (Figure 3.11c). Alternatively, the overall reduction in inhibitory synapses may be compensated for by an increase in presynaptic release in the remaining synapses, leading to an mIPSC frequency similar to that seen in control neurons.

3.3.5 zDHHC9 Promotes Inhibitory Synapse Formation Through the Palmitoylation of

TC10

The enzymatic function of zDHHC9 is important for the promotion of inhibitory synapses

(Figure 3.9a,b), however inhibitory synapse formation does not appear to depend on the only known zDHHC9 substrate, Ras (Figure 3.9d). We therefore took a candidate approach to identify the substrate regulating zDHHC9-mediated inhibitory synapse formation. We focused on another 130

small GTPase, TC10, shown to be a substrate for both prenylation and palmitoylation

(Michaelson et al., 2001; Watson et al., 2003) and play a role in the clustering of gephyrin

(Mayer et al., 2013). To determine whether TC10 can be palmitoylated by zDHHC9, we transfected HEK293T cells with either wildtype TC10 or palmitoylation-defective TC10 (TC10

C206, 209S) in the presence or absence of zDHHC9/GCP16. Using the Acyl-Rac assay to determine palmitoylation levels (Badrilla, Leeds, UK), we found that TC10 palmitoylation is significantly increased in the presence of zDHHC9/GCP16 compared to cells expressing TC10 alone (Figure 3.13a). There was no demonstrable palmitoylation of TC10C206, 209S, validating both the specificity of the Acyl-Rac assay as well as the previously identified sites of palmitate attachment for TC10 (Watson et al., 2003). Moreover, in hippocampal neurons, knockdown of zDHHC9 decreases while zDHHC9/ GCP16 overexpression increases the palmitoylation of endogenous TC10 in neurons (Figure 3.13b). We also validated TC10 shRNA in hippocampal neurons (Figure 3.13c). We next determined whether the palmitoylation of TC10 is involved in zDHHC-mediated promotion of dendrite outgrowth and synapse formation. Hippocampal neurons were transfected at 10 DIV with eGFP plus the indicated constructs and fixed and analyzed at 13 DIV. Dendrite length was not significantly altered following TC10 knockdown or overexpression, and TC10 did not impact zDHHC9’s effects on dendrite length (Figure 3.11d).

TC10 also did not significantly affect the density of PSD-95 puncta (Figure 3.13e). Neurons expressing TC10 shRNA exhibited significantly fewer gephyrin puncta compared to control cells, and cells expressing both zDHHC9 and TC10 shRNA exhibited a similar decrease in gephyrin puncta density (Figure 3.11e), suggesting that these two proteins function in the same pathway. The TC10 knockdown phenotype was rescued by co-expressing TC10R, but not TC10

C206,209SR, indicating that TC10 needs to be palmitoylated to promote inhibitory synapse 131

132

Figure 3.13 zDHHC9 Promotes Inhibitory Synapse Formation through the Palmitoylation of the Small GTPase, TC10 (a) Acyl-RAC assays from HEK293T cells transfected with the indicated constructs. The cleaved bound fraction (cBF) indicates that there is an increase in TC10 palmitoylation in cells co-transfected with zDHHC9+GCP16. The palmitoylation mutant, TC10 C206,209S is not palmitoylated demonstrating the specificity of the assay. The cleaved unbound fraction (cUF) represents levels of non-palmitoylated TC10 which is lower in cells co-transfected with zDHHC9+GCP16. The preserved bound fraction (pBF) represents the amount of non-specific binding to the resin and the preserved unbound fraction (pUF) represents the total input. (b) zDHHC9 knockdown decreases while zDHHC9+GCP16 expression increases the palmitoylation of TC10 in hippocampal neurons. (c) Validation of TC10 shRNA. The value at the bottom of the blot represents percent protein remaining in shRNA transfected lysate. N=3 separate blots from 3 separate cultures. (d) Knockdown or overexpression of TC10 has no discernable effects on dendritic growth. (e-f) TC10 knockdown decreases while TC10 overexpression increases the density of gephyrin puncta, with no effects on PSD-95 puncta. This regulation of inhibitory synapse density alters the ratio of excitatory:inhibitory synapses. Asterisks denote significant changes in PSD-95 density and hashtags denote significant in gephyrin density compared to control. N=40 per condition, >3 separate cultures. */# P<0.05, **/## P<0.01, ***/### P<0.001; one-way ANOVA, B:F=38.78, C:PSD-95 F=11.84, Gephyrin F=21.73, D: PSD-95 F=14.70, Gephyrin F=37.19, E:F=39.28, Tukey’s post-hoc; mean ± SEM. (g-h) FRAP analysis demonstrates an increase in mobility of TC10 following zDHHC9 knockdown. Points with error bars represent the mean ± SEM, and solid lines represent a single exponential fit. (h) Statistical tests compare the plateau values from exponential fits ± SEM. N=20 neurons, >3 cultures. P<0.001; one-way ANOVA, F:F=26.79, G:F=22.64, Tukey’s post-hoc; mean ± SEM.

formation. This is further supported by the fact that TC10R overexpression significantly increased the density of gephyrin puncta while overexpression of TC10 C206,209SR does not

(Figure 3.13e). All of these phenotypes were similar when comparisons were made to total synapses as opposed to synapse density, as TC10 had no effect on dendritic length (Figure 3.14).

Figure 3.14 Total TC10 Synapses

TC10 knockdown does not impact total number of excitatory synapses but significantly reduces the total number of inhibitory synapses. Asterisks denote significant changes in PSD-95 density and hashtags denote significant changes in Gephyrin density compared to control. N=40 neurons per condition, >3 cultures. */# P<0.05, **/## P<0.01, ***/### P<0.001; one-way ANOVA, PSD-95 F=43.14, Gephyrin F=73.00, Tukey’s post-hoc; mean ± SEM. 133

Together, this demonstrates that palmitoylation of TC10 is required for zDHHC9-mediated promotion of inhibitory synapse formation and the maintenance of the ratio between excitatory and inhibitory synapse density (Figure 3.13f).

The palmitoylation of TC10 is important for its localization at the plasma membrane, where it is believed to regulate gephyrin clustering (Mayer et al., 2013). To clearly demonstrate that TC10 localization to the membrane is dependent upon zDHHC9 we performed FRAP assays in the presence or absence of zDHHC9 (Figure 3.13g,h). Neurons were transfected with the indicated constructs and the mobility of TC10 within a ROI determined as shown previously

(Fig. 3d; Brigidi et al., 2014; 2015). We confirmed that zDHHC9 knockdown enhanced TC10 mobility (Figure 3.13g,h) and that this could not be rescued by zDHHS9 or the palmitoylation- defective human variants, R148W and P150S. Together, this demonstrates that zDHHC9- mediated palmitoylation of TC10 stabilizes TC10 at the membrane where it can act to promote gephyrin clustering and the formation of inhibitory synapses (Mayer et al., 2013).

3.3.6 zDHHC9 Knockout Mice Exhibit Enhanced Synaptic Excitability

Primary hippocampal cultures are an excellent system to investigate the cellular effects of knocking down zDHHC9 as well as elucidating downstream pathways involved. However, it is important to also determine the impact of zDHHC9 loss-of-function in vivo. Male C57BL/6J mice were bred with female heterozygous zDHHC9 knockout mice (MEM695N1) obtained from the Mutant Mouse Resource and Research Center (MMRRC; University of California, Davis) and zDHHC9 knockout males used for experimentation. zDHHC9 null mice exhibited a reduction in total Ras levels in hippocampus (Figure 3.15a,b), similar to that observed in the bone marrow and spleen of zDHHC9 knockout mice (Liu et al., 2014). In contrast, TC10 levels as well as those of other excitatory and inhibitory synaptic proteins known to be substrates for 134

Figure 3.15 zDHHC9 Knockout Mice Demonstrate Reduced Palmitoylation of Ras and TC10

(a,b) Western blots from zDHHC9 knockout mice against the indicated antibodies. zDHHC9 knockout mice have reduced levels of total Ras compared to controls, with no changes in other palmitoylated synaptic substrates (PSD-95, Gephyrin, and GABAγ2) or TC10 levels. (c,d) Acyl-RAC assays from zDHHC9 knockout mice using the indicated antibodies. The cleaved bound fractions (cBF) indicate that zDHHC9 knockout mice have reduced levels of Ras and TC10 palmitoylation compared to wild-type littermates with no changes in other synaptic palmitoylated substrates. WT n = 3 animals, zDHHC9 knockout = 3 animals.

135

palmitoylation, including PSD-95, Gephyrin, and the γ2 subunit of the GABA receptor, were unchanged (Fig. 3.15a,b). zDHHC9 null mice exhibited a significant reduction in the palmitoylation of both Ras and TC10 but with no reduction in the palmitoylation of PSD-95,

Gephyrin or GABARγ2, suggesting that the observed changes in synapse density are not due to decreased palmitoylation of these excitatory and inhibitory synapse proteins specifically (Figure

3.15c,d).

We next investigated whether the loss of zDHHC9 was associated with changes in the electrophysiological properties of CA1 neurons in acute hippocampal slices. In zDHHC9 knockout mice, whole cell current-clamp showed no significant differences in action potential firing threshold, membrane potential, input resistance, or tonic firing frequency (Figure 3.16), suggesting zDHHC9 knockout does not affect CA1 neuron active and passive properties.

We then performed whole cell voltage-clamp recordings in acute hippocampal slices in the absence of the Na+ channel blocker, tetrodotoxin (TTX), and observed a significant decrease in the interevent interval (i.e. an increase in frequency) and increase in amplitude of both sEPSCs and sIPSCs (Figure 3.17a-b, e-h). To distinguish action potential-dependent from action potential-independent release, we also recorded the mEPSCs and mIPSCs in the presence of

TTX. Similar to spontaneous activity, we observed a decrease in the interevent interval (i.e. an increase in frequency) and increase in amplitude of mEPSCs in zDHHC9 knockout CA1 neurons when compared to wild type (Figure 3.17c-d, i-l). Unexpectedly, we also observed an increase in both amplitude and frequency of mIPSCs in zDHHC9 knockout CA1 neurons, potentially indicative of increased interneuron activity. This has been previously documented as a mechanism for elevated excitability and termed “non-linear network effects” (Kapfer et al.,

2007), in which a reduction of dendritic inhibition makes cells more excitable leading to the 136

Figure 3.16 Intrinsic Action Potential Firing and Membrane Properties of CA1 Neurons in Wild-Type and zDHHC9 Knockout Mice (a) Representative whole-cell current clamp traces from a zDHHC9 CA1 neuron in response to 1s hyperpolarizing and depolarizing current injections. (b) Current-frequency analysis displaying mean action potential firing frequency over the 1-second recording period at each current injection. (c-f) Histograms displaying mean values for action potential firing threshold (c), current-frequency slope value for linear fit of individual cells’ firing frequency in response to current injection (D), resting membrane potential (e) and input resistance (f). WT: N = 3 animals, 10 cells; zDHHC9: N = 4 animals, 17 cells.

more effective recruitment of GABAergic feedback inhibition specifically mediated by soma- targeting interneurons (Miles et al., 1996; Jedlicka et al., 2009).

3.3.7 zDHHC9 Knockout Mice Display Seizure-Like Cortical Spike Activity

Mutations in zDHHC9 have been linked to Rolandic epilepsy, a non-convulsive epileptic syndrome most commonly observed in children (Baker et al., 2015). Electroencephalography

(EEG) recordings in Rolandic epilepsy patients reveal centro-temporal spikes originating from or near the central sulcus of the brain (Jung et al., 2003). zDHHC9 knockout mice and age matched

137

Figure 3.17 zDHHC9 Knockout Mice Exhibiti Increased Both Spontaneous and Miniature Synaptic Activity and Seizure-Like Events

(a,b) Representative whole-cell voltage-clamp recordings from CA1 neurons in acute brain slices for sEPSCs (a) and sIPSCs (b). (e,g) zDHHC9 KO neurons display a significant increase in frequency (decrease in interevent interval) for both (e) sEPSCs (WT = 4974±788 ms, Zdhhc9 KO = 452±41 ms; P=1 x 10-6 t-test) and (g) sIPSCs (WT = 397±33 ms, Zdhhc9 KO = 297±25 ms; P=0.015 t-test). (f,h) zDHHC9 KO neurons display a significant increase in peak amplitude of both (f) sEPSCs (WT = 23.7±1.3 pA, zDHHC9 KO = 42.5±0.6; P=3 x 10-7 t-test) and (h) sIPSCs (WT = 79.7±1.6 pA, zDHHC9 KO = 83.5±1.1; P=0.044 t-test). (c,d) Representative whole-cell voltage- clamp recordings from CA1 neurons in acute brain slices for mEPSCs (c) and mIPSCs (d). (i,k) zDHHC9 KO neurons display a signiﬁcant increase in frequency (decrease in interevent interval) for both (i) mEPSCs (WT = 2544.8±229.8 ms, zDHHC9 KO = 982.6±62.7 ms; P=0.0001 t-test) and (k) mIPSCs (WT = 144.1±3.1 ms, zDHHC9 KO = 121.4±2.0 ms; P=0.001 t-test). (j,l) zDHHC9 KO neurons display a signiﬁcant increase in peak amplitude of both (j) mEPSCs (WT = 17.8±1.0 pA, zDHHC9 KO = 21.1±0.6; P=0.02 t-test) and (l) mIPSCs (WT = 26.1±0.2 pA, zDHHC9 KO = 27.8±0.2; P=0.012 t-test). sEPSCs: WT n = 3 animals, 5 cells; zDHHC9 KO n = 4 animals, 8 cells; sIPSCs: WT n = 3 animals, 6 cells; zDHHC9 KO n = 4 animals, 7 cells. mEPSCs: WT n = 5 animals, 9 cells; zDHHC9 KO n = 4 animals, 8 cells; mIPSCs: WT n = 5 animals, 10 cells; zDHHC9 KO n = 4 animals, 10 cells. (m) Representative voltage trace from a visual cortex electrode in a zDHHC9 mouse. Lower trace shows expanded time scale from upper trace. (N-P) Analysis of seizure-like oscillatory spikes occurring in a two-hour recording period in WT (n = 3) and zDHHC9 KO (n = 4) mice for (N) total number (WT = 9, zDHHC9 KO = 188), (O) Mean duration (WT = 1.61±0.16 s, zDHHC9 = 3.97±0.6 s; P=0.047) and (P) spike cycle frequency. * P<0.05, **P<0.01, ***P<0.001, t-test. 138

littermate controls were evaluated to determine whether abnormal spiking existed as an in vivo correlate of increased synaptic excitability. Depth electrodes were placed in the right visual cortex, left dentate gyrus, left somatosensory (S1) cortex and right CA1 hippocampus to record local field potentials in P50-P70 mice. Periods of spontaneous abnormal spiking akin to seizure- like activity, was observed in all zDHHC9 KO mice (Figure 3.17m-p). This was observed in the right visual cortex in all 4 knockout mice and occasionally in the left dentate gyrus in one knockout mouse. The left S1 cortex and right CA1 hippocampus did not display spontaneous spiking. Over the 2 hour recording period, individual variability was observed with 3 of 4 mice displaying 10-15 seizure-like events and one mouse displaying 161 seizure-like events. While some low-level spiking was also observed in wild-type littermate controls, there were significantly fewer spike events and lower total spiking period duration (Figure 3.17m-p).

Together with brain slice recordings, this demonstrates that loss of zDHHC9 in vivo corresponds to an increase in synaptic excitability and an associated non-convulsive seizure-like activity.

3.4 Discussion

Here, we report that zDHHC9 regulates dendritic outgrowth and inhibitory synapse formation through the palmitoylation of two small GTPases, Ras and TC10, respectively. zDHHC9 mediated-palmitoylation of these GTPases increases their affinity for membranes, allowing for precise spatial and/or temporal regulation of downstream effectors. Indeed, we demonstrate that zDHHC9-mediated palmitoylation of Ras leads to decreased mobility, activation of the ERK signaling pathway, and promotion of dendritic growth. In contrast, decreased zDHHC9 activity results in a larger pool of mobile Ras and activation of the JNK pathway, leading to the retraction of dendritic processes. These findings are consistent with our observation that zDHHC9 localizes to Golgi satellites and outposts believed to be involved in the 139

coordination of dendritic branching and morphogenesis (Ye et al., 2007; Arthur et al., 2015; Ori-

McKenny et al., 2012). We also identify a new substrate for zDHHC9 palmitoylation, TC10, and demonstrate that zDHHC9-mediated palmitoylation of TC10 is essential for the formation/maintenance of inhibitory synapses. Moreover, we demonstrate that zDHHC9 knockout mice exhibit increased network excitability and related seizure-like activity. Together, these results demonstrate that zDHHC9 is essential for network formation by fine-tuning the dendritic arbor, regulating inhibitory synapse formation, and maintaining E:I synaptic balance.

