University of California, San Diego

UNIVERSITY OF CALIFORNIA, SAN DIEGO

Characterizing Postranslational Regulatory Mechanisms of the Ubiquitin Proteasome

System

A dissertation submitted in partial satisfaction of the

requirements for the degree of Doctor of Philosophy

Chemistry

Frankie Robert Gonzales

Committee in charge:

Professor Gentry Patrick, Chair Professor Elizabeth Komives, Co-Chair Professor Gourisankar Ghosh Professor Judy Kim Professor Dong-Er Zhang Professor Brian Zid

2017

Frankie Robert Gonzales, 2017

The dissertation of Frankie Robert Gonzales is approved, and it is acceptable in quality and form for publication on microfilm and electronically:

Co-Chair

Chair

University of California, San Diego 2017

iii

DEDICATION

I dedicate this work to my mom who has supported me in every aspect of my life. And to the countless educators who have taken special interest in my and helped advance me and

my career far further than I had ever expected. Mitch Markowitz, Dr. Jennifer Whiles-

Lillig, and Dr. Betsy Komives have been among these mentors who have propped me up

and I owe my entire educational path and career to them.

TABLE OF CONTENTS

Signature Page ...... iii

Dedication ...... iv

Table of Contents ...... v

List of Abbreviations ...... ix

List of Figures ...... x

List of Tables ...... xii

Acknowledgements ...... xiii

Vita ...... xiv

Abstract of the Dissertation ...... xvi

CHAPTER I: INTRODUCTION ...... 1

A. Protein degradation in eukaryotic cells ...... 2

B. Structure of the 26S proteasome...... 3

C. Functions of the ubiquitin proteasome system...... 4

D. Regulation of the UPS...... 6

E. Protein homeostasis at synapses...... 6

F. References ...... 9

CHAPTER II: INVESTIGATING PROTEASOME SUBUNIT RPT6 PHOSPHORYLATION DYNAMICS USING NOVEL MOUSE LINES THAT ABROGATE OR MIMIC PHOSPHORYLATION ...... 12

A. Abstract ...... 14

B. Introduction ...... 15

C. Materials and Methods ...... 17 Generation of S120A and S120D knock-in mice ...... 17

Antibodies and reagents ...... 17

Proteasome purification ...... 18

Peptidase and ATPase assays ...... 18

Western blot and native gel assays ...... 19

Slice Electrophysiology ...... 19

Whole-cell Electrophysiology ...... 21

Behavior ...... 21

Spine density, and immunohistochemistry ...... 22

Preparation and transfection of organotypic slice cultures ...... 22

Time-Lapse Imaging and Image Analysis for spine dynamics ...... 22

D. Results ...... 23 1. Generation, purification, and analysis of Rpt6 S120A and S120D mutant

proteasome composition and activity ...... 23

2. Abrogation of phosphorylation dynamics do not effect basal synaptic

transmission nor LTP paradigms of synaptic plasticity...... 25

3. Small changes in proteasome ATPase and peptidase kinetic activity have

no effect on fear conditioning, nor context discrimination paradigms of

learning and memory ...... 27

4. Elimination of phosphorylation dynamics do not alter basal spine density 28

5. Elimination of phosphorylation dynamics does not have a significant effect

on spine outgrowth ...... 29

E. Discussion ...... 30

E. References ...... 32

F. Figures and Figure Legends ...... 35

CHAPTER III: PROTEASOME PHOSPHORYLATION REGULATES COCAINE-INDUCED SENSITIZATION ...... 44

A. Abstract ...... 46

B. Introduction ...... 47

C. Materials and methods ...... 48 Generation of S120D Knockin mice ...... 48

Antibodies and Reagents ...... 49

Neuronal cultures ...... 49

Proteasome activity assays ...... 49

Western blot analysis ...... 50

qPCR ...... 50

Behavioral Sensitization ...... 51

Histology ...... 51

Data Analysis ...... 52

D. Results ...... 52 1. Cocaine increases Rpt6 S120 phosphorylation and proteasome activity ..... 52

2. Generation of Rpt6 S120D Knockin (KI) mice ...... 53

3. Rpt6 S120 phosphorylation and peptidase activity is increased in NAc and

PFC in cocaine treated wild type mouse brains, but not in S120D mutant

mice ...... 54

4. Behavioral sensitization is completely absent in Rpt6 S120D KI mutant

mice ...... 55

E. Discussion ...... 56

vii

F. Conclusions ...... 59

G. References ...... 59

CHAPTER IV: PHOSPHORYLATION OF THE 19S REGULATORY PARTICLE ATPASE SUBUNIT, RPT6, MODIFIES SUSCEPTIBILITY TO PROTEOTOXIC STRESS AND PROTEIN AGGREGATION ...... 68

A. Abstract ...... 69

B. Introduction ...... 70

C. Materials and methods ...... 73 Yeast strains and knock-in strategy ...... 73

Plasmids ...... 74

Immunoprecipitations and immunoblots ...... 74

Growth assays ...... 75

Viability assays ...... 75

Flow cytometry ...... 76

Bud profiling ...... 76

Native-PAGE peptidase and 26S fluorogenic peptidase activity assays ...... 76

D. Results ...... 77 1. Rpt6 is phosphorylated in S. cerevisiae in response to stress ...... 77

2. Decreased proteasome activity in the myc-rpt6 S119A mutant ...... 83

3. Altered Rpt6 phosphorylation exacerbates Htt polyQ aggregation and

proteotoxicity ...... 84

4. Compromised senescence in myc-rpt6-S119A cells expressing Htt103Q ..... 87

E. Discussion ...... 89

F. References…………...……...………………………………………………...…...... 94

viii

LIST OF ABBREVIATIONS

AMC Amino-Methyl Coumarin

ATP Adenosine Triphosphate

CaMKII Calcium Calmodilin Kinase II

DTT Dithiothreitol

EDTA Ethylenediaminetetraacetic Acid

LTD Long Term Depression

LTP Long Term Potentiation mEPSC mini Excitatory Postsynaptic Current

PB Phosphate Buffer

PFA Paraformaldehyde

UBL Ubiquitin-like domain

UIM Ubiquitin Interacting Motif

UPS Ubiquitin Proteasome System

WT Wild Type

LIST OF FIGURES

Figure 1.1: A graphical representation of the 26S proteasome ...... 3

Figure 1.2: Atomic structure of the base and lid of the 19S regulatory particle in complex with the 20S core particle. (Huang, Luan, Wu, & Shi, 2016) ...... 4

Figure 2.1: Proteasome protein degradation and ATP hydrolysis kinetics are significantly modulated in Rpt6 S120A and S120D knock-in mice ...... 35

Figure 2.2: Basal synaptic transmission and LTP are unaltered in Rpt6 S120A and S120D knock-in mice...... 36

Figure 2.3a: Context discrimination and Fear Memory were unaffected in Rpt6 S120A and S120D knock-in mice ...... 37

Figure 2.3b: Immediate, context, and cued memory were unaffected in Rpt6 S120A and S120D knock-in mice ...... 38

Figure 2.4: Spine density is not altered in S120 knock-in mice ...... 39

Figure 2.5: New spine generation was not altered in Rpt6 S120A mouse neurons ...... 40

Figure 2.S1. Generation of Rpt6 S120D Knockin (KI) mice ...... 41

Figure 2.S2: Performance on the elevated plus maze was not impaired in S120 knock-in mice ...... 42

Figure 3.1a: Cocaine increases Rpt6 S120 phosphorylation and proteasome activity via the dopaminergic pathway ...... 62

Figure 3.1b: Cocaine increases Rpt6 S120 phosphorylation and proteasome activity via the dopaminergic pathway ...... 63

Figure 3.2: Generation of Rpt6 S120D Knockin (KI) mice ...... 64

Figure 3.3: Cocaine increases Rpt6 phosphorylation and proteasome activity in NAc and PFC in wildtype but not Rpt6 S120D KI mutant mice ...... 65

Figure 3.4: Disruption of locomotor sensitization in S120D mice ...... 66

Figure 4.1: The rpt6-S119A mutants had increased susceptibility to proteotoxic stress and displayed abnormal morphology...... 80

Figure 4.2: Proteasome activity was decreased in the myc-rpt6-S119A mutant ...... 86

Figure 4.3: Altered Rpt6 phosphorylation and proteasome function exacerbated Htt polyQ aggregation and proteotoxicity ...... 88

Figure 4.4: Senescent myc-rpt6-S119A cells expressing Htt103Q had a delayed loss of viability relative to myc-RPT6 upon chronological aging ...... 91

LIST OF TABLES

Table S2.1: Proteasome interacting proteins...... 43

xii

ACKNOWLEDGEMENTS

Chapter II, in part, is a reprint that the dissertation author was the principal researcher and author of. Gonzales FR and Patrick GN, (2017). Investigating protesome subunit Rpt6 phosphorylation dynamics using novel mouse lines that abrogate or mimic phosphorylation. (in press .)