Ras and TC10 share a high degree of similarity at the C-terminus hypervariable region, which contains targeting information to regulate subcellular localization (Michaelson et al.,

2001). Four distinct classes of palmitoylated proteins have been categorized, including a class farnesylated prior to their palmitoylation in the C-terminal region (Resh, 1996; Resh, 1999;

Salaun et al., 2010). The identification of additional substrates for zDHHC9 palmitoylation will determine whether this is a consensus sequence for recognition by zDHHC9, and potentially bioinformatically querying databases for similarly prenylated and palmitoylated proteins may reveal additional substrates for zDHHC9 palmitoylation.

Our finding that TC10 does not play a role in dendrite outgrowth initially appears contradictory to previous reports implicating TC10 in neurite outgrowth (Pommereit & Wouters,

2007; Fujita et al., 2013). However, these studies examined PC12 cell lines lacking TC10 at the time of differentiation, while the current study utilized primary hippocampal neurons transfected at 10 DIV with TC10 shRNA. As such, TC10 may play different roles in different cell types and/or stages of neuronal development.

The high degree of colocalization of zDHHC9 with both excitatory and inhibitory synapses places it in a unique position to respond to local changes in synaptic activity. Although 140

our work demonstrates that Ras palmitoylation is not a critical regulator of excitatory synapse density, it is possible that dynamic palmitoylation of Ras can impact the function of excitatory synaptic function, as differential downstream activators of Ras are known to regulate AMPA receptor content at synapses (Qin et al., 2005; Zhu et al., 2002; Hu et al., 2008). Furthermore, our electrophysiological data demonstrate that downregulation of zDHHC9 results in increased glutamatergic and GABAergic synaptic transmission.

Multiple studies have identified loss-of-function zDHHC9 variants in patients with X- linked intellectual disability (Raymond et al., 2007; Masurel-Paulet et al., 2014; Mitchell et al.,

2014; Tzschach et al., 2015; Han et al., 2017) and have further suggested that zDHHC9 loss of function is associated with an elevated risk of epilepsy (Baker et al., 2015). Interestingly, functional disruption of collybistin, a TC10 guanine nucleotide exchange factor, has also been associated with X-linked intellectual disability and epilepsy (Striano et al., 2017; Papadopoulous et al., 2015; Shimojima et al., 2011). Despite this, the underlying cellular and molecular mechanisms remain elusive. It is widely believed that ID is the result of alterations in dendritic growth and/or “synaptopathies”, as a large number of gene variants identified in ID patients encode for proteins involved in synapse structure and function (Valnegri et al., 2012). Indeed, over 50% of ID associated proteins are enriched at synaptic compartments (Ropers & Hamel,

2005) and disruptions in E:I balance are commonly observed in patients with epileptogenic and neurodevelopmental disorders (Ziburkus et al., 2013; Eichler & Meier, 2008). Interestingly, zDHHC9 is regulated by microRNA 134 (miRNA-134) in somatostatin-positive (SST+) interneurons (Chai et al., 2013), which make up approximately 30% of all inhibitory cells in the cortex and hippocampus (Rudy et al., 2011; Jinno & Kosaka, 2006). As impairments in interneuron function are believed to be a root cause for seizure activity by affecting E:I balance 141

(Marx et al., 2013; Liu et al., 2014), it is possible that additional effects corresponding to zDHHC9 dysregulation in cortical interneurons also contribute to the etiology of intellectual disability and epilepsy.

Our work raises the possibility that the abnormal dendritic and synaptic phenotypes resulting from disrupted palmitoylation by zDHHC9 contributes to the pathogenesis of intellectual disability and epilepsy observed in humans. However, the changes recorded in hippocampal pyramidal neurons in acute slice may not be related to the seizure activity recorded in vivo. Nonetheless, these findings enhance our understanding of zDHHC9 function and its interaction with specific substrates (Figure 3.18) and may provide a new approach to therapeutic intervention for patients with zDHHC9 mutations.

Figure 3.18 Model for zDHHC9 Function

Loss of zDHHC9 function leads to significantly smaller and less complex dendritic arbors, with the same number of excitatory synapses (more densely packed) and a significant reduction in inhibitory synapses, ultimately leading to a shift in the E:I balance. zDHHC9 mediates dendrite outgrowth through the palmitoylation of Ras, and inhibitory synapse formation and/or maintenance through palmitoylation of TC10. 142

Chapter 4: Conclusion

The work presented in this dissertation provides many new insights in the role of palmitoylation enzymes and their substrates in neural circuit formation and neurodevelopmental disorders. Specifically, I demonstrate how the palmitoylation enzymes, zDHHC15 and zDHHC9, contribute to the etiology of intellectual disability through differential mechanisms that converge on changes in dendritic growth and synaptic (excitatory:inhibitory) balance. Our findings from

Chapter 2 show that loss of zDHHC15 selectively impairs the trafficking of PSD-95 and the maturation of dendritic spines, revealing a key role in excitatory synapse formation and/or maintenance. While this is true in vitro, the consequences of loss of zDHHC15 in vivo are yet to be determined. Loss of zDHHC15 also resulted in impaired growth of 10 DIV hippocampal neurons, suggesting that zDHHC15 also serves to regulate dendritic development through an as of yet identified substrate. As circuit formation is dependent on both dendritogenesis and synaptogenesis, further research into the associated signaling pathways and substrates of zDHHC15 is clearly warranted. Another intriguing observation from this study was the finding that zDHHC15 knockdown can be rescued by overexpression of another zDHHC protein, zDHHC3. This may be a form of redundancy, as it has been demonstrated that palmitoylation deficient PSD-95 is diffusely distributed in dendrites and the cell body (Topinka & Bredt, 1998), and previous research has demonstrated that palmitoylation is required for axon exclusion of

PSD-95 (El-Husseini et al., 2001). However, endogenous zDHHC3 does not fully support PSD-

95 palmitoylation when zDHHC15 is depleted. This, coupled with the finding that zDHHC3 overexpression leads to mislocalization of the enzyme, suggests that these enzymes normally inhabit specific subcellular locations, and perhaps function independently in regulating PSD-95

143

palmitoylation. This is further supported by a study that found zDHHC2 to be responsible for local palmitate cycling at the synapse in response to activity (Noritake et al., 2009).

These data suggest that the primary factor that determines zDHHC substrate interactions appears to be the subcellular localization of the zDHHC in question, and the extent to which substrates are trafficked through that particular compartment. These interactions can in turn determine the distribution of neuronal palmitoylated proteins. Golgi-localized proteins, such as zDHHC15 and zDHHC9, help facilitate protein trafficking and membrane targeting, while once a palmitoylated protein reaches its destination (such as the synapse), PATs such as zDHHC2 and zDHHC5, which are more mobile, maintain the equilibrium of palmitoylated proteins (Noritake et al., 2009; Brigidi et al., 2014).

While the effects of zDHHC15 palmitoylation on synaptic function are primarily manifested at excitatory synapses, the findings from Chapter 3 demonstrate that zDHHC9 predominantly regulates inhibitory synapses. Furthermore, we discovered that zDHHC9 palmitoylation of Ras regulated dendritic growth through an ERK/JNK phosphorylation pathway, and that zDHHC9 palmitoylation of a novel substrate, TC10, regulated inhibitory synapse formation and/or maintenance. It is possible that these two processes are synergistically regulated, as dendritic GABAergic synapse number increases in parallel with neuronal growth, such that GABAergic synapse density remains relatively constant, but the total number of synapses increases as arborization extends (Swanwick et al., 2004; Swanwick et al., 2006). As the ratio of excitation and inhibition must be carefully regulated, further investigation into the mechanisms of palmitoylation and its regulation to maintain synaptic homeostasis, the compensatory alterations that prevent aberrant signaling and restore circuit set points (Turrigiano

& Nelson 2004; Davis, 2006) are needed. 144

Together, these studies demonstrate how changes in the balance between excitation and inhibition at the neuronal level are important for proper brain function and disruption to this process may be involved in the etiology of intellectual disability.

4.1 The Role of Palmitoylation in Excitatory and Inhibitory Synapse Balance

It has been proposed that changes in the relative strength of excitation and inhibition underlies many diseases and disorders of the nervous system (reviewed in Sohal & Rubenstein,

2019). However, typical views have taken a unidimensional approach, focusing just on levels of excitation and inhibition. More recently, E:I balance has been approached with a more multidimensional view, acknowledging the contribution of various mechanisms that impact developmental, maintenance, and/or plasticity-related processes (Figure 4.1) (Sohal &

Rubenstien, 2019). Both excitatory and inhibitory synapse function depends on the precise alignment of pre- and post-synaptic specializations, as well as the regulated recruitment of proteins to synaptic sites, and this is dependent on gene expression and protein translation as well as regulated trafficking. Palmitoylation, due its dynamic, reversible nature, is a strong candidate for regulating the recruitment, localization, and maintenance of synaptic proteins, contributing to signaling pathways that enable precise control over cell function.

The majority of previous research has focused on the role of palmitoylation in synaptic plasticity, a process through which neuronal activity modulates protein composition at the synapse leading to alterations in the number of synapses or synaptic strength (Kang et al., 2008;

Neves et al., 2008; Huganir & Nicoll, 2013). From this work, a number of synaptic proteins that are substrates for palmitoylation have been demonstrated to contribute to the dynamic and functional modulation of synapses, including PSD-95, Gephyrin, CDC42, and receptor (NMDA,

AMPA, GABA) subunits (Noritake et al., 2009; Yokoi et al., 2012; Brigidi et al., 2014; Fukata et 145

Figure 4.1 Unidimensional vs Multidimensional Views of E:I Balance

In a traditional, unideminsional framework (top), E:I balance refers to the overall level of circuit activity. Dysfunction in inhibition or excitation causes a shift at the overall level of circuit activity, altering the signal to noise ratio. As an example, excessive “noise” may render the circuits less responsive to signals. A reduced signal to noise ratio may reduce the strength of signals, perhaps due to excessive inhibition. In the updated, multidimensional view, E:I balance is conceptualized in a multidimensional space. Different mechansisms affect circuits at various levels to ultimately contribute to pathophysiology. In this example, abnormal development due to altered gene expression or protein translation leads to altered or dysregulated compensatory or homeostatic processes, which causes the system to settle on the “new” altered pathological state. Reproduced with permission from Sohal & Rubenstein, 2019.

146

al., 2013; Itoh et al., 2018; Hayashi et al., 2005; Thomas et al., 2012, 2013; Thomas & Hayashi,

2013, Dejanovic et al., 2014). While the majority of investigations into synaptic plasticity and excitability have focused on post-synaptic excitatory synapses, it is becoming increasingly clear that regulation of pre-synaptic function and inhibitory synapses are critically important to maintain proper neural function. For instance, GABAergic synaptic plasticity can modify presynaptic GABA release or the number, sensitivity, or responsiveness of the postsynaptic GABA receptors, leading to changes in excitability and neuronal circuit function (Castillo et al., 2011;

McBain & Kauer, 2009; Gaiarsa et al., 2002). The clustering and density of both PSD-95 and gephyrin scaffolding complexes can be modified in response to activity, and so the plasticity of a few strategically located inhibitory synapses have important consequences for determining neuronal output (Gidon & Segev, 2012). Additionally, inhibitory interneuron subtypes display different E:I ratios, as parvalbumin, calbindin D and calretinin interneurons have E:I ratios of

14:1, 3:1, and 2:1 respectively (Gulyas et al., 1999). This demonstrates that each cell type may have different requirements for excitation and inhibition, and therefore individual neurons must have mechanisms to correct their excitatory and inhibitory synapses in response to alterations in synapse strength and/or number.

Most studies to date have demonstrated the role of palmitoylation enzymes in regulating either excitatory or inhibitory synapses individually, while few have examined how palmitoylation affects the delicate balance between these two synaptic subtypes, or alter developmental processes (such as Ras-MAPK signaling) in neurons. Indeed, Ras and downstream pathways regulate survival, growth, and differentiation, with dysregulation of Ras or major effector pathways causing severe nervous system dysfunction and dysregulation (reviewed in Zhong, 2017). Palmitoylation, as a dynamic process, can regulate, coordinate, and fine-tune 147

neuronal signaling pathways both temporally (in response to activity) or spatially (specific subcellular localizations). As such, changes in palmitoylation can lead to an array of deficits that are could include the selective loss of excitatory or inhibitory synapses, deficits in compensatory or homeostatic mechanisms, and/or alterations in gene function or protein translation culminating in E:I imbalance and pathology (see Figure 4.1).

4.2 Palmitoylation in Intellectual Disability

In Chapters 2 and 3, this dissertation identifies some plausible mechanisms for how zDHHC15 and zDHHC9 may contribute to the phenotype of intellectual disability by affecting

PSD-95 trafficking and excitatory synapse formation or Ras/TC10 palmitoylation and dendrite growth/inhibitory synapse formation, respectively. Intellectual disability is a neurodevelopmental disorder characterized by delays in cognitive development due to the disruption of processes involved in brain development and/or function (Raymond et al., 2007). The proper regulation of neural development and circuit formation is essential for normal neuronal function (McAllister,

2000) and defects in both dendritic growth and synaptic function have been ascribed to intellectual disability (Purpura et al., 1975; Kaufman & Moser, 2000; Humeau et al., 2009).

Indeed, more than 50% of ID-related genes appear to be enriched in the pre- and/or post-synaptic compartments (Ropers & Hamel, 2005) and the majority of these are known to be involved in processes affecting synapse formation, regulation of spine morphology, or dendritic growth

(Verpelli et al., 2013). Even mild alterations in these genes can lead to drastic changes in synapse morphology and function.

Determining the developmental and functional changes in neural circuitry as a result of these changes is critical to understand the molecular determinants of intellectual disability.

Current practices screen 114 genes for XLID in patients (Greenwood Genetic Center, 148

Figure 4.2 Genes Screened for in X-Linked Intellectual Disability (XLID)

(a) The top GO terms enriched in the candidate genes for XLID. The majority of genes are involved in synaptic or dendritic processes, providing further evidence that these are the most common deficits observed in ID. (b) The percentage of XLID genes known, predicted, and unknown to be palmitoylated. Palmitoylated proteins are significantly enriched in the candidate genes. Palmitoylation prediction was performed by Swisspalm.

Greenwood, SC), and these genes are predominantly associated with synapses and dendrites. Of these, 27 encode known palmitoylated proteins, and an additional 52 of these genes are predicted

149

to be palmitoylated by Swisspalm (Blanc et al., 2015) (Figure 4.2). This means that 69% of the candidates for XLID may be palmitoylated, far exceeding the expected value as only ~10% of the proteome is estimated to be palmitoylated (Saunders et al., 2015). While many of these proteins are likely palmitoylated by different PATs, these observations indicate that palmitoylation is important for neural function and cognition. It is also important to remember that deficits in depalmitoylation may be equally pathogenic, contributing to dysfunction in cellular signaling. In order to target palmitoylation for therapeutic use, we must further understand the complicated cascades involved in palmitoylation and depalmitoylation regulation.

4.3 Potential for Palmitoylation as a Therapeutic Target

In this dissertation, I have focused primarily on the association between zDHHC15 and zDHHC9 with intellectual disability. However, mutations and changes in expression of the zDHHC family are associated with numerous other neurodevelopmental and neurological disorders and diseases, as well as other disorders such as cancers and infectious diseases (Resh,

2017; Chavda et al., 2014; Saunders et al., 2015). Although limited information exists on zDHHC15, zDHHC9 is implicated in colorectal/gastric cancer (Mansilla et al., 2007) and leukemia (Liu et al., 2014), while Ras as an oncogene is perhaps one of the most studied proteins of the last several decades. Interestingly, oncogenic H-Ras appears to undergo an increased rate of palmitate turnover when compared to wildtype H-Ras (Baker et al., 2003). As elevated levels of zDHHC9 may be oncogenic, and Ras oncogenes are known to have altered activity, a detailed understanding of the effects of zDHHC9/Ras is necessary for the development of therapeutics.

For example, both activators and inhibitors could be potentially useful or harmful depending on cell type and normal levels of expression, as both upregulation and downregulation of zDHHC9 and Ras are associated with disease, and this is likely the case for other substrate-enzyme pairs. 150

Furthermore, due to the overlap and redundancy of PAT function, any clinical modulators will need to be highly specific and titrated to the specific levels required for cell function. Gene therapies may be the most useful in this regard, but whether or not these gene therapies can be efficacious after development remains to be seen.