Chapter III, in part, is a reprint that the dissertation author was the principal researcher and author of. Gonzales FR, Howell KK, Dozier LE, Anagnostaras SG,

Patrick GN. (2017). Molecular and Cellular Neuroscience . (in press)

Chapter IV, in full, is a reprint that the dissertation author was the second researcher and author of. The material appears in PLOS ONE . Marquez-Lona E.M.,

Torres-Machorro A.L., Gonzales F.R., Pillus L., Patrick G.N. (2017) Phosphorylation of the 19S regulatory particle ATPase subunit, Rpt6, modifies susceptibility to proteotoxic stress and protein aggregation. PLOS ONE 12(6): e0179893

xiii

VITA

2009 Bachelors of Science, Chemistry, Sonoma State University, Rohnert Park, Ca

2009 Bachelors of Science, Biology, Sonoma State University, Rohnert Park, Ca

2014 Masters of Science, Chemistry, University of California, San Diego, San Diego,

2017 Doctor of Philosophy, Chemistry, University of California, San Diego, San

Diego, Ca

PUBLICATIONS

Gonzales, F.R., Scudder S.L., Howell, K.K., Dozier, L.E., Anagnostaras, S.G., Patrick, G.N. (2017). Investigating Protesome subunit Rpt6 phosphorylation dynamics using novel mouse lines that abrogate or mimic phosphorylation. (in submission)

Gonzales, F.R., Howell, K.K., Dozier, L.E., Anagnostaras, S.G., Patrick, G.N. (2017). Proteasome phosphorylation regulates cocaine-induced sensitization. J. Mol. Cell. Neuroscience., 2017.

Marquez-Lona E.M., Torres-Machorro A.L., Gonzales F.R., Pillus L., Patrick G.N. (2017) Phosphorylation of the 19S regulatory particle ATPase subunit, Rpt6, modifies susceptibility to proteotoxic stress and protein aggregation. PLOS ONE 12(6): e0179893

FIELDS OF STUDY

Major Field: Biochemistry Studies in Biochemistry and Neurobiology Professor Gentry Patrick

HONORS AND AWARDS

xiv

2001 Stanley Miller Award, Department of Chemistry, UCSD, San Diego, Ca.

2009 B.S., Chemistry with honors, Sonoma State University, Rohnert Park, Ca.

ABSTRACT OF THE DISSERTATION

Proteasome Regulation by Rpt6 Phosphorylation

Frankie Robert Gonzales

Doctor of Philosophy in Chemistry

University of California, San Diego, 2017

Professor Gentry Patrick, Chair

Professor Elizabeth Komives, Co-Chair

Protein homeostasis in is critical to maintain cell health and viability. Protein homeostasis can be divided into two major categories: protein synthesis, and protein degradation. One of the major pathways of protein degradation is via the Ubiquitin

Proteasome System (UPS), which, at its core, involves the proteasome; a multi-subunit cellular machine that unfolds and degrades proteins and breaks them down into small peptides to be utilized by the cell for further protein synthesis. Regulation of this machinery is vital to maintain protein homeostasis. We have discovered a posttranslational modification to the AAA ATPase Rpt6 subunit that modulates proteasome activity. Here, we have made mutants that abrogate or mimic this phosphorylation, in order to allow us to further uncover the regulatory mechanisms that

xvi

govern protein breakdown and homeostasis in eukaryotic cells.

Chapter II investigates the basic cellular effects in the CNS of eliminating phosphorylation dynamics of Rpt6 in novel mouse models that either block Rpt6 phosphorylation or mimic it. Behavioral, electrophysiological, biochemical, and other techniques are utilized in investigating the effects of Rpt6 S120 phosphorylation in the mammalian brain.

In Chapter III, the loss of cocaine sensitization in Rpt6 S120D mice is explored.

We find that treatment of cultured neurons with cocaine causes the phosphorylation and activation of CaMKIIα, and concomitant Rpt6 phosphorylation and subsequent increase in proteasome activity. This process is shown to be activated via monaminergic systems in neurons, specifically the inhibition of dopamine uptake. This process is abrogated when Rpt6 S120 phosphorylation is mimicked when mutated to aspartic acid, particularly in the nucleus accumbens and pre-frontal cortex, providing further insight into the pathogenesis of cocaine sensitization.

Chapter IV presents a study of the dynamics Rpt6 phosphorylation and cellular susceptibility to proteotoxic stress. Rpt6 S120 knock-in mutants of S. Cerevisiae are utilized to investigate how cells respond to various proteotoxic stressors. We find Rpt6 is phosphorylated in response to such stressors, and when phosphorylation is abrogated, cells are not able to respond to such stresses and become more susceptible to protein aggregation and cell death.

xvii

Chapter I: Introduction

A. Protein degradation in eukaryotic cells Proteins are the affecter molecules in eukaryotic cells, carrying out the duties encoded by the genome in order to maintain basic cellular functions. Protein homeostasis, the balance between protein synthesis and degradation, is critical in the maintenance of cell health and proliferation. Amino acids, the building blocks of proteins, are energetically costly and in some cases, impossible for cells to synthesize. Therefore, recycling proteins is a vital function in maintaining homeostasis. One major pathway in the degradation and recycling of proteins is the Ubiquitin Proteasome System (UPS).

The UPS utilizes a small protein tag, ubiquitin, that when covalently linked to proteins, flags them for degradation, trafficking, or recycling (Ciechanover, Heller, Elias,

Haas, & Hershko, 1980; Hershko, Ciechanover, Heller, Haas, & Rose, 1980). When destined for degradation, ubiquitinated proteins are broken down by the proteasome; a multi-subunit cellular machine with multiple proteolytic properties that break down proteins into short chain peptides that can be recycled for use in the synthesis of new proteins. The UPS rapidly degrades short-lived and misfolded proteins as well as the bulk breakdown of longer-lived structural and enzymatic proteins (Fagan, Waxman, &

Goldberg, 1987; Waxman, Fagan, & Goldberg, 1987).

The UPS consists of a series of ubiquitin conjugating enzymes, which modify proteins with ubiquitin. This modification can be singular (mono-ubiquitination), or can consist of two or more ubiquitin proteins, either in polyubiquitin chains, or in many mono-ubiquitin modifications at multiple sites on a protein. They type of ubiquitination often determines the fate of the modified protein. It was first thought that the rate of protein hydrolysis depended solely on the rate of ubiquitination of targeted proteins

(Chin, Kuehl, & Rechsteiner, 1982). However, it has become clear that modifications and proteasome-interacting-proteins (PIPs) can affect the enzymatic rate of the proteasome itself (Djakovic, Schwarz, Barylko, DeMartino, & Patrick, 2009; Guo, Huang, & Chen,

2017; Lee et al., 2010; Lokireddy, Kukushkin, & Goldberg, 2015; Myeku, Wang, &

Figueiredo-Pereira, 2012).

B. Structure of the 26S proteasome.

Figure 1.1: A graphical representation of the 26S proteasome

The 26S proteasome consists of a 20S core, where the proteolytic enzymatic activity takes place, and two regulatory 19S particles, located at either opening of the barrel-shaped 20S core particle. The barrel consists of two α rings and two β rings. The core particle is capable of three types of proteolytic activity: trypsin-like, chymotrypsin- like, and peptidyl-glutamyl peptide-hydrolyzing (PHGH), each with their own unique substrate specificities.