4.4 Epilepsy

Epilepsy is most commonly understood to be a result of disruptions in the normal balance between excitation and inhibition, with periods of asynchrony generating abnormal activity

(Fritschy, 2008). Other common neurodevelopmental disorders that are also believed to result from E:I imbalance, including autism and ID, are often comorbid with epilepsy (Schoenfeld et al., 1999; Camfield & Camfield, 2002), although teasing apart the mechanism has proven difficult as separating the effects of seizures from pre-existing pathology is difficult. Indeed, seizures themselves are known to affect neuronal excitability and neuronal growth, and the idea that “seizures beget seizures” has been reinforced since the late 1800’s (Gowers, 1888). In

Chapter 3, we showed that loss of zDHHC9 led to decreased inhibition in vitro, which was observed as seizure-like activity in vivo. Recent findings have supported a role for palmitoylation in seizures and epilepsy, as mice expressing a palmitoylation deficient form of GluA1 demonstrated increased seizure susceptibility due to increased hyperexcitability (Itoh et al.,

2018). As an interesting aside, expression of proteins that promote inhibitory synapse formation, such as Semaphorin 4D, have been hypothesized to suppress seizures in vivo (Acker et al., 2018).

This taken with the comorbid data suggests that perhaps the comorbidities are the result of hyperexcitability, while the hypoexcitable alterations in gene function do not contribute to epileptic or seizure phenotypes. This is further evidenced by the idea that mutations in zDHHC9 or collybistin, a TC10 activator, present with higher than normal incidence of epilepsy/seizures 151

by creating a reduction in inhibition leading to elevated excitability (Baker et al., 2015;

Papadopoulus et al., 2015; Shimojima et al., 2011; Kalscheuer et al., 2009).

Palmitoylation itself has also been strongly implicated in the seizures, as kainite-induced seizures lead to the increased palmitoylation of a number of synaptic proteins, including PSD-95 and AKAP79/150 (Kang et al., 2008; Keith et al., 2012). Additionally, deficiency of AMPAR- palmitoylation increases the susceptibility to seizures (Itoh et al., 2018). Interestingly, the anti- seizure medication valproic acid is also known to inhibit the palmitoylation of several synaptic proteins (Kay et al., 2015). Together, this demonstrates that targeting protein palmitoylation in the generation of and treatment of seizures warrants further investigation.

Our findings emphasize the importance of studying the function of palmitoylation enzymes and palmitoylated proteins in the maintenance and alteration of the excitatory:inhibitory ratio in multiple brain regions and cell types, to better understand how these proteins contribute to diverse pathologies. Although the well-studied circuitry of the hippocampus is an excellent model for molecular manipulations in vitro, the same molecular machinery and cell biology cannot be extrapolated to all brain regions and cell types.

4.5 zDHHC9 and Ras Signaling in Neurons

Although Ras was not the main focus of this dissertation, the findings in Chapter 3 provide useful insight and prompt a discussion of Ras and the relationship between zDHHC9 and

Ras signaling in neurons. Chapter 3 demonstrates that loss of zDHHC9 causes mislocalization and redistribution of Ras from its normal membrane localization in vitro, resulting in differential downstream signaling believed to be mediated by Ras subcellular localization. Furthermore, we saw that zDHHC9 knockout results in a similar reduction in Ras palmitoylation/membrane localization, and surprisingly also causes a reduction in Ras protein levels in vivo. This may be 152

the due to a potential interplay between palmitoylation and ubiquitination/phosphorylation, which has been shown in other systems to alter the equilibrium of protein stability (Blaskovic et al., 2013; Swaney et al., 2013).

Ras signaling affects numerous and diverse cellular processes both during development and after maturation, including survival, growth, differentiation, intracellular transport, gene expression, and LTP (Zhong, 2016). It should come as no surprise then that alterations in Ras signaling have profound effects on the nervous system, and indeed an entire pathological categorization is defined around deficits in various proteins linked to Ras signaling, collectively termed “RASopathies” (Rauen, 2013). However, Ras signaling has far reaching and unknown consequences on numerous other cellular pathways, including Rho, mTOR, and CDC42 signaling (Chew et al., 2014; Mendoza et al., 2011; Wang et al., 2005), which themselves also have numerous functions and pathways. Extensive research has demonstrated that Ras signals differently based on the interaction with GEFs/GAPs and subcellular localization (Plowman et al., 2005; Hancock, 2003; Prior & Hancock, 2012; Zhang et al., 2016; Roose et al., 2005; Chiu et al., 2002; Freedman et al., 2006; Chai et al., 2013). While this work has demonstrated that prenylation of Ras is enough to sort it to the Golgi, where further palmitoylation targets it to the plasma membrane, this dissertation extends this work by demonstrating that this occurs in hippocampal neurons and that this process can be, at least in part, regulated by zDHHC9. Ras has also been linked to presynaptic vesicle release and plasticity (Kushner et al., 2005; Arendt et al., 2004; Seeger et al., 2004) and axonal growth (Fivaz et al., 2008), although whether zDHHC9 plays a role in these processes remains to be seen and is a possible avenue for future experiments.

153

4.6 Significance, Strengths, and Limitations

The work presented in this dissertation provides a plausible mechanism for how loss of the palmitoylation enzymes, zDHHC15 or zDHHC9, may contribute to the pathology and manifestation of XLID and/or epilepsy. Furthermore, it provides further advancement to the field of palmitoylation with regards to nervous system function, specifically dendritic growth and maintenance, and synapse maturation and maintenance. In Chapter 2 it was determined how loss of zDHHC15 impaired dendritic growth, impaired spine maturation, and inhibited the trafficking of PSD-95. In Chapter 3 it was determined that loss of zDHHC9 promoted active retraction of the dendritic arbor, and that zDHHC9 mediated dendritic growth through Ras palmitoylation and activation or suppression of downstream kinases, namely ERK and JNK. It was then demonstrated that loss of zDHHC9 affected the palmitoylation of a newly identified substrate,

TC10, and that this impaired inhibitory synapse formation and/or maintenance. Additionally, two of the known clinical mutations of zDHHC9 were shown to work as loss-of-function, demonstrating the importance of zDHHC9 to nominal cellular function. Together, this dissertation has demonstrated how loss of protein palmitoylation can adversely affect protein trafficking and protein function and how this may contribute to cellular dysfunction that can lead to cognitive deficits as a result of a shift in the balance between excitation and inhibition.

A strength of the work described in this dissertation is that we were able to investigate the cellular mechanisms of neurobiology in neurons without the influence of other major cell types through the use of dissociated hippocampal cultures, which contain relatively few interneurons.

The culture system also allows for cell autonomous investigation of genetic manipulations.

Additionally, we are able to follow up our in vitro investigation with an examination in vivo,

154

which provided compelling evidence for our in vitro findings through electrophysiological methods.

A limitation of our culture system is that we are using relatively immature (DIV 13 cultured from E18 rats) hippocampal neurons to investigate dendritic spine/synapse development. While this stage is appropriate for studies of dendritic development and arborization, it may be early for the elucidation of effects on spine development and maturation.

It is possible that some of the deficits observed may normalize at later stages of development, or possible that the deficts are further compounded. Determining whether these processes are reflective of a developmentally delay or a permanent alteration is an interesting avenue for future study.

One caveat of using dissociated hippocampal cultures and a limitation of this study is that the findings presented here may not be completely representative of all aspects of dendritic growth, synapse function, or cellular interactions that occur in vivo. While loss of zDHHC9 reduced inhibitory synapse formation and/or maintenance in vitro in hippocampal cells, the seizure phenotype we report was detected only in the cortex in vivo. Further characterization of the zDHHC9 knockout animals will determine if this is the result of network interactions or a cortical specific mechanism. Utilizing techniques such as AAV-GFP stereotactic injections in the hippocampus of WT vs zDHHC9 knockout animals, or breeding of zDHHC9 knockout and

Thy1-GFP mice, in conjunction with imaging modalities such as CLARITY could help reconcile the effects of zDHHC9 knockout on dendritic and synaptic properties in vivo.

Another limitation is that we have not fully investigated or characterized the zDHHC enzyme-substrate pairs, and any observed effects could be secondary to the impaired palmitoylation of a number of different proteins, or deficits in “palmitoylation cascades” 155

(Abrami et al., 2017) as the palmitoylation (and depalmitoylation enzymes) are themselves palmitoylated. By comparing the knockout animals with wildtype littermates and utilizing mass spectrometry, a complete list of putative substrates could be generated to better understand the function of zDHHC15 or zDHHC9. For instance, other studies have demonstrated the role of zDHHC PATs in dendritic growth and spine density, notably zDHHC8 (Mukai et al., 2008). The similar phenotype of this research (reduced mature spines and reduced total number of dendrites) suggests that perhaps there may be common substrates for zDHHC8 and zDHHC15 that regulate these processes. A potential candidate for follow-up studies includes the stathmins, as zDHHC15 has been shown to be a PAT for stathmin 2 and 3 (Levy et al., 2011), while loss of zDHHC15 inhibits lysosomal sorting. Lysosomal impairments have been demonstrated to decrease dendritic spine number, specifically excitatory synapses (Goo et al., 2017) and to regulate the structural plasticity of spines (Padamsey et al., 2017). When considering our experiments and results, we must always be cognizant of the possibility that depletion of a particular palmitoylation enzyme may be affecting diverse cellular processes and numerous mechanisms may be at work to generate the observed phenotype.

In the case of zDHHC9, both confirmed substrates for zDHHC9 palmitoylation (Ras and

TC10) have been linked with ID and epilepsy (Papadopoulous et al., 2015; Shimojima et al.,

2011; Kalscheuer et al., 2009; Baker et al., 2015; Raymond et al., 2007; Mitchell et al., 2014;

Masurel-Paulet et al., 2014; Han et al., 2017; Tzschach et al., 2017; Rauen, 2013; Jindal et al.,

2015), suggesting that this enzyme and its substrates are critically involved in normal brain function and plasticity. It remains a possibility, and is indeed quite likely, that these proteins play a role in the regulation and function of additional cells in the nervous system beyond neurons.

Indeed, zDHHC9 is the most highly expressed palmitoylation enzyme in oligodendrocytes and is 156

expressed in mature oligodendrocytes at a level nearly 5 fold of its expression in neurons, while zDHHC15 has its highest expression in oligodendrocyte precursor cells (OPCs) (Figure 4.3)

(Zhang et al., 2014). Deficits in myelin have been observed to cause a number of psychiatric and nervous system disorders, including intellectual disability (Fields, 2008). We cannot rule out the possibility that deficits in myelin are also contributing to the observed phenotypes in vivo, and indeed the zDHHC9 knockout mice display a 36% reduction in corpus callosum volume revealed by MRI (Kouskou et al., 2018). Further investigations stemming from this work will be beneficial to better understand how cognitive functions such as learning and memory are impaired when zDHHC9 is lost.

Figure 4.3 Relative Expression of ZDHHC15 and ZDHHC9 from Ben-Barres RNAseq

Both ZDHHC15 and ZDHHC9 have expression profiles indicating their involvement in OPCs and mature oligodendrocytes.

4.7 Final Remarks

The field of palmitoylation is still in its nascent stages, with the first enzymes having only been discovered 20 years ago (Putilina et al., 1999) and the first crystal structure being solved very recently in early 2018 (Rana et al., 2018). Advancements in technology and techniques have pushed the field forward, allowing us to perform large-scale proteomic assays to determine 157

enzyme-substrate pairs. Despite this, our knowledge of the reverse reaction, depalmitoylation, is

(at present) very poorly understood. In order to truly understand the role of palmitoylation in the nervous system, the full catalog of depalmitoylation enzymes and palmitoylation- depalmitoylation cycles will allow a more thorough understanding of the mechanisms through which palmitoylation regulates nervous system function. It also remains an open question about how the palmitoylation/depalmitoylation enzymes are regulated, as other post-translational modification enzymes (such as kinases) can be regulated by numerous factors, including other post-translational modifications. Investigation into the regulation of palmitoylation and depalmitoylation enzymes by their own post-translational modifications, such as phosphorylation

(e.g. Moritz et al., 2015; Lievens et al., 2016), palmitoylation cascades (e.g. Abrami et al., 2017) and accessory proteins (e.g. Swarthout et al., 2005; Woodley et al., 2018) will help us parse apart the intricacies of this important biological process.

158

Bibliography

Abe, T., Kato, M., Miki, H., Takenawa, T., & Endo, T. (2003). Small GTPase TC10 and its homologue RhoT induce N-WASP-mediated long process formation and neurite outgrowth. Journal of Cell Science, 116, 155-168.

Ahearn, I.M., Tsai, F.D., Court, H., Zhou, M., Jennings, B.C., Ahmed, M., Fehrenbacher, N., Linder, M.E., and Philips, M.R. (2011). FKBP12 Binds to Acylated H-Ras and Promotes Depalmitoylation. Molecular Cell, 41, 173–185.

Alabi, A.A., & Tsien, R.W. (2012). Synaptic vesicle pools and dynamics. Cold Spring Harb Perspect Biol, 4, a013680.

Anderson, S.S. & Bi, G.Q. (2000). Axon formation: a molecular model for the generation of neuronal polarity. Bioessays, 22, 172-179.

Aran, V., and Prior, I.A. (2013). Compartmentalized Ras signaling differentially contributes to phenotypic outputs. Cell Signal, 25(9), 1748-1753.

Aridor, M., Guzik, A.K., Bielli, A. and Fish, K.N. (2004) Endoplasmic reticulum export site formation and function in dendrites. J. Neurosci. 24, 3770–3776.

Arstikaitis, P., Gauthier-Campbell, C., Carolina Gutierrez Herrera, R., Huang, K., Levinson, J.N., Murphy, T.H., Kilimann, M.W., Sala, C., Colicos, M.A., and El-Husseini, A. (2008). Paralemmin-1, a modulator of filopodia induction is required for spine maturation. Mol. Biol. Cell, 19, 2026–2038.

Arthur, A.L., Yang, S.Z., Abellaneda, A.M., and Wildonger, J. (2015). Dendrite arborization requires the dynein cofactor NudE. J Cell Sci., 128(11), 2191-2201.

Baker, K., Astle, D.E., Scerif, G., Barnes, J., Smith, J., Moffat, G., Gillard, J., Baldeweg, T., and Raymond, F.L. (2015). Epilepsy, cognitive deficits and neuroanatomy in males with ZDHHC9 mutations. Ann Clin Transl Neurol. 2(5), 559-569.

Bamji, S.X. (2005). Cadherins: Actin with the cytoskeleton to form synapses. Neuron, 47, 175- 178.

Banker, G.A. & Cowan, W.M. (1977). Rat hippocampal neurons in dispersed cell culture. Brain Research, 126, 397-342.

Bar-Sagi, D. & Feramisco, J.R. (1985). Microinjection of the ras oncogene protein into PC12 cells induces morphological differentiation. Cell, 42, 841-848.

Bartels, D.J., Mitchell, D.A., Dong, X., & Deschenes, R.J. (1999). Erf2, a novel gene product that affects the localization and palmitoylation of Ras2 in Saccharomyces cerevisiae. Mol. Cell. 159

Biol., 19, 6775-6787.

Bartlett, W.P. & Banker, G.A. (1984). An electron microscopic study of the development of axons and dendrites by hippocampal neurons in culture. I. Cells which develop without intercellular contacts. J Neurosci, 4, 1944-1953.

Bartlett, W.P. & Banker, G.A. (1984). An electron microscopic study of the development of axons and dendrites by hippocampal neurons in culture. II. Synaptic relationships. J Neurosci, 4, 1954-1965.

Bhattacharyya, R., Barren, C., & Kovacs, D.M. (2013). Palmitoylation of amyloid precursor protein regulates amyloidogenic processing in lipid rafts. J. Neurosci., 33, 11169-11183.

Bivona, T.G. and Philips, M.R. (2003). Ras pathway signaling on endomembranes. Curr Opin Cell Biol., 15(2), 136-142.

Björkblom, B., Ostman, N., Hongisto, V., Komarovski, V., Filén, J.J., Nyman, T.A., Kallunki, T., Courtney, M.J. and Coffey, E.T. (2005). Constitutively active cytpolasmic c-Jun N-terminal kinase 1 is a dominant regulator of dendritic architecture: role of microtubule-associated protein 2 as an effector. Neurosci., 25(27), 6350-6361.

Blanc, M., David, F., Abrami, L., Miglozzi, D., Armand, F., Burgi, J., & van der Goot, F.G. (2015). SwissPalm: Protein palmitoylation database. F1000Res, 4, 261. (eCollection).

Borgese, N., Francolini, M., & Snapp, E. (2006). Endoplasmic reticulum architecture: structures in flux. Curr Opin Cell Biol, 18, 358-364.

Bosmans, F., Milescu, M., and Swartz, K.J. (2011). Palmitoylation influences the function and pharmacology of sodium channels. Proc Natl Acad Sci U.S.A., 108(50), 20213-20218.

Bourne, H.R., Sanders, D.A., & McCormick, F. (1990). The GTPase superfamily: a conserved switch for diverse cell functions. Nature, 348, 125-132.

Bowen, A.B., Bourke, A.M., Hiester, B.G., Hanus, C., and Kennedy, M.J. (2017). Golgi- independent secretory trafficking through recycling endosomes in neuronal dendrites and spines. eLife, 6, e27362.