The 19S regulatory particle consists of a base composed of a heterohexameric ring of the AAA ATPases Rpt1-6. These subunits possess coiled tails on their N-termini which interact with the α-rings of the 20S core particle and allow for proper alignment and gate opening of the enzymatic core. These ATPases provide the mechanical energy necessary to remove ubiquitin, unfold substrates, and translocate them into the core particle for proteolysis. Above the heterohexameric ring are a series of Rpn subunits, including Rpn10 and Rpn13 (ubiquitin receptors), and Rpn11 (a de-ubiquitinating enzyme). Rpn9, Rpn5, Rpn6, Rpn7, Rpn3, and Rpn12 form a horseshoe shaped structure at the top of the lid and covers half of the Rpt subunits.

Figure 1.2: Atomic structure of the base and lid of the 19S regulatory particle in complex with the 20S core particle. (Huang, Luan, Wu, & Shi, 2016)

C. Functions of the ubiquitin proteasome system. The ubiquitin proteasome system modifies target proteins with ubiquitin, a small protein found in all eukaryotic cells. It was once believed that the UPS was merely a large trash can of the cell that ubiquitinated and degraded proteins nearly

5 indiscriminately. It is now know that the UPS is a highly developed and specific system that identifies proteins that need to be degraded, trafficked, or recycled based on cellular needs.

Ubiquitin modification does not always lead to substrate degradation. Based on the type of ubiquitin modification, the substrate faces multiple potential fates. Ubiquitin attaches via a terminal glycine residue attached to a lysine residue on the substrate. The addition of a single ubiquitin is known as monoubiquitination. However, polyubiquitination can occur by adding subsequent ubiquitins in a homotypic chain via various lysine residues in the ubiquitin sequence. One of the most common types of polyubiquitination that often leads to degradation is K48 ubiquitination. However, many other chain linkages have been identified, including K6 (DNA repair), K11/K27/K29

(degradation), K33 (stress response), and K63 (endocytosis/NFκB activation). These chains can be linear or branched and often the composition of these chains can determine the fate of a substrate (Ikeda & Dikic, 2008; Iwai & Tokunaga, 2009; Swatek &

Komander, 2016).

Ubiquitination of a substrate is carried out via a cascade involving three classes of ligases; E1, E2, and E3. E1 enzymes activate ubiquitin by forming a thiodiester bond between its active site and the ubiquitin c-terminal glycine, E2 ligases conjugate the ubiquitin, and E3 ligases act as scaffolds to transfer the ubiquitin to a substrate. The E3 ligases provide for substrate specificity.

D. Regulation of the UPS. It was previously believed that the rate limiting step for protein degradation via the UPS was the rate of substrate ubiquitination. It has become clear, however, that the enzymatic rate of proteasomal substrate degradation can be modulated in various ways.

Many proteasome interacting proteins (PIPs) have been identified that, when interacting with the proteasome, can change the rate of proteasome activity. Usp14 has been shown to inhibit proteasome function, even in mutants that eliminate its de-ubiquitinating activity (Lee et al., 2010). Similarly, Hul5 and Ecm29 are recruited to the proteasome in response to aberrant assembly or activity deficiency in a stress response to render them inactive (Park, Kim, Tian, Gygi, & Finley, 2011). Even free chains of ubiquitin have been observed to increase proteasome activity.

In addition to proteasome interacting proteins, post-translational modifications directly on proteasome substrates has been shown to regulate activity. It has been demonstrated that activated protein kinase A (PKA) can phosphorylate the regulatory subunit Rpn6 at serine 14 and enhance proteasome functions including substrate degradation and ATP hydrolysis. Indeed, phosphorylation of many 26S subunits are phosphorylated by a large range of kinases that have the ability to modulate proteasome activity (Guo et al., 2017). The discovery that proteasomes do not constantly operate at maximal capacity, and that modifications or protein interactions can regulate and activate proteasome activity represents a major pathway which can be perturbed as a research tool or used as novel therapies in disease scenarios, as many diseases are characterized by an imbalance between protein synthesis and degradation.

E. Protein homeostasis at synapses .

The mammalian brain contains billions of neurons that make trillions of synaptic connections. These connections are highly dynamic, constantly forming and breaking connections based on environmental inputs. This dynamic response to changes is known as synaptic plasticity, and is crucial in maintaining cell viability, learning and memory.

This constant remodeling of synapses means that protein turnover is especially important.

What’s more, is that synapses can be long distances away from the cell body, therefore the need for local protein synthesis and degradation is vital.

The ubiquitin proteasome system is critical in maintaining protein homeostasis and therefore synaptic plasticity in neurons, as Ehlers et al found that large cohorts of synaptic proteins have their half-lives controlled by the UPS (Ehlers, 2003). It has been shown that the use of proteasome inhibitors like lactacystin can abrogate memory formation in live animals (Lopez-Salon et al., 2001). Furthermore, the use of proteasome inhibitors on rat hippocampal slices prevents these neurons from being able to maintain induced long term potentiation (LTP) or long term depression (LTD), two paradigms essential in the formation of long and short term memory formation, indicating that proteasome activity can have major consequences in synaptic activity (Fonseca, Vabulas,

Hartl, Bonhoeffer, & Nagerl, 2006; Hou et al., 2006). Conversely, proteasomal activity has been shown to be regulated by synaptic activity. When synaptic activity is stimulated with bicuculline, a competitive antagonist of GABAA receptors, proteasome activity is significantly increased. Proteasomes are also sequestered at synapses in response to increased synaptic activity (Bingol & Schuman, 2006), and the use of proteasome inhibitors in cultured neurons block the activity-dependent outgrowth of new spines

(Hamilton et al., 2012), indicating a tight relationship between neuronal activity and the functions of the UPS.

The entire UPS is critical in the maintenance of synapses, plasticity and memory formation. It has been demonstrated by various knock-outs and mutations to multiple components of the UPS that a disruption in the ubiquitination, de-ubiquitination, or degradation by the UPS causes an array of dysfunctions and diseases in the brain. For example, the E3 ligase E6AP has been implicated in a range diseases including cancers and neurological disorders such as Angelman’s Syndrome (El Hokayem & Nawaz,

2014). Along with the proteasome, many UPS components are also sequestered at synapses in response to increased synaptic strength. Nedd4-1, an E3 ubiquitin ligase, is diffuse under basal conditions in cultured hippocampal neurons, and becomes punctate in spines in response to AMPAR activation (Scudder et al., 2014). Finally, deubiquitinating enzymes (DUBs) are also closely linked to synaptic strength and remodeling. One substrate of Usp14, a proteasome-associated DUB, are GABAA receptors, and a mutation in Usp14 causes abnormal receptor turnover in the ataxia a J mouse, which have seizure and rapid muscular degeneration (Lappe-Siefke et al., 2009)

The UPS in involved in both structural remodeling as well as composition of surface receptors and down-stream signaling components. Under AMPAR stimulation,

GluA1 receptors are ubiquitinated and their internalization and trafficking regulated via the endosome sorting complexes required for transport (ESCRTs) (Schwarz, Hall, &

Patrick, 2010) where they can either be recycled back to the surface to degraded.

Therefore, the UPS can quickly change the composition of surface receptors at spines in

9 response to stimuli, thus regulating synaptic strength and a feedback mechanism known as synaptic scaling.

F. References Bingol, B., & Schuman, E. M. (2006). Activity-dependent dynamics and sequestration of proteasomes in dendritic spines. Nature, 441 (7097), 1144-1148. doi:10.1038/nature04769

Chin, D. T., Kuehl, L., & Rechsteiner, M. (1982). Conjugation of ubiquitin to denatured hemoglobin is proportional to the rate of hemoglobin degradation in HeLa cells. Proc Natl Acad Sci U S A, 79 (19), 5857-5861.

Ciechanover, A., Heller, H., Elias, S., Haas, A. L., & Hershko, A. (1980). ATP- dependent conjugation of reticulocyte proteins with the polypeptide required for protein degradation. Proc Natl Acad Sci U S A, 77 (3), 1365-1368.