Bozdagi O, Valcin M, Poskanzer K, Tanaka H, & Benson DL (2004). Temporally distinct demands for classic cadherins in synapse formation and maturation. Mol Cell Neurosci. . 27, 509–521. Brambilla, R., Gnesutta, N., Minichiello, L., White, G., Roylance, A.J., Herron, C.E., Ramsey, M., Wolfer, D.P., Cestari, V., Rossi-Arnaud, C., Grant, S.G., Chapman, P.F., Lipp, H.P., Sturani, E., & Klein, R. (1997). A role for the Ras signalling pathway in synaptic transmission and long-term memory. Nature, 390, 281-286.

160

Brewer GJ. Isolation and culture of adult rat hippocampal neurons. (1997). J. Neurosci. Methods., 71, 143–155.

Brigidi, G.S., Santyr, B., Shimell, J., Jovellar, B., and Bamji, S.X. (2015). Activity-regulated trafficking of the palmitoyl-acyl transferase DHHC5. Nat Commun, 6, 8200.

Brigidi, G.S., Sun,Y., Beccano-Kelley, D., Pitman, K., Mobasser, M., Borgland, S.L., Milnerwood, A.J., and Bamji, S.X. (2014). Palmitoylation of δ-catenin by DHHC5 mediates activity-induced synapse plasticity.Nat Neurosci., 17(4), 522-532.

Brose, N. (2013). Why we need more synaptogenic cell-adhesion prtoeins. PNAS, 110, 3717- 3718.

Brown, D.A. (2006). Lipid rafts, detergent-resistant membranes, and raft targeting signals. Physiology, 21, 430-439.

Bu, D.F., Erlander, M.G., Hitz, B.C., Tillakaratne, N.J., Kaufman, D.L., Wagner-McPherson, C.B., Evans, G.A. & Tobin, A.J. (1992). Two human glutamate decarboxylases, 65-kDa GAD and 67-kDa GAD, are each encoded by a single gene. PNAS, 89, 2115-2119.

Camp, L.A. & Hoffman, S.L. (1993). Purification and properties of a palmitoyl-protein thioesterase that cleaves palmitate from H-Ras. J. Biol. Chem., 268, 22566-22574.

Casanova, J.R., Nishimura, M., Owens, J.W., and Swann, J.W. (2012). Impact of seizures on developing dendrites: implications for intellectual developmental disabilities. Epilepsia, 53 Suppl 1, 116-124.

Chai, S., Cambronne, X.A., Eichhon, S.W., and Goodman, R.H. (2013). MicroRNA-134 activity in somatostatin interneurons regulates H-Ras localization by repressing the palmitoylation enzyme, DHHC9. Proc Natl Acad Sci U.S.A., 110(44), 17898-17903.

Chamberlain, L.H. & Shipston, M.J. (2015). The physiology of protein S-acylation. Physiol Rev, 95, 341-376.

Chapman, E.R., Blasi, J., An, S., Brose, N., Johnston, P.A., Sudhof, T.C., & Jahn, R. (1996). Fatty acylation of synaptotagmin in PC12 cells and synaptosomes. Biochem Biophys Res Comm, 225, 326-332.

Charollais, J., and Van Der Goot, F.G. (2009). Palmitoylation of membrane proteins (Review). Mol Membr Biol, 26, 55-66.

Charron G., Wilson J., & Hang H. C. (2009a). Chemical tools for understanding protein lipidation in eukaryotes. Curr. Opin. Chem. Biol., 13, 382–391.

161

Charron G., Zhang M. M., Yount J. S., Wilson J., Raghavan A. S., Shamir E., et al. (2009b). Robust fluorescent detection of protein fatty-acylation with chemical reporters. J. Am. Chem. Soc., 131, 4967–4975.

Chen, H.J., Rojas-Soto, M., Oguni, A., & Kennedy, M.B. (1998). A synaptic Ras-GTPase activating protein (p135 SynGAP) inhibited by CaM kinase II. Neuron, 20, 895-904.

Chen, S.X. & Haas, K. (2011). Function directs form of neuronal architecture. Bioarchitecture, 1, 2-4.

Cheng, D., Hoogenraad, C.C., Rush, J., Ramm, E., Schlager, M.A., Duong, D.M., Xu, P., Wijayawardana, S.R., Hanfelt, J., Nakagawa, T., Sheng, M., & Peng, J. (2006). Relative an absolute quantification of postsynaptic density proteome isolated from rat forebrain and cerebellum. Mol. Cell. Proteomics, 5, 1158-1170.

Chiu, V.K., Bivona, T., Hach, A., Sajous, J.B., Silletti, J., Wiener, H., Johnson, R.L., Cox, A.D. and Philips, M.R. (2002). Ras signaling on the endoplasmic reticulum and the Golgi. Nat Cell Biol. 4(5), 343-350.

Cho, E., and Park, M. (2016). Palmitoylation in Alzheimer’s disease and other neurodegenerative diseases. Pharmacol Res., 111, 133-151.

Choy, E., Chiu, V.K., Silletti, J., Feokistov, M., Morimoto, T., Michaelson, D., Ivanov, I.E., and Philips, M.R. (1999). Endomembrane trafficking of ras: the CAAX motif targets proteins to the ER and Golgi. Cell, 98(1), 69-80.

Cline, H. & Haas, K. (2008). The regulation of dendritic arbor development and plasticity by glutamatergic synaptic input: a review of the synaptotrophic hypothesis. J. Physiol., 586, 1509- 1517.

Collins, M.O., Husi, H., Yu, L., Brandon, J.M., Anderson, C.N., Blackstock, W.P., Choudhary, J.S., & Grant, S.G. (2006). Molecular characterization and comparison of the components and multiprotein complexes in the postsynaptic proteome. J. Neurochem., 97, 16-23.

Colon-Ramos, D.A. (2009). Chapter 2 – synapse formation in developing neural circuits. Current Topics in Developmental Biology, 87, 53-79.

Corty, M.M., Tam, J., & Grueber, W.B. (2016). Dendritic diversification through transcription factor-mediated suppression of alternative morphologies. Development, 143, 1351-1362.

Corvi, M.M., Soltys, C.L., & Berthiaume, L.G. (2001). Regulation of mitochondrial carbamoyl-phosphate synthetase 1 activity by active site fatty acylation. J. Biol. Chem., 276, 45704-45712.

Cox, A.D. & Der, C.J. (2010). Ras history: the saga continues. Small GTPases, 1, 2-27. 162

Craig, A.M. & Lichtman, J.W. (2001). Synapse formation and maturation. T.C. Sudhof, W.M. Cowan, C.F. Stevens (Eds.), Synapses, The Johns Hopkins University Press, Baltimore , pp. 1-88.

Cragg, B.G. (1975). The density of synapses and neurons in normal, mentally defective ageing human brains. Brain: A Journal of Neurology, 98(1), 81-90.

Cremer H, Chazal G, Goridis C, & Represa A. (1997). NCAM is essential for axonal growth and fasciculation in the hippocampus. Mol Cell Neurosci., 8, 323–335 Cubelos B, Sebastian-Serrano A, Beccari L, Calcagnotto ME, Cisneros E, et al. (2010). Cux1 and Cux2 regulate dendritic branching, spine morphology, and synapses of the upper layer neurons of the cortex. Neuron, 66, 523–35. Cui, Y., Costa, R.M., Murphy, G.G., Elgersma, Y., Zhu, Y., Gutmann, D.H., Parada, L.F., Mody, I., & Silva, A.J. (2008). Neurofibromin regulation of ERK signaling modulates GABA release and learning. Cell, 135, 549-560. da Silva, J.S., and Dotti, C.G. (2002). Breaking the neuronal sphere: regulation of the actin cytoskeleton in neuritogenesis. Nat. Rev. Neurosci., 3, 694–704.

Dalva, M.B., McClelland, A.C., & Kayser, M.S. (2007). Cell adhesion molecules: signalling functions at the synapse. Nat. Rev. Neurosci., 8, 206-220.

Davis, G.W. (2006). Homeostatic control of neural activity: from phenomenology to molecular design. Annu. Rev. Neurosci. 29, 307-323.

De Camilli, P., Haucke, V., Takei, K., & Mugnaini, E. (2001). The structure of synapses. T.C. Sudhof, W.M. Cowan, C.F. Stevens (Eds.), Synapses, The Johns Hopkins University Press, Baltimore , pp. 1-88. de Ligt J., Willemsen M. H., van Bon B. W. M., Kleefstra T., Yntema H. G., Kroes T., et al. (2012). Diagnostic exome sequencing in persons with severe intellectual disability. N. Engl. J. Med., 367, 1921–1929.

Dean, C. & Dresbach, T. (2006). Neuroligins and neurexins: linking cell adhesion, synapse formation and cognitive function. Trends Neurosci, 29, 21-29.

Dejanovic, B., Semtner, M., Ebert, S., Lamkemeyer, T., Neuser, F., Lüscher, B., Meier, J.C., and Schwarz, G. (2014). Palmitoylation of gephyrin controls receptor clustering and plasticity of GABAergic synapses. PLoS Biol., 12(7), e1001908.

Denker, A. & Rizzoli, S.O. (2010). Synaptic vesicle pools: an update. Front Synaptic Neurosci, 2, 135.

163

Dietrich, L.E. & Ungermann, C. (2004). On the mechanism of protein palmitoylation. EMBO Rep, 5, 1053-1057.

Dichter, M.A. (1975). Physiological properties of vertebrate nerve cells in tissue culture. UCLA Forum Med Sci, 18, 101-114.

Dityatev A, Dityateva G, Sytnyk V, Delling M, Toni N, Nikonenko I, et al. (2004). Polysialylated neural cell adhesion molecule promotes remodeling and formation of hippocampal synapses. J Neurosci., 24, 9372–9382. Dong, X., Shen, K., Bülow, H.E. (2015). Intrinsic and extrinsic mechanisms of dendritic morphogenesis. Annu. Rev. Physiol., 77, 271-300.

Dotti, C.G., Sullivan, C.A. & Banker, G.A. (1988). The establishment of polarity by hippocampal neurons in culture. J Neurosci, 8, 1454-1468.

Drisdel, R.C. & Green, W.N. (2004). Labeling and quantifying sites of protein palmitoylation. Biotechniques, 36, 276-285.

Dupraz, S., Grassi, D., Bernis, M.E., Sosa, L., Bisbal, M., Gastaldi, L., Jausoro, I., Caceres, A., Pfenninger, K.H. & Quiroga, S. (2009). The TC10-Exo70 complex is essential for membrane expansion and axonal specific in developing neurons. Journal of Neuroscience, 29, 13292- 13301.

Eberwine, J.H. (1999). Dendritic localization of mRNAs and local protein synthesis. In: Dendrites (Stuart, G., Spruston, N., & Hausser, M. Eds), 68-84. Oxford: Oxford University Press.

Ehrlich, I., Klein, M., Rumpel, S., and Malinow, R. (2007). PSD-95 is required for activity- driven synapse stabilization. Proceedings of the National Academy of Sciences of the United States of America, 104, 4176-4181.

Eichler, S.A. and Meier, J.C. (2008). E-I balance and human diseases – from molecules to networking. Front Mol Neurosci., 1, 2.

Eisenberg, S., Beckett, A.J., Prior, I.A., Dekker, F.J., Hedberg, C., Waldmann, H., Ehrlich, M., and Henis, Y.I. (2011). Raft protein clustering alters N-Ras membrane interactions and activation pattern. Mol Cell Biol., 31(19), 3938-3952.

Eisenberg, S., Laude, A.J., Beckett, A.J., Mageean, C.J., Aran, V., Hernandez-Valladares, M., Henis, Y.I. and Prior, I.A. (2013). The role of palmitoylation in regulating Ras localization and function. Biochem Soc Trans., 41(1), 79-83.

El-Husseini, A.E.-D., & Bredt, D.S. (2002). Protein palmitoylation: a regulator of neuronal development and function. Nat. Rev. Neurosci., 3, 791-802.

164

El-Husseini Ael, D., Craven, S.E., Brock, S.C., and Bredt, D.S. (2001). Polarized targeting of peripheral membrane proteins in neurons. The Journal of Biological Chemistry, 276, 44984- 44992.

El-Husseini, A.E., Craven, S.E., Chetkovich, D.M., Firestein, B.L., Schnell, E., Aoki, C., and Bredt, D.S. (2000a). Dual palmitoylation of PSD-95 mediates its vesiculotubular sorting, postsynaptic targeting, and ion channel clustering. The Journal of Cell Biology, 148, 159-172.

El-Husseini, A.E., Schnell, E., Chetkovich, D.M., Nicoll, R.A., and Bredt, D.S. (2000b). PSD- 95 involvement in maturation of excitatory synapses. Science, 290, 1364-1368.

El-Husseini, A-D., Schnell, E., Dakoji, S., Sweeney, N., Zhou, Q., Prange, O., Gauthier- Campbell, C., Aguilera-Moreno, A., Nicoll, R.A. and Bredt, D.S. (2002). Synaptic strength regulated by palmitate cycling on PSD-95. Cell, 108(6), 849-863.

Elias, G.M., Funke, L., Stein, V., Grant, S.G., Bredt, D.S., & Nicoll, R.A. (2006) Synapse- specific and developmentally regulated targeting of AMPA receptors by a family of MAGUK scaffolding proteins. Neuron, 52, 307-320.

English, J.D. & Sweatt, J.D. (1997). A requirement for the mitogen-activated protein kinase cascade in hippocampal long term potentiation. J. Biol. Chem., 272, 19103-19106.

Essrich, C., Lorez,M., Benson, J.A., Fritschy, J.M., & Luscher, B. (1998). Postsynaptic clustering of major GABAA receptor subtypes requires the gamma-2-subunit and gephyrin. Nat. Neurosci, 1, 563-571.

Fang, C., Deng, L., Keller, C.A., Fukata, M., Fukata, Y., Chen, G., and Luscher, B. (2006). GODZ-mediated palmitoylation of GABA(A) receptors is required for normal assembly and function of GABAergic inhibitory synapses. The Journal of neuroscience : the official journal of the Society for Neuroscience, 26, 12758-12768.

Fannon AM, & Colman DR. (1996). A model for central synaptic junctional complex formation based on the differential adhesive specificities of the cadherins. Neuron, 17, 423– 434.

Fernandez-Alfonso, T. & Ryan, T.A. (2008). A heterogeneous “resting” pool of synaptic vesicles that is dynamically interchanged across boutons in mammalian CNS synapses. Brain Cell Biol, 36, 87-100.

Ferreira, T., Ou, Y., Li, S., Giniger, E., and van Meyel, D.J. (2014). Dendrite architecture organized by transcriptional control of the F-actin nucleator Spire. Development, 141, 650- 660.

Fiala, J.C., Spacek, J., & Harris, K.M. (2002). Dendritic spine pathology: cause of consequence of neurological disorders. Brain Res. Brain Res. Rev, 39, 29-54. 165

Fields, R.D. (2008). White matter in learning, cognition, and psychiatric disorders. Trends Neurosci., 31, 361-370.

Fivaz, M., Bandara, S., Inoue, T., & Meyer, T. (2008). Robust neuronal symmetry breaking by Ras-triggered local positive feedback. Current Biology, 18, 44-50.

Fletcher, T.L. & Banker, G.A. (1989). The establishment of polarity by hippocampal neurons: the relationship between the stage of a cell's development in situ and its subsequent development in culture. Developmental Biology, 136, 446-454.

Foletti, D.L., Prekeris, R., and Scheller, R.H. (1999). Generation and maintenance of neuronal polarity: mechanisms of transport and targeting. Neuron, 23, 641–644.

Forrester, M.T., Hess, D.T., Thompson, J.W., Hultman, R., Moseley, M.A., Stamler, J.S. & Casey, P.J. (2011). Site-specific analysis of protein S-acylation by resin-assisted capture. J Lipid Res, 52, 393-398.

Freund, T.F. & Katona, I. (2007). Perisomatic inhibition. Neuron, 56, 33-42.

Fromer, M., Pocklington, A.J., Kavanagh, D.H., Williams, H.J., Dwyer, S., Gormley, P., Georgieva, L., Rees, E., Palta, P., Ruderfer, D.M et al. (2014). De novo mutations in schizophrenia implicate synaptic networks. Nature, 506, 179-184.

Fujita A, Koinuma S, Yasuda S, Nagai H, Kamiguchi H, Wada N, and Nakamura, T. (2013). GTP Hydrolysis of TC10 Promotes Neurite Outgrowth through Exocytic Fusion of Rab11- and L1-Containing Vesicles by Releasing Exocyst Component Exo70. PLoS ONE, 8(11): e79689.

Fukata, M., Fukata, Y., Adesnik, H., Nicoll, R.A., and Bredt, D.S. (2004). Identification of PSD-95 palmitoylating enzymes. Neuron, 44, 987-996.

Fukata, Y., and Fukata, M. (2010). Protein palmitoylation in neuronal development and synaptic plasticity. Nature reviews Neuroscience, 11, 161-175.