Djakovic, S. N., Schwarz, L. A., Barylko, B., DeMartino, G. N., & Patrick, G. N. (2009). Regulation of the proteasome by neuronal activity and calcium/calmodulin- dependent protein kinase II. J Biol Chem, 284 (39), 26655-26665. doi:10.1074/jbc.M109.021956

Ehlers, M. D. (2003). Activity level controls postsynaptic composition and signaling via the ubiquitin-proteasome system. Nat Neurosci, 6 (3), 231-242. doi:10.1038/nn1013

El Hokayem, J., & Nawaz, Z. (2014). E6AP in the brain: one protein, dual function, multiple diseases. Mol Neurobiol, 49 (2), 827-839. doi:10.1007/s12035-013-8563- y

Fagan, J. M., Waxman, L., & Goldberg, A. L. (1987). Skeletal muscle and liver contain a soluble ATP + ubiquitin-dependent proteolytic system. Biochem J, 243 (2), 335- 343.

Fonseca, R., Vabulas, R. M., Hartl, F. U., Bonhoeffer, T., & Nagerl, U. V. (2006). A balance of protein synthesis and proteasome-dependent degradation determines the maintenance of LTP. Neuron, 52 (2), 239-245. doi:10.1016/j.neuron.2006.08.015

Guo, X., Huang, X., & Chen, M. J. (2017). Reversible phosphorylation of the 26S proteasome. Protein Cell, 8 (4), 255-272. doi:10.1007/s13238-017-0382-x

Hamilton, A. M., Oh, W. C., Vega-Ramirez, H., Stein, I. S., Hell, J. W., Patrick, G. N., & Zito, K. (2012). Activity-dependent growth of new dendritic spines is regulated by the proteasome. Neuron, 74 (6), 1023-1030. doi:10.1016/j.neuron.2012.04.031

Hershko, A., Ciechanover, A., Heller, H., Haas, A. L., & Rose, I. A. (1980). Proposed role of ATP in protein breakdown: conjugation of protein with multiple chains of the polypeptide of ATP-dependent proteolysis. Proc Natl Acad Sci U S A, 77 (4), 1783-1786.

Hou, L., Antion, M. D., Hu, D., Spencer, C. M., Paylor, R., & Klann, E. (2006). Dynamic translational and proteasomal regulation of fragile X mental retardation protein controls mGluR-dependent long-term depression. Neuron, 51 (4), 441-454. doi:10.1016/j.neuron.2006.07.005

Huang, X., Luan, B., Wu, J., & Shi, Y. (2016). An atomic structure of the human 26S proteasome. Nat Struct Mol Biol, 23 (9), 778-785. doi:10.1038/nsmb.3273

Ikeda, F., & Dikic, I. (2008). Atypical ubiquitin chains: new molecular signals. 'Protein Modifications: Beyond the Usual Suspects' review series. EMBO Rep, 9 (6), 536- 542. doi:10.1038/embor.2008.93

Iwai, K., & Tokunaga, F. (2009). Linear polyubiquitination: a new regulator of NF- kappaB activation. EMBO Rep, 10 (7), 706-713. doi:10.1038/embor.2009.144

Lappe-Siefke, C., Loebrich, S., Hevers, W., Waidmann, O. B., Schweizer, M., Fehr, S., . . . Kneussel, M. (2009). The ataxia (axJ) mutation causes abnormal GABAA receptor turnover in mice. PLoS Genet, 5 (9), e1000631. doi:10.1371/journal.pgen.1000631

Lee, B. H., Lee, M. J., Park, S., Oh, D. C., Elsasser, S., Chen, P. C., . . . Finley, D. (2010). Enhancement of proteasome activity by a small-molecule inhibitor of USP14. Nature, 467 (7312), 179-184. doi:10.1038/nature09299

Lokireddy, S., Kukushkin, N. V., & Goldberg, A. L. (2015). cAMP-induced phosphorylation of 26S proteasomes on Rpn6/PSMD11 enhances their activity and the degradation of misfolded proteins. Proc Natl Acad Sci U S A, 112 (52), E7176-7185. doi:10.1073/pnas.1522332112

Lopez-Salon, M., Alonso, M., Vianna, M. R., Viola, H., Mello e Souza, T., Izquierdo, I., . . . Medina, J. H. (2001). The ubiquitin-proteasome cascade is required for mammalian long-term memory formation. Eur J Neurosci, 14 (11), 1820-1826.

Myeku, N., Wang, H., & Figueiredo-Pereira, M. E. (2012). cAMP stimulates the ubiquitin/proteasome pathway in rat spinal cord neurons. Neurosci Lett, 527 (2), 126-131. doi:10.1016/j.neulet.2012.08.051

Park, S., Kim, W., Tian, G., Gygi, S. P., & Finley, D. (2011). Structural defects in the regulatory particle-core particle interface of the proteasome induce a novel

proteasome stress response. J Biol Chem, 286 (42), 36652-36666. doi:10.1074/jbc.M111.285924

Schwarz, L. A., Hall, B. J., & Patrick, G. N. (2010). Activity-dependent ubiquitination of GluA1 mediates a distinct AMPA receptor endocytosis and sorting pathway. J Neurosci, 30 (49), 16718-16729. doi:10.1523/JNEUROSCI.3686-10.2010

Scudder, S. L., Goo, M. S., Cartier, A. E., Molteni, A., Schwarz, L. A., Wright, R., & Patrick, G. N. (2014). Synaptic strength is bidirectionally controlled by opposing activity-dependent regulation of Nedd4-1 and USP8. J Neurosci, 34 (50), 16637- 16649. doi:10.1523/JNEUROSCI.2452-14.2014

Swatek, K. N., & Komander, D. (2016). Ubiquitin modifications. Cell Res, 26 (4), 399- 422. doi:10.1038/cr.2016.39

Waxman, L., Fagan, J. M., & Goldberg, A. L. (1987). Demonstration of two distinct high molecular weight proteases in rabbit reticulocytes, one of which degrades ubiquitin conjugates. J Biol Chem, 262 (6), 2451-2457.

Chapter II: Investigating proteasome subunit Rpt6 phosphorylation dynamics using novel mouse lines that abrogate or mimic phosphorylation

Frankie R. Gonzales* 1†, Samantha L Scudder* 2, Kristin K. Howell 2, Lara E. Dozier 1, Stephan G. Anagnostaras 2 and Gentry N. Patrick 2

Author Affiliations: 1,2 Section of Neurobiology, Division of Biological Sciences,

University of California San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0347;

2Molecular Cognition Laboratory, Department of Psychology, University of California

San Diego, 9500 Gilman Drive, LA Jolla, CA 92093-0109;

† To whom correspondence should be addressed: Frankie R Gonzales, Division of

Chemistry and Biochemistry, University of California San Diego, 9500 Gilman Drive, La

Jolla, CA 92093-0347, [email protected], 707-486-9317

A. Abstract It is now apparent that protein degradation in the CNS via the Ubiquitin

Proteasome System (UPS) plays a crucial role in the development, maintenance, and remodeling of synapses. The proteasome subunit Rpt6, an AAA ATPase subunit of the

19S regulatory particle, has emerged as an important site for regulation of 26S proteasome function in neurons. It has been shown that phosphorylation of Rpt6 on serine

120 (S120) can stimulate the catalytic rate of substrate degradation by the 26S proteasome, making it an attractive target to explore regulation of proteasome function in neurons. Here, we created two mouse models which feature mutations at S120 that block or mimic phosphorylation at this site. We find that peptidase and ATPase activities are upregulated in the phospho-mimetic state, and downregulated in the phospho-dead state

(S120 mutated to aspartic acid (S120D) or alanine (S120A), respectively). However, these mutations had no effect on basal synaptic transmission, LTP, spine density, or behavior. While it has been shown that perturbations in Rpt6 S120 phosphorylation dynamics and proteasome activity affect the maintenance of plasticity, structure and function of synapses in various ways , we find that in a mouse model that blocks or mimics phosphorylation at this site, either compensatory mechanisms negate these effects, or small variations in proteasome activity are not enough to affect large scale changes in synaptic structure, activity, plasticity, or behavior.

B. Introduction Protein homeostasis, the balance between protein synthesis and degradation, is crucial in maintaining cell health and viability. The UPS is a predominant pathway for which proteins are broken down and recycled in eukaryotic cells, making it pivotal in their development, maintenance, and survival. The 26S proteasome is a multi-subunit complex which breaks down proteins targeted for degradation , and consists of a 20S catalytic core, as well as a 19S regulatory particle (RP). The 19S regulatory particle includes a six subunit hexameric ring of AAA ATPases (Rpt1-6) which serve as the interface with the alpha subunits of the core particle (CP), as well as provide the mechanical energy necessary to deubiquitinate substrates and move them into the catalytic core. Protein homeostasis is particularly important in the CNS, as rapid response to external cues, maintaining and remodeling synaptic connections, and constantly adjusting the protein composition of pre- and postsynaptic compartments is critical in synaptic function and dysfunction.

It was shown that large cohorts of pre and post synaptic proteins are degraded in response to chronic activity blockade and upregulation. Additionally, we found that the blockade of action potentials (APs) with tetrodotoxin inhibited proteasome peptidase activity. In contrast, up-regulation of activity with bicuculline, which blocks GABAergic synaptic transmission, dramatically increases peptidase activity. Ca 2+ /calmodulin- dependent protein kinase II alpha (CaMKIIα) has been directly linked to this phenomenon, as it was demonstrated that CaMKIIα acts as a scaffold protein necessary for proteasome translocation into spines and also stimulates proteasome peptidase activity by phosphorylating serine 120 on Rpt6, an ATPase subunit on the 19S regulatory particle

16 of the proteasome in an activity-dependent manner . In subsequent investigations, we found that homeostatic scaling of synaptic strength induced by chronic application of bicuculline or tetrodotoxin is both mimicked and occluded by altered Rpt6 phosphorylation. Furthermore, it was demonstrated that the acute inhibition of the proteasome blocks activity-dependent new spine generation. Together, these data suggest that CaMKII-dependent phosphorylation of Rpt6 at S120 maybe an important regulatory mechanism for proteasome-dependent control of synaptic remodeling in slow homeostatic plasticity.

We wanted to further investigate the role of Rpt6 phosphorylation and proteasome function in neurons. Historically, proteasome inhibitors have been used to understand the role of protein degradation in synaptic plasticity and behavior. However, they are quite toxic to cells and therefore have a limited utility. Many loss of function mutations in proteasome subunits cause lethality in various organisms, therefore the need for more subtle mutant models which partially alter proteasome function is quite high. Here, we have generated two novel knock-in mouse models which eliminate S120 phosphorylation dynamics by mutating Rpt6 S120 to either alanine (S120A) which would block phosphorylation, or to aspartate (S120D), which mimics phosphorylation due to its negatively charged carboxyl group. With these mouse models, we set out to examine the effects of altered Rpt6 S120 phosphorylation.

We find that while the phospho-mimetic mutant exhibits a significant increase in the kinetic rate of ATP hydrolysis and substrate degradation, and the phospho-dead mutant displays a significant decrease thereof, we did not observe any significant changes in spine outgrowth dynamics, basal synaptic transmission, LTP, nor basal learning and

17 memory. These findings are in stark contrast to previous data which showed significant differences in LTP, synaptic strength, and activity-dependent spine outgrowth.

Compensation mechanisms may be mitigating for the discreet changes in proteasome activity, or it could simply be that the small variations in activity are just not significant enough to affect systemic changes in synaptic structure, function, or behavior.

C. Materials and Methods Generation of S120A and S120D knock-in mice. We generated Rpt6 phospho- mimetic (ser120 to aspartic acid; S120D), and phospho-dead (ser120 to alanine; S120A)

KI mice (iTL; www.genetargeting.com). The targeting vectors were linearized and transfected by electroporation into BA1 (C57Bl/6 x 129/SvEv) (Hybrid) embryonic stem cells. Selected clones were expanded for southern blot analysis to identify recombinant

ES clones (data not shown). The ES clones were microinjected into C57BL/6 blastocysts. After germline transmission, the Neo cassette was removed by mating to

C57BL/6 FLP mice. Tail DNA was analyzed by PCR to identify heterozygous mice and verify deletion of the Neo cassette. Mutant heterozygous mice were backcrossed to

C57BL/6. By visual inspection, Rpt6 S120D and S120A homozygous mutants

(confirmed by PCR and sequencing) obtained by crossing heterozygous mutants, displayed normal body size, feeding, and mating behaviors. The intercross of heterozygotes resulted in production of wild-type, heterozygous, and homozygous offspring at the expected 1:2:1 Mendelian ratio. All procedures were approved by the

UCSD IACUC and compliant with the NRC Guide.

Antibodies and reagents. Anti-20S core α subunits monoclonal mouse antibody

(mAb MCP231) and Rpt6 regulatory subunit monoclonal mouse antibody (mAb p45-

110) were purchased from Enzo Life Sciences. Custom rabbit (pAb; Clone #07) anti-

Rpt6 phospho-specific antibody for serine 120 (pS120) was previously generated commercially (ProSci) against a synthetic phosphorylated peptide . N-succinyl-Leu-Leu-

Val-Tyr-7-amido-4-methylcoumarin (Suc-LLVY-AMC) substrate was used (BACHEM).

GST-Ubl and His-UIM constructs were a gift from the Alfred Goldberg lab. Malachite

Green and Adenosine triphosphate were purchased my Sigma Aldrich. PhosSTOP kinase/phosphatase inhibitor and cOmplete protease inhibitors were purchased from

Roche.

Proteasome purification. Purification was conducted as previously described , with some modifications. Whole brain lysates were dounce homogenized in affinity purification buffer (APB: 25 mM Hepes–KOH, pH 7.4, 10% glycerol, 5 mM MgCl 2, 1 mM ATP, and 1 mM DTT) and lysates cleared by ultra-centrifugation at 100,000g for 60 minutes. Lysate was incubated with purified GST-Ubl recombinant protein at a concentration of 0.2 mg’ml, then GSH-agarose beads were added and incubated for 2 hours at 4°C. The slurry was loaded onto a 20ml column and washed twice with 10 ml

APB. Proteasomes were eluted with purified UIM (2 mg/ml) in two incubations of 250ul each. The resulting purified proteasomes were then measured for protein concentration, aliquoted and frozen at -80°C.

Peptidase and ATPase assays. Peptidase assays were conducted according to methods previously described ). For peptidase assays, purified proteasomes were mixed with assay buffer (50 mM HEPES (pH 7.8), 10 mM NaCl, 1.5 mM MgCl2, 1 mM EDTA,

1 mM EGTA, 250 mM sucrose, 5mM DTT, 2mM ATP), and suc-LLVY-AMC substrate, and 100ul were loaded into a Costar black 96 well microassay plate (in triplicate), and the

19 kinetic rate of cleavage was monitored by the increase in fluorescence (Excitation:

360nm; Emission:465nm) at 37°C with a microplate fluorimeter (Perkin-Elmer HTS7000

Plus) every 30 seconds for two hours. The slope of the plot was calculated as the relative kinetic rate of substrate cleavage by the proteasome. The Malachite green assay, a colorimetric assay used to measure the evolution of inorganic phosphate from a mixture of pure proteasomes and ATP, was adapted from previous studies to quantify ATP hydrolysis. The assay was conducted in triplicate on clear Costar 96 well plates, and the absorption at 660nm was measured via plate reader spectrophotometer at different time points. The slope of the graph was calculated as the kinetic rate of ATP hydrolysis.

Western blot and native gel assays. Equal amounts of purified proteasomes were resolved by SDS-PAGE, then transferred onto nitrocellulose membranes. Membranes were then probed for proteasome α-subunits, total Rpt6, and phospho-Rpt6. In-gel activity assays were performed by loading equal amounts of purified proteasomes onto native gradient gels (4-12%) and resolving overnight at 4°C. The gel was then soaked in developing buffer (50mM Tris-HCl pH=7.4, 5mM MgCl2, 0.5mM EDTA, 1mM ATP) supplemented with 50 µM suc-LLVY-AMC substrate for 30 minutes at room temperature. The gel was then imaged in a Protein Simple FluorChem E imaging system, resulting in fluorescent bands. Gels were then transferred to nitrocellulose and probed for core subunits, Rpt6, and phospho-Rpt6.