Fukata, Y., Dimitrov, A., Boncompain, G., Vielemeyer, O., Perez, F., and Fukata, M. (2013). Local palmitoylation cycles define activity-regulated postsynaptic subdomains. J Cell Biol., 202(1), 145-161.

Gardiol, A., Racca, C., & Triller, A. (1999). Dendritic and postsynaptic protein synthetic machinery. J. Neurosci, 19, 168-179.

Gartner U., Alpar A., Seeger G., Heumann R., and Arendt, T. (2004) Enhanced Ras activity in pyramidal neurons induces cellular hypertrophy and changes in afferent and intrinsic connectivity in synRas mice. Int J Dev Neurosci., 22, 165–173.

166

Gatto, C.L. & Broadie, K. (2010). Genetic controls balancing excitatory and inhibitory synaptogenesis in neurodevelopmental disorder models. Front. Synaptic Neurosci., 2, 4.

Gauthier-Campbell, C., Bredt, D.S., Murphy, T.H., and El-Husseini, A.E.-D. (2004). Regulation of dendritic branching and filopodia formation in hippocampal neurons by specific acylated protein motifs. Mol. Biol. Cell., 15, 2205–2217.

Gecz, J., Shoubridge, C., Corbett, M. (2009). The genetic landscape of intellectual disability arising from chromosome X. Trends Genet., 25, 308-316.

George, J., Soares, C., Montersino, A., Beique, J.C. and Thomas, G.M. (2015). Palmitoylation of LIM Kinase-1 ensures spine-specific actin polymerization and morphological plasticity. Elife, 4, e06327.

Gerrow, K., Romorini, S., Nabi, S.M., Colicos, M.A., Sala, C., & El-Husseini, A. (2006). A preformed complex of postsynaptic proteins is involved in excitatory synapse development. Neuron, 49, 547-562.

Goddard, A.D. and Watts, A. (2012). Regulation of G protein-coupled receptors by palmitoylation and cholesterol. BMC Biol., 10, 27.

Golgi, C. (1873). Sulla struttura della grigia del cervello. Gazetta Medica Italiana. Finger, S. (Ed). Reviewed in “Minds behind the brain – a history of pioneers and their discoveries”, Oxford University Press, Oxford, 204-206.

Gomis-Ruth, S., Wierenga, C.J., & Bradke, F. (2008). Plasticity of polarization: changing dendrites into axons in neurons integrated in neuronal circuits. Curr. Biol., 18, 992-1000.

Goodwin, J.S. and Kenworthy, A.K. (2005). Photobleaching approaches to investigate diffusional mobility and trafficking of Ras in living cells. Methods, 37(2), 154-164.

Goodwin, J.S., Drake, K.R., Rogers, C., Wright, L., Lippincott-Schwartz, J., Philips, M.R., and Kenworthy, A.K. (2005a). Depalmitoylated Ras traffics to and from the Golgi complex via a nonvesicular pathway. J Cell Biol., 170(2), 261-272.

Goslin, K., & Banker, G. (1989). Experimental observations on the development of polarity by hippocampal neurons in culture. J. Cell. Biol., 108, 1507-1516.

Graf, E.R., Zhang, X., Jin, S.X., Linhoff, M.W., & Craig, A.M. (2004). Neurexins induce differentiation of GABA and glutamate postsynaptic specializations via neuroligins. Cell, 646, 1013-1026.

Greaves, J. and Chamberlain, L.H. (2006). Dual role of the cysteine-string domain in membrane binding and palmitoylation-dependent sorting of the molecular chaperone cysteine- string protein. Mol Biol Cell, 17(11), 4748-4759. 167

Greaves, J., & Chamberlain, L.H. (2011a). DHHC palmitoyl transferases: substrate interactions and (patho)physiology. Trends Biochem. Sci., 36, 245-253.

Greaves, J. and Chamberlain, L.H. (2011). Differential palmitoylation regulates intracellular patterning of SNAP25. J Cell Sci., 124(8), 1351-1360.

Greaves, J., Carmichael, J.A., and Chamberlain, L.H. (2011). The palmitoyl transferase DHHC2 targets a dynamic membrane cycling pathway: regulation by a C-terminal domain. Molecular Biology of the Cell, 22, 1887-1895.

Greaves, J., Gorleku, O.A., Salaun, C., and Chamberlain, L.H. (2010). Palmitoylation of the SNAP25 protein family: specificity and regulation by DHHC palmitoyl transferases. J Biol Chem., 285, 24629-24638.

Greaves, J., Salaun, C., Fukata, Y., Fukata, M., and Chamberlain, L.H. (2008). Palmitoylation and membrane interactions of the neuroprotective chaperone cysteine-string protein. The Journal of Biological Chemistry, 283, 25014-25026.

Greger, I.H. & Esteban, J.A. (2007). AMPA receptor biogenesis and trafficking. Curr. Opin. Neurobiol, 17, 289-297.

Gu, B. and Daltone, K.A. (2017). Models and detection of spontaneous recurrent seizures in laboratory rodents. Zool Res., 38(4), 171-179.

Gulledge, A.T., Kampa, B.M., & Stuart, G.J. (2005).Synaptic integration in dendritic trees. J. Neurobiol., 64, 75-90.

Ha, S., and Redmond, L. (2008). ERK mediates activity dependent neuronal complexity via sustained activity and CREB-mediated signaling. Dev Neurobiol., 68(14), 1565-1579.

Haider, B., & McCormick, D.A. (2009). Rapid neocortical dynamics: cellular and network mechanisms. Neuron, 62, 171-189.

Hall, A. (1990). The cellular functions of small GTP-binding proteins. Science, 249, 635-640.

Han, J.Y., Lee, I.G., Shin, S., Kim, M., Jang, J.H., and Park, J. (2017). The first patient with sporadic X-linked intellectual disability with de novo ZDHHC9 mutation identified by next- generation sequencing. Eur J Med Genet., 60(10), 499-503.

Hangya, B., Pi, H.J., Kvitsiani, D., Ranade, S.P., & Kepecs, A. (2014). From circuit motif to computations: mapping the behavioral repertoire of cortical interneurons. Curr Opin Neurobil, 26, 117-124.

Hanus, C., Kochen, L., Dieck S., T., Racine, V., Sibarita, J.B., Schuman, E.M. & Ehlers, M.D. (2014). Synaptic control of secretory trafficking in dendrites. Cell Rep., 7, 1771-1778. 168

Hanus, C., Geptin, H., Tushev, G., Garg, S., Alvarez-Castelao, B., Sambandan, S., Kochen, L., Hafner, A.S., Langer, J.D., & Schuman, E.M. (2016). Unconventional secretory processing diversifies neuronal ion channel properties. eLife, 5, e20609.

Harvey, K., Duguid, I.C., Alldred, M.J., Beatty, S.E., Ward, H., Keep, N.H., Lingenfelter, S.E., Pearce, B.R., Lundgren, J., Owen, M.J. et al. (2004). The GDP-GTP exchange factor collybistin: an essential determinant of neuronal gephyrin clustering. J Neurosci., 24(25), 5816-5826.

Hausser, M., Spruston, N. and Stuart, G.J. (2000). Diversity and dynamics of dendritic signaling. Science, 290(5492), 739-744.

Hayashi, S. & Takeichi, M. (2015). Emerging role of protocadherins: from self-avoidance to enhancement of motility. J Cell Sci, 128, 1455-1464.

Hayashi, T., Rumbaugh, G., Huganir, R.L. (2005). Differential regulation of AMPA receptor subunit trafficking by palmitoylation of two distinct sites. Neuron, 47(5), 709-723.

Hayashi, T., Thomas, G.M., and Huganir, R.L. (2009). Dual palmitoylation of NR2 subunits regulates NMDA receptor trafficking. Neuron, 64(2), 213-226.

Hemsley, P.A. (2009). Protein S-acylation in plants (Review). Mol Membr Biol., 26(1), 114- 125. Herms, J. & Dorostkar, M.M. (2016). Dendritic spine pathology in neurodegenerative diseases. Annual Review of Pathology, 11, 221-250.

Heumann, R., Goemans, C., Bartsch, D., Lingenhöhl, K., Waldmeier, P.C., Hengerer, B., Allegrini, P.R., Schellander, K., Wagner, E.F., Arednt, T et al. (2000). Transgenic activation of Ras in neurons promotes hypertrophy and protects from lesion-induced degeneration. J Cell Biol., 151(7), 1547-1548.

Horch, H. W., Krttgen, A., Portbury, S.D., and Katz, L.C. (1999). Destabilization of cortical dendrites and spines by BDNF. Neuron, 23, 353–364.

Hornemann, T. (2015). Palmitoylation and depalmitoylation defects. J. Inherit. Metab. Dis., 38, 179-186.

Horton, A.C. and Ehlers, M.D. (2003) Dual modes of endoplasmic reticulum-to-Golgi transport in dendrites revealed by live-cell imaging. J. Neurosci. 23, 6188–6199.

Horton, A.C., Racz, B., Monson, E.E., Lin, A.L., Weinberg, R.J. & Ehlers, M.D. (2005) Polarized secretory trafficking directs cargo for asymmetric dendrite growth and morphogenesis. Neuron, 48, 757–771

169

Hseuh, Y.P. & Sheng, M. (1999). 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, 274, 532-536.

Hu, H., Gan, J., & Jonas, P. (2014). Interneurons. Fast-spiking, parvalbumin+ GABAergic interneurons: from cellular design to microcircuit function. Science, 345, 1255263.

Hu, H., Qin, Y., Bochorishvili, G., Zhu, Y., van Aelst, L., and Zhu, J.J. (2008). Ras signaling mechanisms underlying impaired GluR1-dependent plasticity associated with Fragile X Syndrome. J Neurosci., 28(31), 7847-7862.

Hua, J.Y. and Smith, S.J. (2004). Neural activity and the dynamics of central nervous system development. Nature Neuroscience, 7, 327–332.

Huang, K., & El-Husseini, A. (2005). Modulation of neuronal protein trafficking and function by palmitoylation. Current Opinion in Neurobiology, 15, 527-535.

Huang, K., Sanders, S.S., Kang, R., Caroll, J.B., Sutton, L., Wan, J., Singaraja, R., Young, F.B., Liu, L., El-Husseini, A., Davis, N.G., & Hayden, M.R. (2011). Wild-type HTT modulates the enzymatic activity of the neuronal palmitoyl transferase HIP14. Hum. Mol. Genet., 20, 3356-2265.

Huang, K., Yanai, A., Kang, R., Arstikaitis, P., Singaraja, R.R., Metzler, M., Mullard, A., Haigh, B., Gauthier-Campbell, C., Gutekunst, C.A., Hayden, M.R. & El-Husseini, A. (2004). Huntingtin-interacting protein HIP14 is a palmitoyl transferase involved in palmitoylation and trafficking of multiple neuronal proteins. Neuron, 44, 977-986.

Huganir, R.L. & Nicoll, R.A. (2013). AMPARs and synaptic plasticity: the last 25 years. Neuron, 80, 704-717.

Hume, R.I. and Purves, D. (1981). Geometry of neonatal neurones and the regulation of synapse elimination. Nature, 293(5832), 469-71.

Humeau, Y., Gambino, F., Chelly, J. and Vitale, N. (2009). X-linked mental retardation: focus on synaptic function and plasticity. J Neurochem., 109(1), 1-14.

Hunt, D.L. & Castillo, P.E. (2012). Synaptic plasticity of NMDA receptors: mechanisms and functional implications. Curr. Opin. Neurobiol., 22, 496-508.

Huttenlocher, P.R. (1970). Dendritic development and mental defect. Neurology 20(4), 381. Huttenlocher, P.R. (1974). Dendritic development in neocortex of children with mental defect and infantile spasms. Neurology, 24, 203-210.

Ikeda, K., & Bekkers, J.M. (2009). Counting the number of releasable synaptic vesicles in a presynaptic terminal. PNAS, 106, 2945-2950. 170

Jacobs, B., Schall, M., Prather, M., Kapler., F., Driscoll, L., Baca, S., Jacobs, J., Ford, K., Wainwright, M., & Treml, M. (2001). Regional dendritic and spine variation in human cerebral cortex: a quantitative golgi study. Cereb Cortex, 11, 558-571.

Jan, Y-N. & Jan, L.Y. (2003). The control of dendrite development. Neuron, 40, 229-242.

Jeffries, O., Geiger, N., Rowe, I.C., Tian, L., McClafferty, H., Chen, L., Bi, D., Knaus, H.G., Ruth, P. and Shipston, M.J. (2010). Palmitoylation of the S0-S1 linker regulates cell surface expression of voltage- and calcium-activated potassium (BK) channels. J Biol Chem, 285(43), 33307-33314.

Jiang, M., Lee, C.L., Smith, K.L., and Swann, J.W. (1998). Spine loss and other persistent alterations of hippocampal pyramidal cell dendrites in a model of early-onset epilepsy. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 18, 8356-8368.

Jindal, G.A., Goyal, Y., Burdine, R.D., Rauen, K.A. & Shvartsman, S.Y. (2015). RASopathies: unraveling mechanisms with animal models. Dis Model Mech., 8, 769-782.

Jinno, S., and Kosaka, T. (2006). Cellular architecture of the mouse hippocampus: a quantitative aspect of chemically defined GABAergic neurons with stereology. Neurosci Res., 56(3), 229-245.

Joberty, G., Perlungher, R.R., & Macara, I.G. (1999). The Borgs, a new family of Cdc42 and TC10 GTPase-interacting proteins. Molecular and Cellular Biology, 19, 6585-6597.

Joseph, A. & Turrigiano, G.G. (2017). All for one but not one for all: excitatory synaptic scaling and intrinsic excitability are coregulated by CaMKIV, whereas inhibitory synaptic scaling is under independent control. J. Neurosci, 37, 6778-6785.

Jung, K.Y., Kim, J.M., and Kim, D.W. (2003). Patterns of interictal spike propagation across the central sulcus in benign rolandic epilepsy. Clin Electroencephalogr., 34(3), 153-157.

Kaech S, Banker G. Culturing hippocampal neurons. (2006). Nat. Protoc,. 1, 2406–2415.

Kalscheuer, V.M., Musante, L., Fang, C., Hoffmann, K., Fuchs, C., Carta, E., Deas, E., Venkateswarlu, K., Menzel, C., Ullmann, R., Tommerup, N., Dalprà, L., Tzschach, A., Selicorni, A., Lüscher, B., Ropers, H.H., Harvey, K., & Harvey, R.J. (2009). A balanced chromosomal translocation disrupting ARHGEF9 is associated with epilepsy, anxiety, aggression, and mental retardation. Hum. Mutat., 30, 61-68.

Kang, R. (2004). Presynaptic Trafficking of Synaptotagmin I Is Regulated by Protein Palmitoylation. Journal of Biological Chemistry, 279, 50524–50536.

Kang, R., Wan, J., Arstikaitis, P., Takahashi, H., Huang, K., Bailey, A.O., Thompson, J.X., Roth, A.F.. Drisdel, R.C. et al. (2008). Neural palmitoyl-proteomics reveals dynamic synaptic 171

palmitoylation. Nature, 456(7224), 904-90.

Kanzaki, M., Mora, S., Hwang, J.B., Saltiel, A.R., & Pessin, J.E. Atypical protein kinase C (PKCzeta/lambda) is a convergent downstream target of the insulin-stimulated phosphatidylinositol 3-kinase and TC10 signaling pathways. (2004). Journal of Cell Biology, 164, 279-290.

Kato, H., Allen, N.D., Emson, P.C., and Kiyama, H. (2000). GAP-43 N-terminal translocation signal targets beta-galactosidase to developing axons in a panneuronal transgenic mouse line. Brain Res. Dev. Brain Res., 121, 109–112.

Kaufmann, W.E. and Moser, H.W. (2000). Dendritic anomalies in disorders associated with mental retardation. Cereb Cortex., 10(10), 981-991.

Kaur, I., Yarov-Yarovoy, V., Kirk, L.M., Plambeck, K.E., Barragan, E.V., Ontiveros, E.S., and Diaz, E. (2016). Activity-dependent palmitoylation controls SynDIG1 stability, localization, and function. J Neurosci., 36(29), 7562-7568.

Kawase, K., Nakamura, T., Takaya, A., Aoki, K., Namikawa, K., Kiyama, H., Inagaki, S., Takemoto, H., Saltiel, A.R., and Matsuda, M. (2006). GTP hydrolysis by the Rho family GTPase TC10 promotes exocytic vesicle fusion. Dev. Cell., 11, 411-421.

Kay, H.Y., Greene, D.L., Kang, S., Kosenko, A., & Hoshi, N. (2015). M-current preservation contributes to anticonvulsant effects of valproic acid. J. Clin. Invest., 125, 3904-3914.

Keith, D.J., Sanderson, J.L., Gibson, E.S., Woolfrey, K.M., Robertson, H.R., Olszewski, K., Kang, R., El-Husseini, A., & Dell’acqua, M.L. (2012). Palmitoylation of A-kinase anchoring protein 79/150 regulates dendritic endosomal targeting and synaptic plasticity mechanisms. J Neurosci, 32, 7119-7136.

Keller, C.A., Yuan, X., Panzanelli, P., Martin, M.L., Alldred, M., Sassoè-Pognetto M, and Lüscher B. (2004). The gamma2 subunit of GABA(A) receptors is a substrate for palmitoylation by GODZ. J Neurosci, 24(26), 5881-5891.

Kepecs, A., & Fishell, G. (2014). Interneuron cell types are fit to function. Nature, 505, 318- 326.

Kim, J.H., Liao, D., Lau, L.F., & Huganir, R.L. SynGAP: a synaptic RasGAP that associates with the PSD-95/SAP90 family. Neuron, 20, 683-691. Kneussel, M., Brandstatter, J.H., Laube, B., Stahl, S., Muller, U., & Betz, H. (1999). Loss of postsynaptic GABA(A) receptor clustering in gephyrin-deficient mice. J. Neuro., 19, 9289- 9297.

Ko, P.J., and Dixon, S.J. (2018). Protein palmitoylation and cancer. EMBO reports 19.

172

Kolkova, K., Novitskaya, V., Pedersen, N., Berezin, V. & Bock, E. (2000). Neural cell adhesion molecule-stimulated neurite outgrowth depends on activation of protein kinase C and the Ras-mitogen-activated protein kinase pathway. J Neurosci, 20, 2238-2246.

Konur, S. and Ghosh, A. (2005). Calcium signaling and the control of dendritic development. Neuron, 46, 401–405.

Kornau, H.C., Schenker, L.T., Kennedy, M.B., & Seeburg, P.H. (1995). Domain interaction between NMDA receptor subunits and the postsynaptic density protein PSD-95. Science, 269, 1737-1740.

Korycka, J., Lach, A., Heger, E., Boguslawska, D.M., Wolny, M., Toporkiewicz, M., Augoff, K., Korzeniewski, J., & Sikorski, A.F. (2012). Human DHHC proteins: a spotlight on the hidden player of palmitoylation. Eur. J. Cell. Biol., 91, 107-117.

Kostiuk, M.A., Corvi, M.M., Keller, B.O., Plummer, G., Prescher, J.A., Hangauer, M.J., Bertozzi, C.R., Rajaiah, G., Falck, J.R., and Berthiaume, L.G. (2008). Identification of palmitoylated mitochondrial proteins using a bio-orthogonal azido-palmitate analogue. The FASEB Journal, 22, 721–732.

Krijnse-Locker, J., Parton, R.G., Fuller, S.D., Griffiths, G., & Dotti, C.G. (1995). The organization of the endoplasmic reticulum and the intermediate compartment in cultured rat hippocampal neurons. Mol. Biol. Cell., 6, 1315-1332.

Kurochkina, N. & Guha, U. (2013). SH3 domains: modules of protein-protein interactions. Biohys Rev, 5, 29-39.

Kumar, V., Zhang, M-X., Swank, M.W., Kunz, J., and Wu, G-Y. (2005). Regulation of dendritic morphogenesis by Ras-PI3K-Akt-mTOR and Ras-MAPK signaling pathways. J Neurosci, 25(49), 11288-11299.

Kushner, S.A., Elgersma,Y., Murphy, G.G., Jaarsma, D., van Woerden, G.M., Hojjati, M.R., Cui, Y., LeBoutillier, J.C., Marrone, D.F., Choi, E.S. et al. (2005). Modulation of presynaptic plasticity and learning by the H-ras/extracellular signal-regulated kinase/synapsin I signaling pathway. J. Neurosci., 25, 9721-9734.

Kutzleb, C., Sanders, G., Yamamoto, R., Wang, X., Lichte, B., Petrasch-Parwez, E., and Kilimann, M.W. (1998). Paralemmin, a prenyl-palmitoyl-anchored phosphoprotein abundant in neurons and implicated in plasma membrane dynamics and cell process formation. The Journal of Cell Biology, 143, 795–813.

Laux, T., Fukami, K., Thelen, M., Golub, T., Frey, D., and Caroni, P. (2000). GAP43, MARCKS, and CAP23 modulate PI(4,5)P(2) at plasmalemmal rafts, and regulate cell cortex actin dynamics through a common mechanism. The Journal of Cell Biology, 149, 1455–1472.

173

Lee, H.-J. & Zheng, J.J. (2010). PDZ domains and their bindings partners: structure, specificity, and modification. Cell Communication and Signaling, 8, 8.

Lee, M.C., Miller, E.A., Goldberg, J., Orci, L. & Schekman, R. (2004) Bi-directional protein transport between the ER and Golgi. Annu. Rev. Cell Dev. Biol., 20, 87–123.

Lein, E.S., Hawrylycz, M.J., Ao, N., Ayres, M., Bensinger, A., Bernard, A., Boe, A.F., Boguski, M.S., Brockway, K.S., Byrnes, E.J. et al. (2007). Genome-wide atlas of gene expression in the adult mouse brain. Nature, 445(7124), 168-176.

Lemke J. R., Riesch E., Scheurenbrand T., Schubach M., Wilhelm C., Steiner I., et al. (2012). Targeted next generation sequencing as a diagnostic tool in epileptic disorders. Epilepsia, 53 1387–1398.

Lesca G., Till M., Labalme A., Vallee D., Hugonenq C., Philip N., et al. (2011). De novo Xq11.11 microdeletion including ARHGEF9 in a boy with mental retardation, epilepsy, macrosomia, and dysmorphic features. Am. J. Med. Genet. A., 155A, 1706–1711.

Levental, I., Grzybek, M., & Simons, K. (2010). Greasing their way: lipid modifications determine protein association with membrane rafts. Biochemistry, 49, 6305-6316.

Levi, S., Logan, S.M., Tovar, K.R., & Craig, A.M. (2004). Gephyrin is critical for glycine receptor clustering but not for the formation of functional GABAergic synapses in hippocampal neurons. J. Neurosci., 24, 207-217.

Levinson, J.N. & El-Husseini, A. (2005). New players tip the scales in the balance between excitatory and inhibitory synapses. Mol. Pain. 1, 12.

Levinson, J.N., Li, R., Kang, R., Moukhles, H., El-Husseini, A. & Bamji, S.X. (2010). Postsynaptic scaffolding molecules modulate the localization of neuroligins. Neuroscience, 165, 782-793.

Levy, A.D., Devignot, V., Fukata, Y., Fukata, M., Sobel, A., & Chauvin, S. (2011). Subcellular Golgi localization of stathmin family proteins is promoted by a specific set of DHHC palmitoyl transferases. Molecular Biology of the Cell, 22, 1930-1942.

Li N, Zhao CT, Wang Y, & Yuan XB. (2010). The transcription factor Cux1 regulates dendritic morphology of cortical pyramidal neurons. PLOS ONE, 5, e10596.

Lievens, P.M., Kuznetsova, T., Kochlamazashvili, G., Cesca, F., Gorinski, N., Galil, D.A., Cherkas, V., Ronkina, N., Lafera, J., Gaestel, M. et al. (2016). ZDHHC3 tyrosine phosphorylation regulates neural cell adhesion molecular palmitoylation. Mol Cell Biol., 36(17), 2208-2225.

174

Lin, D.T. & Conibear, E.(2015a). ABHD17 proteins are novel protein depalmitolases that regulate N-Ras palmitate turnover and subcellular localization. eLife, 4, e11306.

Lin, D.T. & Conibear, E. (2015b). Enymatic protein depalmitoylation by acyl protein thioesterases. Biochem. Soc. Trans., 43, 193-198.

Linder, M.E. & Deschenes, R.J. (2007). Palmitoylation: policing protein stability and traffic. Nat Rev Mol Cell Biol, 8, 74-84.

Linhares, N.D., Valadares, E.R., da Costa, S.S., Arantes, R.R., de Oliveira, L.R., Rosenberg, C., Vianna-Morgante, A.M., and Svartman, M. (2016). Inherited Xq13.2-q21.31 duplication in a boy with recurrent seizures and pubertal gynecomastia: Clinical, chromosomal and aCGH characterization. Meta Gene, 9, 185-190.

Little, E.B., Edelman, G.M. & Cunningham, B.A. (1998). Palmitoylation of the cytoplasmic domain of the neural cell adhesion molecule N-CAM serves as an anchor to cellular membranes. Cell Adhes Commun, 6, 415-430.

Liu, G. (2004). Local structural balance and functional interaction of excitatory and inhibitory synapses in hippocampal dendrites. Nat Neurosci., 7(4), 373-379.

Liu, H, Abecasis, G.R., Heath, S.C., Knowles, A., Demars, S., Chen, Y-J., Roos, J.L., Rapoport, J.L., Gogos, J.A. & Karayiorgou, M. (2002). Genetic variation in the 22q11 locus and susceptibility to schizophrenia. PNAS, 99, 16859-16864.

Liu, Y.Q., Yu, F., Liu, W.H., He, X.H., and Peng, B.W. (2014). Dysfunction of hippocampal interneurons in epilepsy. Neurosci Bull., 30(6), 985-998.

Lobo, S., Greentree, W.K., Linder, M.E. and Deschenes, R.J. (2002). Identification of a Ras Palmitoyltransferase in Saccharomyces cerevisiae. J Biol Chem, 277(43), 41268-41273.

Long, J.Z. & Cravatt, B.F. (2011). The metabolic serine hydrolases and their functions in mammalian physiology and disease. Chem. Rev., 111, 6022-6063.

Lu, W., Shi, Y., Jackson, A.C., Bjorgan, K., During, M.J., Sprengel, R., Seeburg, P.H. & Nicoll, R.A. (2009). Subunit composition of synaptic AMPA receptors revealed by a single- cell genetic approach. Neuron, 62, 254-268.

Luscher, B., Fuchs, T., & Kilpatrick, C.L. (2011). GABAA receptor mediated trafficking- mediated plasticity of inhibitory synapses. Neuron, 70, 385-409.

Lynch, S.J., Snitkin, H., Gumper, I., Philips, M.R., Sabatini, D., and Pellicer, A. (2015). The differential palmitoylation states of N-Ras and H-Ras determine their distinct golgi sub- compartment localizations. J Cell Phyiol., 230(3), 610-619.

175

Machado, C.O., Griesi-Oliveira, K., Rosenberg, C., Kok, F., Martins, S., Passos-Bueno, M.R., & Sertie, A.L. (2016). Collybistin binds and inhibits mTORC1 signaling: a potential novel mechanism contributing to intellectual disability and autism. Eur. J. Hum. Genet., 24, 59-65.

Magee, A.I. & Courtneidge, S.A. (1985). Two classes of fatty acid acylated proteins exist in eukaryotic cells. EMBO Journal, 4, 1137-1144.

Mancarci, B.O., Toker, L., Tripathy, S.J., Li, B., Rocco, B., Sibille, E., and Pavlidis, P. (2017). Cross-laboratory analysis of brain cell type transcriptomes with applications to interpretation of bulk tissue data. eNeuro, 4(6), ENEURO.0212-17.2017.

Mansouri, M.R., Marklund, L., Gustavsson, P., Davey, E., Carlsson, B., Larsson, C., White, I., Gustavson, K.H., and Dahl, N. (2005). Loss of ZDHHC15 expression in a woman with a balanced translocation t(X;15)(q13.3;cen) and severe mental retardation. European Journal of Human Genetics : EJHG, 13, 970-977.

Marco E. J., Abidi F. E., Bristow J., Dean W. B., Cotter P., Jeremy R. J., et al. (2008). ARHGEF9 disruption in a female patient is associated with X linked mental retardation and sensory hyperarousal. J. Med. Genet,. 45 100–105.

Margeta, M.A. & Shen, K. (2009). Molecular mechanisms of synaptic specificty. Mol Cell Neurosci, 43, 261-267.

Marin-Padilla, M. (1972). Structural abnormalities of the cerebral cortex in human chromosomal aberrations: a Golgi study. Brain Research, 44, 625-629.

Martinez, F., Rosello, M., Mayo, S., Monfort, S., Oltra, S., and Orellana, C. (2014). Duplication at Xq13.3-q21.1 with syndromic intellectual disability, a probable role for the ATRX gene. American Journal of Medical Genetics Part A, 164A, 918-923.

Martinez-Cerdeno, V. (2017). Dendrite and spine modifications in autism and related neurodevelopmental disorders in patients and animal models. Dev. Neurobiol., 77, 393-404.

Marx, M., Haas, C.A., and Häussler, U. (2013). Differential vulnerability of interneruons in the epileptic hippocampus. Front Cell Neurosci., 1, 5.

Masurel-Paulet, A., Kalscheuer, V.M., Lebrn, N., Hu, H., Levy, F., Thauvin-Robinet, C., Darmecy-Stamboul, V., El Chehadeh, S., Thevenon, J., Chancenotte, S. et al. (2014). Expanding the clinical phenotype of patients with a ZDHHC9 mutation. Am J Med Genet A., 164A(3), 789-795.

Matallanas, D., Sanz-Moreno, V., Arozarena, I., Calvo, F., Agudo- Ibáñez, L., Santos, E., Berciano, M.T., and Crespo, P. (2006). Distinct utilization of effectors and biological outcomes resulting from site-specific Ras activation: Ras functions in lipid rafts and Golgi complex are dispensable for proliferation and transformation. Mol Cell Biol., 26(1), 100-116. 176

Maulik, P.K., Mascarenhas, M.N., Mathers, C.D., Dua, T., & Saxena, S. (2011). Prevalence of intellectual disability: a meta-analysis of population-based studies. Res. Dev. Disabil., 32, 419- 436.

Mayer, S., Kumar, R., Jaiswal, M., Soykan, T., Ahmadian, M.R., Brose, N., Betz, H., Rhee, J.S., and Papadopoulos, T. (2013). Collybistin activation by GTP-TC10 enhances postsynaptic gephyrin clustering and hippocampal GABAergic neurotransmission. Proc Natl Acad Sci U.S.A., 110(51), 20795-20800.

McCormack, S.G., Stornetta, R.L., & Zhu, J.J. (2006). Synaptic AMPA receptor exchange maintains bidirectional plasticity. Neuron, 50, 75-88.

McCormick, D.A. (2002). Cortical and subcortical generators of normal and abnormal rhythmicity. Int. Rev. Neurobiol., 49, 99-114.

McIntire, S.L., Reimer, R.J., Schuske, K., Edwards, R.H. & Jorgensen, E.M. (1997). Identification and characterization of the vesicular GABA transporter. Nature, 389, 870-876.

Meijering, E., Jacob, M., Sarria, J.C., Steiner, P., Hirling, H., and Unser, M. (2004). Design and validation of a tool for neurite tracing and analysis in fluorescence microscopy images. Cytometry Part A : the Journal of the International Society for Analytical Cytology, 58, 167- 176.

Michaelson, D., Silletti, J., Murphy, G., D’Eustachio, P., Rush, M., and Philips, M.R. (2001). Differential localization of Rho GTPases in live cells: regulation by hypervariable regions and RhoGDI binding. J Cell Biol., 152(1), 111-126.

Mill, P., Lee, A.W., Fukata, Y., Tsutsumi, R., Fukata, M., Keighren, M., Porter, R.M., McKie, L., Smyth, I., and Jackson, I.J. (2009). Palmitoylation regulates epidermal homeostasis and hair follicle differentiation. PLoS Genet, 5, e1000748.

Miller, F.D. & Kaplan, D.R. (2003). Signaling mechanisms underlying dendrite formation. Curr Opin Neurobiol, 13, 391-398.

Miller, G. A. & Penfield, W. (1956). The magic number seven plus or minus two: some limits on our capacity for processing information. Psychol. Rev., 63, 81-97.

Mitchell, D.A., Hamel, L.D., Ishizuka, K., Mitchell, G., Schaefer, L.M. and Deschenes, R.J. (2012). The Erf4 subunit of the yeast Ras palmitoyl acyltransferase is required for stability of the acyl-Erf2 intermediate and palmitoyl transfer to a Ras2 substrate. J Biol Chem., 287(41), 34337-34348.

Mitchell, D.A., Hamel, L.D., Krishna, D.R., Farh, L., Rettew, L.M., Sanchez, P.R., and Deschenes, R.J. (2014). Mutations in the X-linked intellectual disability gene, zDHHC9, alter 177

autopalmitoylation activity by distinct mechanisms. J Biol Chem., 289(26), 18582-18592.

Mitchell, D.A., Vasudevan, A., Linder, M.E. and Deschenes, R.J. (2006). Protein palmitoylation by a family of DHHC protein S-acyltransferases. J Lipid Res, 47(6), 1118-1127.

Mizumaru, C., Saito, Y., Ishikawa, T., Yoshida, T., Yamamoto, T., Nakaya, T., & Suzuki, T. Suppression of APP-containing vesicle trafficking and production of beta-amyloid by AID/DHHC-12 protein. (2009). J. Neurochem., 111, 1213-1224.