Slice Electrophysiology. Acute hippocampal slices were prepared from 3-8 week old mice with experimenter blind to genotype. Mice were anesthetized with isofluorane prior to decapitation and brain extraction into ice-cold sucrose-containing ACSF (in mM:

83 NaCl, 2.5 KCl, 1 NaH2PO4, 26.2 NaHCO3, 22 glucose, 72 sucrose, 0.5 CaCl2, 3.3

MgSO4). Tissue was sliced coronally into 350 µm slices using a Leica VT1200 vibratome. Slices were recovered in standard ACSF (in mM: 119 NaCl, 5 KCl, 1

NaH2PO4, 26 NaHCO3, 11 glucose, 2 CaCl2, 1 MgSO4) at 34°C for 30 minutes and at room temperature for at least 30 minutes prior to recordings. Slices were transferred to a submerged recording chamber and perfused with room-temperature (basal transmission experiments) or 30°C (potentiation experiments) oxygenated ACSF (with 100 µM picrotoxin).

A cluster stimulating electrode (FHC) was placed in the stratum radiatum of the

CA1 region and current or voltage was injected using an ISO-Flex stimulus isolator

(A.M.P.I.) triggered by a Clampex 10.3 (Molecular Devices) or custom IGOR Pro protocol. Recording electrodes were generated from thin-walled capillary tubing (Warner

Instruments) using a horizontal pipette puller (P-97 Flaming/Brown Micropipette Puller,

Sutter Instruments), resulting in a resistance of 1-3 MΩ, and pipettes were filled with

ACSF. The recording electrode was placed 200-300 µm away from the stimulating electrode in the stratum radiatum, along a pathway parallel to the CA1 pyramidal layer. A second stimulating electrode was placed on the opposite side of the recording electrode to serve as a control pathway in LTP experiments. Field responses were recorded using an

Axopatch 200B amplifier (Molecular Devices) and digitized using a Digidata 1322 digitizer (Molecular devices), and signals were acquired using Clampex 10.3 or IGOR

Pro. Field excitatory post-synaptic potentials (fEPSPs) were evoked using 100 µs pulses.

Input-output relationships were determined by recording fEPSPs at a variety of stimulus intensities ranging from 30 µA to 130 µA and averaging five traces per intensity. Paired pulse facilitation was examined by generating pulses at variable separations (400, 200,

150, 100, and 50 ms) and taking the ratio of the second fEPSP amplitude to the first, with three recordings averaged for each separation. To study long-term potentiation (LTP), a stimulus intensity that produces a response of 0.2 – 0.4 mV was administered once every

15 seconds until a stable baseline of 15 minutes was reached. LTP was induced in one pathway using four sets of 1-second 100 Hz trains separated by 20 seconds and post-LTP responses were recorded from both control and experimental pathways for 40-60 minutes after potentiation.

Whole-cell Electrophysiology. Dissociated hippocampal neurons at DIV 21-24 from either genotype were used for recordings of whole-cell miniature excitatory postsynaptic currents (mEPSCs), with experimenter blind to genotype. Cultures were bathed in a room-temperature HEPES-buffered saline solution (in mM: 119 NaCl, 5 KCl,

2 CaCl2, 2 MgCl2, 30 glucose, 10 HEPES, 0.001 TTX, and 0.01 µM bicuculline, pH

7.2). Recording electrodes had resistances ranging from 2.5 – 4 MΩ and were filled with internal solution optimized for voltage-clamping (in mM: 10 CsCl, 105 CsMeSO3, 0.5

ATP, 0.3 GTP, 10 HEPES, 5 glucose, 2 MgCl2, and 1 EGTA (pH 7.2)). Pyramidal-like neurons were voltage-clamped at -70 mV and any cells with access resistance greater than 25 MΩ were discarded. mEPSC amplitude and inter-event interval were analyzed from 80-150 events per trace using Clampfit 10.3 and differences between genotypes were determined using unpaired t-tests.

Behavior. Fear Conditioning: Animals are trained 10 minutes in Context A: metal bar floors, isopropyl scent, and light. Baseline activity is recorded for the first 2 min while in context A . Animals are presented with 3 30-second tones (90 dB, 2800 Hz) at

22 minutes 3, 4, & 5 . The last 2 sec of the tones co-terminate with a shock . During the last 5 min of training, freezing is recorded as a measure of immediate memory

Spine density, and immunohistochemistry. Adult (P21-P182) homozygous Rpt6

S120A, homozygous S120D, corresponding WT mice were given an anesthetic analgesic solution of ketamine and xylazine prior to intracardial perfusion with 0.9% saline, followed by 4% PFA fixative solution (20mL 0.2M PB, 10mL ddH2O, 10mL 16% PFA).

Brains are postfixed for 1 hr then 100 μm coronal sections were cut using a Vibratome.

Penetrating microelectrodes were pulled from borosilicate capillary glass with filament

(1 mm outer diameter/−0.58 mm inner diameter) and then backfilled with KCl (200 mM) and Alexa 594 hydrazide (10 mM) solution (Invitrogen). Using micromanipulation and visual guidance, CA1 neurons were filled via iontophoresis. Sections were postfixed for

10min prior to mounting with Aqua Polymount (Polysciences Inc). Confocal microsopy was used to image secondary apical dendrites for analysis (see above). The density, length, and width of dendritic protrusions were measured via Z-stacks in ImageJ using custom macros blind to genotype.

Preparation and transfection of organotypic slice cultures. Organotypic hippocampal slices were prepared from litter matched P6-P8 wild-type and S120A KI mice of both sexes, as described . The cultures were transfected 1-2 d before imaging via biolistic gene transfer (180 psi), as previously described . 10-15 µg of EGFP (Clontech) was coated onto 6-8 mg of 1.6 µm gold beads.

Time-Lapse Imaging and Image Analysis for spine dynamics. EGFP-transfected

CA1 pyramidal neurons [7-9 days in vitro (DIV)] at depths of 10-50 µm were imaged using a custom two-photon microscope controlled with ScanImage . Slices were imaged

23 at 29 °C in recirculating artificial cerebrospinal fluid (ACSF) containing in mM: 127

NaCl, 25 NaHCO3, 25 D-glucose, 2.5 KCl, 1.25 NaH2PO4, 1 MgCl2, and 2 CaCl2, aerated with 95%O2/5%CO2. For each neuron, image stacks (512 × 512 pixels; 0.035

μm/pixel) with 1 μm z-steps were collected from 5 to 6 segments of secondary dendrites

(apical and basal) at 15 min intervals. To compare rates of activity-induced new spine formation in WT versus S120A KI mice, slices were treated with 30 µM bicuculline

(Tocris, 1,000x aqueous stock) or vehicle. All shown images are maximum projections of three-dimensional (3D) image stacks after applying a median filter (3 × 3) to the raw image data. Spine formation rates were blindly analyzed in 3D using custom software in

MATLAB.

D. Results 1. Generation, purification, and analysis of Rpt6 S120A and S120D mutant proteasome composition and activity

In our previous studies, we demonstrated neuronal activity and CaMKIIα- dependent phosphorylation of Rpt6 S120 and concomitant increased proteasome activity.

Additionally, only the co-expression of Rpt6 S120A mutant blocked CaMKIIα-dependent increased proteasome activity. Furthermore, expression of Rpt6 S120A and S120D produced opposite effects on synaptic strength. To determine if Rpt6 S120 phosphorylation was functionally relevant, we generated two novel mouse lines that endogenously express either Rpt6 S120A or Rpt6 S120D mutants in order to compare them with wild type. By visual inspection, Rpt6 S120A and Rpt6 S120D homozygous mutants displayed normal body size, feeding, and mating behaviors.

We evaluated proteasome peptidase and ATPase activity purified from Rpt6 WT and mutant mice. The gentle affinity purification protocol utilized more efficiently maintains interactions between the regulatory and catalytic particles of the proteasome, as well as those between the proteasome and its interacting proteins. This purification method, previously described by Besche and Goldberg, employs a ubiquitin-like protein motif

(UBL) to bind to proteasomes, and then competes proteasomes off UBL using a ubiquitin-interacting motif (UIM) . In order to verify efficient purification of WT and mutant proteasomes, we resolved the purification products on SDS -PAGE gels. Western blots show high concentrations of 20S subunits as well as total Rpt6. When we probed for phospho-Rpt6 using our affinity purified phospho-Rpt6 antibody, S120A mutants showed little to no cross-reactivity, while S120D showed marginal cross-reactivity, as expected

(Figure 2A). In order to evaluate composition and functionality of our purified proteasomes, we ran the purification products on a native gel and performed an in-gel fluorescence assay utilizing the LLVY-AMC substrate. After electrophoresis, we bathed the gel in a buffer containing the substrate and imaged it on a Protein Simple UV imager, rendering bands for both single and doubly capped proteasomes (Figure 2B), verifying that our proteasomes were in-tact and fully functional following purification.

In order to more quantitatively evaluate proteasome activity, we utilized kinetics assays to measure peptidase and ATPase activities to determine if they exhibit the same differences as they did in cultures. We found that Rpt6 S120A phospho-dead mutants showed significantly reduced peptidase activity as compared to WT (P<0.001; N=12), while Rpt6 S120D phospho-mimetic mutants exhibit significantly increased peptidase activity (P=0.028; N=12) (Figure 2C).

Rpt6 is one of six ATPase subunits in the 19S regulatory particle. These subunits hydrolyze ATP, providing the mechanical energy necessary to unfold target substrates and move them inside the catalytic core. This is accomplished through a conformational change when ATP is hydrolyzed, therefore it is proposed that the differences observed in peptidase activity as a result of S120 phosphorylation are due to small changes in Rpt6 tertiary structure, affecting the catalytic rate of ATP hydrolysis . We therefore also investigated the ATPase activities of our purified wild type and mutant proteasomes.

Indeed, we found that the S120A mutant displayed a reduced rate of ATPase activity compared to wild type (P=0.006; N=12), and the S120D mutant showed significantly increased ATPase activity (P=0.012; N=12) (Figure 2D). This data shows that peptidase activity, as well as ATPase activity of proteasomes purified from Rpt6 S120A and S120D mice exhibit similar increased and decreased kinetic activity (respectively) that we observed in culture with ectopically expressed mutant Rpt6.

2. Abrogation of phosphorylation dynamics do not effect basal synaptic transmission nor

LTP paradigms of synaptic plasticity.

Basal synaptic transmission. We first assessed whether mutant mice have a typical response curve following increasing intensities of Schaffer collateral (CA3-CA1) stimulation. We observed that slices from both S120D and S120A animals display a normal input-output relationship when compared to slices obtained from age-matched wild-type mice (Figure 2A,B). This indicates that the hippocampal circuit has developed normally and has roughly the same amount of synaptic connectivity as a wild-type hippocampus. Thus, mutations to S120 do not obviously impact the formation and maintenance of excitatory synapses in the hippocampus, and do not appear to strongly

26 impact synaptic strength at these synapses. To assess whether presynaptic release mechanisms were intact in S120D and S120A mice, we next utilized a paired pulse facilitation (PPF) assay. By stimulating Schaffer collateral axons twice at varying pulse separations, we revealed strong facilitation of fEPSPs in wild-type slices at small separations (50 ms, 100 ms). S120D and S120A slices displayed identical levels of facilitation, indicating that presynaptic neurotransmitter release is intact and has normal responses to increased presynaptic calcium (Figure 2C,D).

Long-term potentiation. Since previous work with proteasome inhibitors pointed to an important role for these complexes in mediating LTP, we next sought to determine whether our Rpt6 mutations would affect this plasticity paradigm. We generated acute hippocampal slices from young (P21-P27) age-matched mice and induced LTP in the

CA3-CA1 pathway using four trains of 1-second 100 Hz stimulation. We observed that hippocampal slices from S120D and S120A animals undergo normal levels of fEPSP potentiation compared to wild-type mice (Figure 2E,F).

Due to previous studies suggesting that proteasome activity can modulate LTP induction thresholds, we next used a weaker stimulation paradigm, instead stimulating with a single 1-second 100 Hz train. While both mutants displayed normal levels of slight potentiation 15-30 minutes after LTP induction (Figure 2G,H), we did observe a suppression of post-tetanic potentiation and early LTP in S120D animals (Figure 2E).

However, since potentiation eventually reached wild-type levels, it appears that neither mutation has a strong effect on LTP.

3. Small changes in proteasome ATPase and peptidase kinetic activity have no effect on fear conditioning, nor context discrimination paradigms of learning and memory

Fear Conditioning . One we verified that purified proteasomes from our mutants had significant differences in both peptidase and ATPase activities when compared to wild type, we wanted to know if these differences translated into deficits in behavior and memory in our mouse lines. The effects of mimicking and blocking Rpt6 phosphorylation at serine 120 were examined using Pavlovian fear conditioning to assess immediate and long-term associative memory (Fig. 3, 4). Baseline activity, measured during the first 2 min of training, did not differ between S120D (n=21) mice and WT (n=13) controls,

[F(1,32)= 1.25, p=0.27] (Figure 3A). However, shock reactivity in S120D mice was significantly dampened [F(1,32)= 10.48, p<0.005] (Figure 3A). The difference in shock reactivity did not seem to influence memory for the task. Freezing measured the last 5 min of training served as an indicator of immediate memory. There were no immediate memory differences between S120D and WT mice [F(1,32)=0.004, p=0.95] (Figure 3B).

When assessed 24 h later, we did not find any differences in contextual memory

[F(1,32)= 0.087, p=0.77] (Figure 3C). Similarly, cued memory, measured as average freezing while the tones played, did not differ between groups [F(1,32)= 1.20, p=0.28]

(Figure 3D). S120A (n=16) and WT (n=8) mice displayed similar activity levels at baseline [F(1,22)=0.25, p=0.62] (Figure 3E), but as with S120D mice, shock reactivity was reduced in S120A mice compared to WT littermates [F(1,22)=4.33, p=0.049] (Figure

3E). There were no immediate memory deficits in S120A mice [F(1,22)=1.29, p=0.27]

Figure 3F). Context memory and cued memory were also normal in S120A mice when

28 compared to WT controls. [F(1,22)= 1.645, p=0.21] (Figure 3G); [F(1,22)= 0.119, p=0.73] (Figure 4H), respectively.

Context Discrimination . In order to further examine whether a mutation at serine

120 would produce any perturbations in a hippocampal-dependent memory, we examined both S120D and S120A groups in a context discrimination task. Average shock reactivity was assessed across all eight days in context A+. We did not find any statistically significant differences between either S120D mice (n=4) and WTs (n=6) or S120A mice

(n=3) and WTs (n=7) [F(1,8)= 2.58, p=0.15] (Figure 3I); [F(1,8)= 0.91, p=0.37] (Figure

3J). Each day mice were placed in the chambers, we examined baseline freezing during the first four minutes. There were no baseline freezing differences found between S120D

[F(1,8)=0.11, p=0.75] (Figure 3K) mice or S120A [F(1,8)=0.79, p=0.79] (Figure 3L) mice and their respective WT littermates. In order to assess whether the mice were able to discriminate between the A+ and A- contexts, we took a difference score by subtracting the amount of freezing in A- from that in A+ on consecutive days. Day 1 was excluded since it was the first exposure to the context and there was no freezing as mice had not yet received a shock. No significant differences were found (Figure 3M,N) between either line of knock-ins and their respective control groups [S120D F(1,8)=0.13, p=0.73;

S120A F(1,8)=4.48, p=0.067].

4. Elimination of phosphorylation dynamics do not alter basal spine density .

It has been shown that maintenance of synaptic plasticity and spine outgrowth.

We wanted to investigate if the elimination of phosphorylation dynamics by either blocking phosphorylation in Rpt6-S120A KI mice, or mimicking phosphorylation in

Rpt6-S120D KI mice had any effect on basal spine density. We took coronal sections of

29 perfused mouse brains from both genotypes, micro-injected with Alexa Fluor 594 fluorescent dye, and counted the individual number of spine heads on 20 micron segments of secondary dendritic branches using confocal microscopy and compared them to their age-matched WT littermates (Figure 4A,B). We found that there was no statistically significant difference between the wild type control and the homozygous

Rpt6-S120A KI (P = 0.828, n = 56 for WT and n = 51 for hom S120A) (Figure 4C). We find that there is a statistically significant difference between WT and Rpt6-S120D KI, however the magnitude of the difference is so minute, we can consider it insignificant (P

= 0.039, n = 7 for WT and n = 26 for hom S120D) (Figure 4D).