Moore, A.W., Jan, L.Y., & Jan, Y.N. (2002). Hamlet, a binary genetic switch between single- and multiple- dendrite neuron morphology. Science, 297, 1355-1358.

Mor, A. and Philips, M.R. (2006). Compartmentalized Ras/MAPK signaling. Annu Rev Immunol., 24, 771-800.

Moutton, S., Bruel, A.L., Assoum, M., Chevarin, M., Sarrazin, E., Goizet, C., Guerrot, A.M., Charollais, A., Charles, P., Heron, D., et al. (2018). Truncating variants of the DLG4 gene are responsible for intellectual disability with marfanoid features. Clin Genet., 93, 1172-1178.

Mukai, J., Dhilla, A., Drew, L.J., Stark, K.L., Cao, L., MacDermott, A.B., Karayiorgou, M., and Gogos, J.A. (2008). Palmitoylation-dependent neurodevelopmental deficits in a mouse model of 22q11 microdeletion. Nat Neurosci, 11, 1302-1310.

Murphy, G.A., Solski, P.A., Jillian, S.A., de la Ossa, P.P., D’Eustachio, P., Der, C.J. & Rush, M.G. (1999). Cellular functions of TC10, a Rho family GTPase: regulation of morphology, signal transduction and cell growth. Oncogene, 18, 3831-3845.

Murphy, K.C., Jones, L.A., & Owen, M.J. (1999). High rates of schizophrenia in adults with velo-cardio-facial syndrome. Arch. Gen. Psychiatry, 56, 940-945.

Nadolski, M.J., & Linder, M.E. (2007).Protein lipidation. FEBS J., 274, 5202-5210.

Nakayama, K., Kiyosue, K., & Taguchi, T. (2005). Diminished neuronal activity increases neuron-neuron connectivity underlying silent synapse formation and the rapid conversion of silent to functional synapses. J. Neurosci, 25, 4040-4051.

Neves, G., Cooke, S.F. & Bliss, T.V. (2008). Synaptic plasticity, memory and the hippocampus: a neural network approach to causality. Nat Rev Neurosci, 9, 65-75.

Nicoll, R.A., Tomita, S., & Bredt, D.S. (2006). Auxiliary subunits assist AMPA-type glutamate receptors. Science, 311, 1253-1256.

Niethammer, M., Kim, E., & Sheng, M. (1996). Interaction between the C terminus of NMDA receptor subunits and multiple members of the PSD-95 family of membrane-associated guanylate kinases. J. Neurosci., 16, 2157-2163. 178

Niethammer, P., Delling, M., Sytnyk, V., Dityatev, A., Fukami, K., & Schachner, M. (2002). Cosignaling of NCAM via lipid rafts and the FGF receptor is required for neuritogenesis. J Cell Biol, 157, 521-532.

Noda, M., Ko, M., Ogura, A., Liu, D-G., Amano, T., Takano, T., & Ikawa, Y. (1985). Sarcoma viruses carrying ras oncogenes induce differentiation-associated properties in a neuronal cell line. Nature, 318, 73-75.

Noritake, J., Fukata, Y., Iwanaga, T., Hosomi, N., Tsutsumi, R., Matsuda, N., Tani, H., Iwanari, H., Mochizuki, Y., Kodama, T., et al. (2009). Mobile DHHC palmitoylating enzyme mediates activity-sensitive synaptic targeting of PSD-95. The Journal of Cell Biology, 186, 147-160.

O’Dowd, B.F., and Hnatowich, M., Caron, M.G., Lefkowitz, R.J. & Bouvier, M. (1989). Palmitoylation of the human beta 2-adrenergic receptor. Mutatino of Cys341 in the carboxyl tail leads to an uncoupled nonpalmitoylated form of the receptor. J Biol Chem., 264(13), 7564- 7569.

Ohno, Y., Kashio, A., Ogata, R., Ishitomi, A., Yamazaki, Y., & Kihara, A. (2012). Analysis of substrate specificity of human DHHC protein acyltransferases using a yeast expression system. Mol. Biol. Cell., 23, 4543-4551.

Ohno, Y., Kihara, A., Sano, T., and Igarashi, Y. (2006). Intracellular localization and tissue- specific distribution of human and yeast DHHC cysteine-rich domain-containing proteins. Biochimica et Biophysica Acta., 1761, 474-483.

Ohyama, T., Verstreken, P., Ly, C.V., Rosenmund, T., Raja, A., Tien, A-C., Haueter, C., Schulze, K., and Bellen, H.J. (2007). Huntingtin-interacting protein 14, a palmitoyl transferase required for exocytosis and targeting of CSP to synaptic vesicles. J Cell Biol., 179(7), 1481- 1496.

Omerovic, J. and Prior, I.A. (2009). Compartmentalized signalling: Ras proteins and signalling nanoclusters. FEBS J., 276(7), 1817-1825.

Ori-McKenney, K.M., Jan, L.Y., and Jan, Y.N. (2012). Golgi outposts shape dendrite morphology by functioning as sites of acentrosomal microtubule nucleation in neurons. Neuron, 76(5), 921-930.

Papadopoulous, T., Schemm, R., Grubmüller, H., and Brose, N. (2015). Lipid binding defects and perturbed synaptogenic activity of a collybistin R290H mutant that causes epilepsy and intellectual disability. J Biol Chem., 290 (13), 8256-8270.

Pedro, M.P., Vilcaes, A.A., Gomez, G.A. and Daniotti, J.L. (2017). Individual S-acylated cysteines differentially contribute to H-Ras endomembrane trafficking and acylation/deacylation cycles. Mol Cell Biol., 28(7), 962-974. 179

Peitzsch, R.M. & McLaughlin, S. (1993). Binding of acylated peptides and fatty acids to phospholipid vesicles: pertinence to myristoylated proteins. Biochemistry, 32, 10436-10443.

Percher, A., Ramakrishna, S., Thinon, E., Yuan, X., Young, J.S., & Hang, H.C. (2016). Mass- tag labeling reveals site-specific and endogenous levels of protein S-fatty acylation. PNAS, 113, 4302-4307.

Peters, A., & Kaiserman-Abramof, I.R. (1970). The small pyramidal neuron of the rat cerebral cortex. The perikaryon, dendrites, and spines. Am J Anat, 127, 321-355.

Pierce, J.P., Mayer, T., and McCarthy, J.B. (2001). Evidence for a satellite secretory pathway in neuronal dendritic spines. Curr Biol., 11(5), 351-355.

Piton, A., Redin, C., and Mandel, J.L. (2013). XLID-causing mutations and associated genes challenged in light of data from large-scale human exome sequencing. American Journal of Human Genetics, 93, 368-383.

Plowman, S.J. and Hancock, J.F. (2005). Ras signaling from plasma membrane and endomembrane microdomains. Biochim Biophys Acta., 1746(3), 274-283.

Politis, E.G., Roth, A.F., & Davis, N.G. (2005). Transmembrane topology of the protein palmitoyl transferase Akr1. J. Biol. Chem., 280, 10156-10163.

Polo-Parada L, Bose CM, & Landmesser LT. (2001). Alterations in transmission, vesicle dynamics and transmitter release machinery at NCAM-deficient neuromuscular junctions. Neuron, 32, 815–828. Pommereit, D., and Wouters, F.S. (2007). An NGF-induced Exo70 complex that locally antagonises Cdc42-mediated activation of N-WASP to modulate neurite outgrowth. J Cell Sci., 120, 2694-2670.

Ponimaskin, E., Dityateva, G., Ruonala, M.O., Fukata, M., Fukata, Y., Kobe, F., Wouters, F.S., Delling, M., Bredt, D.S., Schacher, M., and Dityatey, A. (2008). Fibroblast growth factor- regulated palmitoylation of the neural cell adhesion molecule determines neuronal morphogenesis. J Neurosci., 28(36), 8897-8908.

Powers, S., Kataoka, T., Fasano, O., Goldfarb, M., Strathern, J., Broach, J. & Wigler, M. (1984). Genes in S. cerevisiae encoding proteins with domains homologous to the mammalian ras proteins. Cell, 36, 607-612.

Prange, O., Wong, T.P., Gerrow, K., Wang, Y.T., & El-Husseini, A. (2004). A balance between excitatory and inhibitory synapses is controlled by PSD-95 and neuroligin. PNAS, 101, 13915-13920.

180

Prescott, G.R., Gorleku, O.A., Greaves, J.,and Chamberlain, L.H. (2009). Palmitoylation of the synaptic vesicle fusion machinery. J Neurochem., 110(4), 1135-1149. doi:10.1111/j.1471- 4159.2009.06205.x.

Prior, I.A., Lewis, P.D., & Mattos, C. (2012). A comprehensive survey of Ras mutations in cancer. Cancer Res., 72, 2457-2467.

Pulver, A.E., Nestadt, G., Goldberg, R., Shprintzen, R.J., Lamacz, M., Wolyniec, P.S., Morrow, B., Karayiorgou, M., Antonarakis, S.E., & Housman, D. (1994). Psychotic illness in patients diagnosed with velo-cardio-facial syndrome and their relatives. J. Nerv. Ment. Dis., 182, 476–478.

Purkey, A.M., Woolfrey, K.M., Crosby, K.C., Stich, D.G., Chick, W.S., Aoto, J. & Dell’Acqua, M.L. (2018). AKAP150 palmitoylation regulates synaptic incorporation of Ca2+- permeable AMPA receptors to control LTP. Cell. Rep., 25, 974-987.

Purpura, D.P. (1974). Dendritic spine "dysgenesis" and mental retardation. Science, 186, 1126- 1128.

Purpura, D.P. (1975a). Dendritic differentiation in human cerebral cortex: normal and aberrant developmental patterns. Adv Neurol., 12, 91-116.

Purpura, D.P. (1975b). Normal and aberrant neuronal development in the cerebreal cortex of human fetus and young infant. UCLA Forum Med Sci, 18, 141-169.

Purves, D. and Hume, R.I. (1981). The relation of postsynaptic geometry to the number of presynaptic axons that innervate autonomic ganglion cells. J. Neurosci, 1(5), 441-52.

Purves, D., Hadley, R.D., and Voyvodic, J.T. (1986). Dynamic changes in the dendritic geometry of individual neurons visualized over periods of up to three months in the superior cervical ganglion of living mice. J. Neurosci, 6(4), 1051-1060.

Putilina, T., Wong,P., and Gentleman, S. (1999). The DHHC domain: a new highly conserved cysteine-rich motif. Mol. Cell. Biochem, 195, 219-226.

Qanbar, R., and Bouvier, M. (2003). Role of palmitoylation/depalmitoylation reactions in G- protein-coupled receptor function. Pharmacol Ther, 97(1), 1-33.

Qin, Y., Zhu, Y., Baumgart, J.P., Stornetta, R.L., Seidenman, K., Mack, V., van Aelst, L., and Zhu, J.J. (2005). State-dependent Ras signaling and AMPA receptor trafficking. Genes Dev., 19(17), 2000-2015.

Rall, W. (1995). The Theoretical Foundation of Dendritic Function - Selected Papers of Wilfrid Rall with Commentaries. (Cambridge, MA: MIT Press)

181

Rammamorthi, K., & Lin, Y. (2011). The contribution of GABAergic dysfunction to neurodevelopmental disorders. Trends Mol. Med., 17, 452-462.

Ramon y Cajal, S. (1888). Estructura de los centros nerviosos de las aves. Rev. Trim. Hitol. Norm. Pat. Finger, S. (Ed). Reviewed in “Minds behind the brain – a history of pioneers and their discoveries”, Oxford University Press, Oxford, 200-211.

Rauen, K.A. The RASopathies. (2013). Annu. Rev. Genomics Hum. Genet., 14, 355-369.

Raymond, F.L., Tarpey, P.S., Edkins, S., Tofts, C., O’Meara, S., Teague, J., Butler, A., Stevens, C., Barthorpe, S., Buck G., et al. (2007). Mutations in ZDHHC9, which encodes a palmitoyltransferase of NRAS and HRAS, cause X-linked mental retardation associated with Marfanoid habitus. Am J Hum Genet., 80(5), 982-987.

Resh, M.D. (1996). Regulation of cellular signalling by fatty acid acylation and prenylation of sigal transduction proteins. Cell Signal., 8(6), 403-412.

Resh, M.D. (1999). Fatty acylation of proteins: new insights into membrane targeting of myristoylated and palmitoylated proteins. Biochim Biophys Acta., 1451(1), 1-16.

Resh, M.D. (2013). Covalent lipid modifications of proteins. Current Biology, 23, 431-435.

Rizo, J. & Rosenmund, C. (2008). Synaptic vesicle fusion. Nat Struct Mol Biol, 15, 665-674.

Rizzoli, S.O. & Betz, W.J. (2005). Synaptic vesicle pools. Nat Rev Neurosci, 6, 57-69.

Rocks, O., Gerauer, M., Vartak, N., Koch, S., Huang, Z.P., Pechlivanis, M., Kuhlmann, J., Brunsveld, L., Chandra, A., Ellinger, B., et al. (2010). The palmitoylation machinery is a spatially organizing system for peripheral membrane proteins. Cell, 141, 458-471.

Rocks, O., Peyker, A., Kahms, M., Verveer, P.J., Koerner, C., Lumbierres, M., Kuhlmann, J., Waldmann, H., Wittinghofer, A., and Bastiaens, P.I. (2005). An acylation cycle regulates localization and activity of palmitoylated Ras isoforms. Science, 307(5716), 1746-1452.

Rodriguez, A., Ehlenberger, D.B., Hof, P.R., and Wearne, S.L. (2006). Rayburst sampling, an algorithm for automated three-dimensional shape analysis from laser scanning microscopy images. Nature Protocols, 1, 2152-2161.

Ropers, H.H. and Hamel, B.C. (2005). X-linked mental retardation. Nat Rev Genet., 6(1), 46- 57.

Rose, C.R. and Konnerth, A. (2001) Stores not just for storage: intracellular calcium release and synaptic plasticity. Neuron, 31, 519–522.

Rosso, S.B., Sussman, D., Wynshaw-Boris, A., and Salinas, P.C. (2005). Wnt signaling 182

through Dishevelled, Rac, and JNK regulates dendritic development. Nat Neurosci., 8(1), 34- 42.

Rubinson, D.A., Dillon, C.P., Kwiatkowski, A.V., Sievers, C., Yang, L., Kopinja, J., Rooney, D.L., Zhang, M., Ihrig, M.M., McManus, M.T., et al. (2003). A lentivirus-based system to functionally silence genes in primary mammalian cells, stem cells and transgenic mice by RNA interference. Nature Genetics, 33, 401-406.

Rudy, B., Fishell, G., Lee, S., and Hjerling-Leffler, J. (2011). Three groups of interneurons account for nearly 100% of neocortical GABAergic neurons. Dev Neurobiol., 71(1), 45-61.

Rusch, M., Zimmerman, T.J., Burger, M., Dekker, F.J., Gormer, K., Triola, G., Brockmeyer, A., Janning, P., Bottcher, P., Sieber, S.A., et al. (2011). Identification of acylprotein thioesterases 1 and 2 as the cellular targets of the Ras-signaling modulators palmostatin B and M. Agnew. Chemie., 50, 9838-9842.

Sabatini, B.L. & Svoboda, K. (2000). Analysis of calcium channels in singe spines using optical fluctuation analysis. Nature, 408, 589-593.

Sala, C., & Segal, M. (2014). Dendritic spines: the locus of structural and functional plasticity. Physiol. Rev., 94, 141-188.

Salaun, C., Greaves, J., & Chamberlain, L.H. (2010). The intracellular dynamic of protein palmitoylation. J. Cell. Biol., 191, 1229-1238.

Sanders, S.S. & Hayden, M.R. (2015). Aberrant palmitoylation in Huntington disease. Biochem. Soc. Trans., 43, 205-210.

Sanders, S.S., Martin, D.D., Butland, S.L., Lavallee-Adam, M., Calzolari, D., Kay, C., Yates, J.R., 3rd, and Hayden, M.R. (2015). Curation of the Mammalian Palmitoylome Indicates a Pivotal Role for Palmitoylation in Diseases and Disorders of the Nervous System and Cancers. PLoS Comput Biol, 11, e1004405.

Saneyoshi, T. & Hayashi, Y. (2012). The Ca2+ and rho gtpase signaling pathways underlying activity-dependent actin remodeling at dendritic spines. Cytoskeleton, 69, 545–554.

Schmidt, J.W. and Catterall, W.A. (1987). Palmitoylation, sulfation, and glycosylation of the alpha subunit of the sodium channel. Role of post-translational modifications in channel assembly. J Biol Chem, 262, 13713-13723.

Schmidt, M.F., & Schlesinger, M.J. (1979). Fatty acid binding to vesicular stomatitis virus glycoprotein: a new type of post-translational modification of the viral glycoprotein. Cell, 17, 813-819.

Scott, E.K. & Luo, L. (2001). How do dendrites take their shape? Nat. Neurosci., 4, 359-365. 183

Scoville, W.B. & Milner, B. (1957). Loss of recent memory after bilateral hippocampal lesions. J. Neurol. Neurosurg. Psychiatry, 20, 11-21.

Seeger, G., Gartner, U., & Arendt, T. (2004). Transgenic activation of Ras in neurons increases synapse formation in mouse neocortex. J. Neural Transm. (Vienna), 112, 751-761.

Sharma, C., Yang, X.H., and Hemler, M.E. (2008). DHHC2 affects palmitoylation, stability, and functions of tetraspanins CD9 and CD151. Molecular Biology of the Cell, 19, 3415-3425.

Shen, Y., Specht, S.M., De Saint Ghislain, I. Li, R. (1994). The hippocampus: a biological model for studying learning and memory. Prog Neurobiol, 44, 485-496.

Shimojima, K., Sugawara, M., Shichiji, M., Mukaida, S., Takayama, R., Imai, K., and Yamamoto, T. (2011). Loss-of-function mutation of collybistin is responsible for X-linked mental retardation associated with epilepsy. J Hum Genet. 56(8), 561-565.

Shipston, M.J. (2013). Regulation of large conductance calcium- and voltage-activated potassium (BK) channels by S-palmitoylation. Biochem Soc Trans, 41(1), 67-71.

Sigel, E., & Steinmann, M.E. (2012). Structure, function, and modulation of GABAA receptors. Journal of Biological Chemistry, 287, 40224-40231.

Simanshu, D.K., Nissley, D.V., & McCormick, F. (2017). RAS proteins and their regulators in human disease. Cell, 170, 17-33.

Singaraja, R.R., Hadano, S., Metzler, M., Givan, S., Wellington, C.L., Warby, S., Yanai, A., Gutekunst, C.A., Leavitt, B.R., Yi, H., et al. (2002). HIP14, a novel ankyrin domain-containing protein, links huntingtin to intracellular trafficking and endocytosis. Hum. Mol. Genet., 11, 2815-2828.

Sletten, E.M. & Bertozzi, C.R. (2011). From mechanism to mouse: a tale of two biorthogonal reactions. Acc Chem Res, 44, 666-676.

Sohal, V.S. & Rubenstein, J.L.R. (2019). Excitation-inhibition balance as a framework for investigating mechanisms in neuropsychiatric disorders. Molecular Psychiatry, 2019. Epub ahead of print.

Song, S.P., Hennig, A., Schubert, K., Markwart, R., Schmidt, P., Prior, I.A., Böhmer, F.D., and Rubio, I. (2013). Ras palmitoylation is necessary for N-Ras activation and signal propagation in growth factor signalling. Biochem J., 454(2), 323-32.

Stevenson, R.E. & Schwartz, C.E. (2009). X-linked intellectual disability: unique vulnerability of the male genome. Dev Disabil Res Rev, 15, 361-368.

184

Steward, O., Davis, L., Dotti, C., Phillips, L.L., Rao, A., & Banker, G. (1988). Protein synthesis and processing in cytoplasmic microdomains beneath postsynaptic sites on CNS neurons. Mol Neurobiol, 2, 227-261.

Stowers, R.S., & Isacoff, E.Y. (2007). Drosophila huntingtin-interacting protein 14 is a presynaptic protein required for photoreceptor synaptic transmission and expression of the palmitoylated proteins synaptosome-assocated protein 25 and cysteine string protein. J Neurosci, 27, 12874-12883.

Straino, P., and Zara, F. (2017). ARHGEF9 mutations cause a specific recognizable X-linked intellectual disability syndrome. Neurol Genet., 3(3), e159.

Sudhof, T.C. (2008). Neuroligins and neurexins link synaptic function to cognitive disease. Nature, 455, 903-911.

Sun, Y., and Bamji, S.X. (2011). beta-Pix modulates actin-mediated recruitment of synaptic vesicles to synapses. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 31, 17123-17133.

Swann, J.W., Al-Noori, S., Jiang, M., and Lee, C.L. (2000). Spine loss and other dendritic abnormalities in epilepsy. Hippocampus, 10, 617-625.

Swarthout, J.T., Lobo, S., Farh, L., Croke, M.R., Greentree, W.K., Deschenes, R.J. and Linder, M.E. (2005). DHHC9 and GCP16 constitute a human protein fatty acyltransferase with specificity for H- and N-Ras. J Biol Chem., 280(35), 31141-31148.

Takai, Y., Sasaki, T., & Matozaki, T. Small GTP-binding proteins. (2001). Phys. Rev., 81, 153-208.

Takamori, S. (2006). VGLUTs: ‘exciting’ times for glutamatergic research? Neuroscience Research, 55, 343-351.

Takamori, S., Rhee, J.S., Rosenmund, C., & Jahn, R. (2000). Identification of a vescicular glutamate transporter that defines a glutamatergic phenotype in neurons. Nature, 407, 189-194.

Tanabe, K., Tachibana, T., Yamashita, T., Che, Y.H., Yoneda, Y., Ochi, T., Tohyama, M., Yoshikawa, H., & Kiyama, H. (2000). The small GTP-binding protein TC10 promotes nerve elongation in neuronal cells, and its expression is induced during nerve regeneration in rats. J. Neurosci., 20, 4138-4144.

The Huntington's Disease Collaborative Research Group. (1993). A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington's disease chromosomes. Cell, 72, 971–983.

185

Thomas, G.M., Hayashi, T., Chiu, S.L., Chen, C.M and Huganir, R.L. (2012). Palmitoylation by DHHC5/8 targets GRIP1 to dendritic endosomes to regulate AMPA-R trafficking. Neuron, 73(3), 482-496.

Thomas, G.M. & Huganir, R.L. (2004). MAPK cascade signalling and synaptic plasticity. Nat. Rev. Neurosci., 5, 173-183.

Tian, L., Jeffries, O., McClafferty, H., Molyvdas, A., Rowe, I.C., Saleem, F., Chen, L., Greaves, J., Chamberlain, L.H, Knaus, H.G. et al. (2008). Palmitoylation gates phosphorylation-dependent regulation of BK potassium channels. Proc Natl Acad Sci U.S.A., 105(52), 21006-21011.

Togashi H, Abe K, Mizoguchi A, Takaoka K, Chisaka O, & Takeichi M. (2002). Cadherin regulates dendritic spine morphogenesis. Neuron, 35, 77–89.

Tomatis, V.M., Trenchi, A., Gomez, G.A., & Daniotti, J.L. (2010). Acyl-protein thioesterase 2 catalyzes the deacylation of peripheral membrane-associated GAP-43. PLoS One, 5, e15045.

Topinka, J.R. & Bredt, D.S. (1998). N-terminal palmitoylation of PSD-95 regulates association with cell membranes and interaction with K+ channel Kv1.4. Neuron, 20(1), 125-134.

Torre, E.R., & Steward, O. (1996). Protein synthesis within dendrites: glycosylation of newly synthesized proteins of hippocampal neurons in culture. J. Neurosci., 16, 5967-5978.

Tsutsumi, R., Fukata, Y., Noritake, J., Iwanaga, T., Perez, F., and Fukata, M. (2009). Identification of G protein alpha subunit-palmitoylating enzyme. Mol Cell Biol, 29, 435-447.

Turrigiano, G.G., Leslie, K.R., Desai, N.S., Rutherford, L.C., & Nelson, S.B. (1998). Activity- dependent scaling of quantal amplitude in neocortical neurons. Nature, 391, 892-896.

Turrigiano, G.G. & Nelson, S.B. (2004). Homeostatic plasticity in the developing nervous system. Nat. Rev., 5, 97-107.

Tzschach, A., Grasshoff, U., Beck-Woedl, S., Dufke, C., Bauer, C., Kehrer, M., Evers, C., Moog, U., Oehl-Jaschkowitz, B., Di Donato, N.et al. Next-generation sequencing in X-linked intellectual disability. Eur J Hum Genet., 23(11), 1513-1518.

Ueno, K. (2000). Involvement of fatty acid synthase in axonal development in mouse embryos. Genes Cells, 5, 859–869.

Uhlén M, Fagerberg L, Hallström B.M., Lindskog C, Oksvold P, Mardinoglu A, Sivertsson Å, Kampf C, Sjöstedt E, Asplund A, et al. (2015). Tissue-based map of the human proteome. Science, 347(6220), 1260419.

186

Urbanska, M., Blazejczky, M., & Jaworski, J. (2008). Molecular basis of dendritic arborization. Acta. Neurobiol. Exp., 68, 264-288.

Vaillant, A.R., Zanassi, P., Walsh, G.S., Aumont, A., Alonso, A., and Miller, F.D. Signaling mechanisms underlying reversible, activity-dependent dendrite formation. Neuron, 34(6), 985- 998.

Valnegri, P., Sala, C., and Passafaro, M. (2012). Synaptic dysfunction and intellectual disability. Adv Exp Med Biol., 970, 433-449. van Elburg, R.A.J. & van Ooyen, A. (2010). Impact of dendritic size and dendritic topology on burst firing in pyramidal cells. PLS Comput Biol, 6, e1000781.

Vaughn, J.E. (1989). Fine structure of synaptogenesis is the vertebrate central nervous system. Synapse, 3.

Veit, M., Becher, A., and Ahnert-Hilger, G. (2000). Synaptobrevin 2 is palmitoylated in synaptic vesicles prepared from adult, but not from embyronic brain. Mol Cell Neurosci., 15(4), 408-416.

Veit, M., and Schmidt, M.F.G. (2006). Palmitoylation of influenza virus proteins. Berl. Munch. Tierarztl. Wochenschr. 119, 112–122.

Veit, M., Söllner T.H., and Rothman, J.E. (1996). Multiple palmitoylation of synaptotagmin and the t-SNARE SNAP-25. FEBS Lett., 385(1-2), 119-123.

Vissers, L.E.L.M., Gilissen, C., & Veltman, J.A. (2016). Genetic studies in intellectual disability and related disorders. Nature Reviews Genetics, 17, 9-19.

Vetrivel, K.S., Meckler, X., Chen, Y., Nguyen, P.D., Seidah, N.G., Vassar, R., Wong, P.C., Fukata, M., Kounnas, M.Z., and Thinakaran, G. (2009). Alzheimer disease Abeta production in the absence of S-palmitoylation-dependent targeting of BACE1 to lipid rafts. J Biol Chem, 284, 3793-3803.

Waites, C.L., Craig, A.M., & Garner, C.C. (2005). Mechanisms of vertebrate synaptogenesis. Annu. Rev. Neurosci., 28, 251-274.

Wang, F., Chen, X., Shi, W., Yao, L., Gao, M., Yang, Y., and Hao, A. (2015). Zdhhc15b Regulates Differentiation of Diencephalic Dopaminergic Neurons in zebrafish. J Cell Biochem 116, 2980-2991.

Watson, R.T., Furukawa, M., Chiang, S-H., Boeglin, D., Kanzaki, M., Saltiel, A.R., and Pessin, J.E. (2003). The exocytotic trafficking of TC10 occurs through both classical and nonclassical secretory transport pathways in 3T3L1 adipocytes. Mol Cell Biol., 23(3), 961-974.

187

Weber, T., Zemelman, B.V., McNew, J.A., Westermann, B., Gmachl, M., Parlati, F., Sollner, T.H, & Rothman, J.E. (1998). SNAREpins: minimal machinery for membrane fusion. Cell, 92, 759-772.

Wen, Q. & Chklovskii, D.B. (2008). A cost-benefit analysis of neuronal morphology. J.Neurophsiol., 99, 2320-2328.

Wheelock, M.J. & Johnson, K.R. (2003). Cadherins as modulators of cellular phenotype. Annu Rev Cell Dev Biol, 19, 207-235.

Williams, M.E., Wilke, S.A., Daggett, A., Davis, E., Otto, S., Ravi, D., Ripley, B., Bushong, E.A., Ellisman, M.H., Klein, G., & Ghost, A. (2011). Cadherin-9 regulates synapse specific differentiaton in the developing hippocampus. Neuron, 71, 640-655.

Williams, S.R. & Stuart, G.J. (2000). Action potential backpropogation and somato-dendritic distribution of ion channels in thalamocortical neurons. J. Neurosci, 20, 1307-1317.

Woolfrey, K.M., Sanderson, J.L., and Dell’Acqua, M.L. (2015). The palmitoyl acyltransferase DHHC2 regulates recycling endosome exocytosis and synaptic potentiation through palmitoylation of AKAP79/150. J Neurosci., 35(2), 442-456.

Woolfrey, K.M. & Srivastava, D.P. (2016). Control of dendritic spine morphological and functional plasticity by small GTPases. Neural Plasticity, 2016, 3025948, 12 pages.

Woolfrey, K.M., O’Leary, H., Goodell, D.J., Robertson, H.R., Horne, E.A., Coultrap, S.J., Dell’Acqua, M.L., & Bayer, K.U. (2018). CamKII regulates the depalmitoylation and synaptic removal of the scaffold protein AKAP79/150 to mediate structural long-term depression. J. Biol. Chem., 293, 1551-1567.

Wu, G.Y, Zou, D.J., Rajan, I., & Cline, H. (1999). Dendritic dynamics in vivo change during neuronal maturation. J. Neurosci., 19, 4472-4483.

Wu, Q., & Maniatis, T. (1999). A striking organization of a large family of human neural cadherin-like cell adhesion genes. Cell, 97, 779-790.

Yanai, A., Huang, K., Kang, R., Singaraja, R.R., Arstikaitis, P., Gan, L., Orban, P.C., Mullard, A., Cowan, C.M., Raymond, L.A., Drisdel, R.C., Green, W.N., Ravikumar, B., Rubinsztein, D.C., El-Husseini, A., & Hayden, M.R. (2006). Palmitoylation of huntingtin by HIP14 is essential for its trafficking and function. Nat Neurosci, 9, 824-831.

Ye., X. & Carew, T.J. (2010). Small G protein signaling in neuronal plasticity and memory formation: the specific role of ras family proteins. Neuron, 68, 340-361.

Ye, B., Zhang, Y., Song, W., Younger S.H., Jan, L.Y., and Jan, N.Y. (2007). Growing dendrites and axons differ in their reliance on the secretory pathway. Cell 130(4), 717-729. 188

Yokoi, N., Fukata, Y., Sekiya, A., Murakami, T., Kobayashi, K., and Fukata, M. (2016). Identification of PSD-95 Depalmitoylating Enzymes. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 36, 6431-6444.

Yokoi, N., Fukata, Y., Sekiya, A., Murakami, T., Kobayashi, K, & Fukata, M. (2016). Identification of PSD-95 palmitoylating enzymes. J Neurosci, 36, 6431-6444.

Yu, W., & De Blas, A.L. (2008). Gephyrin expression and clustering affects the size of glutamatergic synaptic contacts. J. Neurochem, 104, 830-845.

Yu, W., Jiang, M., Miralles, C.P., Li, R.W., Chen, G., & De Blas, A.L. (2007). Gephyrin clustering is required for the stability of GABAergic synapses. Mol. Cell. Neurosci., 36, 484- 500.

Zeidman, R., Jackson, C.S. & Magee, A.I. (2009). Protein acyl thioesterases (Review). Mol. Membr. Biol., 26, 32-41.

Zhang, F.L. & Casey, P.J. (1996).Protein prenylation: molecular mechanisms and functional consequences. Annu. Rev. Biochem., 65, 241-269.

Zhang, W. & Benson, D.L. (2001). Stages of synapse development defined by dependence on Factin. J Neurosci, 21, 5169-5181.

Zhang, Y., Chen, K., Sloan, S.A., Bennett, M.L., Scholze, A.R., O'Keeffe, S., Phatnani, H.P., Guarnieri, P., Caneda, C., Ruderisch, N., et al. (2014). An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 34, 11929-11947.

Zhu, J.J., Qin, Y., Zhao, M., Van Aelst, L., and Malinow, R. (2002). Ras and Rap control AMPA receptor trafficking during synaptic plasticity. Cell, 110(4), 443-455.

Zhu, Y-C., Li, D, Wang, L., Lu, B., Zheng, J., Zhao, S-L., Zeng, R., and Xiong, Z-Q. (2013). Palmitoylation-dependent CDKL5-PSD-95 interactin regulates synaptic targeting of CDKL5 and dendritic spine development. Proc Natl Acad Sci U.S.A., 110(22), 9118-9123.

Žiburkus, J., Cressman, J.R., amd Schiff, S.J. (2013). Seizures as imbalanced up states: excitatory and inhibitory conductances during seizure-like events. J Neurophysiol., 109(5), 1296-1306.

Zou, D.-J., & Cline, H.T. (1999). Postsynaptic calcium/calmodulin-dependent protein kinase II is required to limit elaboration of presynaptic and postsynaptic neuronal arbors. J Neurosci,19, 8909-8918.

189