5. Elimination of phosphorylation dynamics does not have a significant effect on spine outgrowth .

It has been previously shown that proteasome function is necessary in the generation of new spines when neuronal activity is artificially elevated. We have also shown that increasing activity also increases proteasome peptidase and ATPase activity.

Therefore, we wanted to investigate if the Rpt6 S120A knock-in mutant that cannot be phosphorylated had any effect on the generation of new spines when neuronal activity is elevated. Organotypic hippocampal slices of WT and homozygous S120A mutant mouse brains were prepared, transfected with EGFP, and imaged over time using a two-photon microscope after being treated with either vehicle, or 30 µM bicuculline (Figure 5A).

Spines were counted and compared to their vehicle controls. We found no difference in the basal spine density measured before treatment (P = 0.839, n = 6 for WT and n = 7 for

Hom S120A) (Figure 5B). We also no significant alteration in overall formation of new

30 spines between the genotypes (Figure 5C), nor when we normalized to their respective vehicle controls (P = 0.891, n = 6 for both WT and Hom S120A) (Figure 5D).

E. Discussion Our work has shown that phosphorylation dynamics of AAA APTase subunit

Rpt6 at serine 120 of the proteasome does have a regulatory effect on enzymatic activity, protein half-life control, as well as proteomic interactions with the holoenzyme. We find that phosphorylation increases the kinetic rate of substrate degradation as well as ATP hydrolysis, while dephosphorylation induces opposite effects of a similar magnitude.

While it has become increasingly clear that the UPS and protein half-life regulation play a vital role in synaptic transmission and plasticity, it seems that the small modulations in ATP hydrolysis and substrate degradation through S120 phosphorylation dynamics do not significantly change the trafficking or co-localization of proteasomes in neurons neither in basal conditions, nor in response to neuronal stimulation. In addition, evidence suggests that normal S120 phosphorylation dynamics are not specifically required for neither basal synaptic transmission, nor common synaptic plasticity paradigms, as abrogation of these dynamics result in normal frequency and amplitude of mEPSCs as well as LTP/LTD induction. Similarly, we find eliminating phosphorylation dynamics has no significant effect on basal learning and memory in mice, as no major differences were observed in behavioral protocols such as elevated water maze and

Pavlovian fear memory response.

These results provide evidence that while numerous modules of regulation of the proteasome and its function have been discovered and well characterized, not all of these

31 regulatory mechanisms serve vital roles in all functions of the CNS. This could be due simply to an ability for cells to mitigate for small changes in protein turnover using either alternative mechanisms of protein degradation, such as lysosomal degradation, or by simply changing transcription/translation rates to compensate for minor alterations in

UPS degradation. Recent work has illustrated the interplay and feedback loops between the UPS and the lysosome system, often showing that when one pathway is inhibited, the other is upregulated in order to compensate for the disturbance. Specifically, activation of mTORC1, which inhibits lysosomal degradation, has been shown to increase the number of intact and active proteasomes through the activation of NRF1. Additionally, it has been shown that certain transcription factors for proteasomal subunits are also substrates of the proteasome, creating a feedback loop where inhibition increases the transcription/translation of proteasome subunits, leading to an increase in active holoenzymes. Interestingly, some of these compensation mechanisms are only observed under low concentrations of proteasome inhibitors, and is absent under high levels of inhibition, indicating specialized compensatory mechanisms for minor alterations in activity.

While S120 phosphorylation dynamics showed no significant role in basal learning, memory, and synaptic plasticity paradigms, this does not rule out that there exists highly specialized pathways where downstream effects are dependent on small changes in kinetic activity directly linked to phosphorylation of Rpt6. Our lab has discovered a pathway related to cocaine sensitization, which seems to be highly dependent on Rpt6 becoming phosphorylated by CaMKIIα under cocaine administration.

In that study, we find that cocaine increases CaMKIIα activation, Rpt6 phosphorylation,

32 and subsequent increase in kinetic activity in the nucleus accumbens, as well as in the pre-frontal cortex. This effect was ameliorated when activity was locked in to an already elevated state by the S120D mutant. Since a stimulation in activity was impossible, sensitization to cocaine was occluded.

Taken together, these results suggest that there exists many levels of UPS control in the CNS, and that different mechanisms have different cellular consequences. Rpt6 phosphorylation is one regulatory module that does have significant effects on UPS function, though it is not critical in basal learning and memory. However, specific paradigms of CNS function and response to synaptic cues may still be highly dependent on Rpt6 phosphorylation dynamics.

E. References

Besche, H. C., and A. L. Goldberg. "Affinity Purification of Mammalian 26s Proteasomes Using an Ubiquitin-Like Domain." Methods Mol Biol 832 (2012): 423-32.

Bingol, B., and E. M. Schuman. "Activity-Dependent Dynamics and Sequestration of Proteasomes in Dendritic Spines." Nature 441, no. 7097 (Jun 29 2006): 1144-8.

Bingol, B., C. F. Wang, D. Arnott, D. Cheng, J. Peng, and M. Sheng. "Autophosphorylated Camkiialpha Acts as a Scaffold to Recruit Proteasomes to Dendritic Spines." Cell 140, no. 4 (Feb 19 2010): 567-78.

Ding, Q., E. Dimayuga, and J. N. Keller. "Proteasome Regulation of Oxidative Stress in Aging and Age-Related Diseases of the Cns." Antioxid Redox Signal 8, no. 1-2 (Jan-Feb 2006): 163-72.

Djakovic, S. N., E. M. Marquez-Lona, S. K. Jakawich, R. Wright, C. Chu, M. A. Sutton, and G. N. Patrick. "Phosphorylation of Rpt6 Regulates Synaptic Strength in Hippocampal Neurons." J Neurosci 32, no. 15 (Apr 11 2012): 5126-31.

Djakovic, S. N., L. A. Schwarz, B. Barylko, G. N. DeMartino, and G. N. Patrick. "Regulation of the Proteasome by Neuronal Activity and Calcium/Calmodulin- Dependent Protein Kinase Ii." J Biol Chem 284, no. 39 (Sep 25 2009): 26655-65.

Ehlers, M. D. "Activity Level Controls Postsynaptic Composition and Signaling Via the Ubiquitin-Proteasome System." Nat Neurosci 6, no. 3 (Mar 2003): 231-42.

Graham, S. H., and H. Liu. "Life and Death in the Trash Heap: The Ubiquitin Proteasome Pathway and Uchl1 in Brain Aging, Neurodegenerative Disease and Cerebral Ischemia." Ageing Res Rev 34 (Mar 2017): 30-38.

Hamilton, A. M., W. C. Oh, H. Vega-Ramirez, I. S. Stein, J. W. Hell, G. N. Patrick, and K. Zito. "Activity-Dependent Growth of New Dendritic Spines Is Regulated by the Proteasome." Neuron 74, no. 6 (Jun 21 2012): 1023-30.

Hershko, A., and A. Ciechanover. "The Ubiquitin System." Annu Rev Biochem 67 (1998): 425-79.

Kisselev, A. F., and A. L. Goldberg. "Monitoring Activity and Inhibition of 26s Proteasomes with Fluorogenic Peptide Substrates." Methods Enzymol 398 (2005): 364-78.

Lander, G. C., E. Estrin, M. E. Matyskiela, C. Bashore, E. Nogales, and A. Martin. "Complete Subunit Architecture of the Proteasome Regulatory Particle." Nature 482, no. 7384 (Jan 11 2012): 186-91.

Lanzetta, P. A., L. J. Alvarez, P. S. Reinach, and O. A. Candia. "An Improved Assay for Nanomole Amounts of Inorganic Phosphate." Anal Biochem 100, no. 1 (Nov 15 1979): 95-7.

Matyskiela, M. E., G. C. Lander, and A. Martin. "Conformational Switching of the 26s Proteasome Enables Substrate Degradation." Nat Struct Mol Biol 20, no. 7 (Jul 2013): 781-8.

Pologruto, T. A., B. L. Sabatini, and K. Svoboda. "Scanimage: Flexible Software for Operating Laser Scanning Microscopes." Biomed Eng Online 2 (May 17 2003): 13.

Sha, Z., and A. L. Goldberg. "Proteasome-Mediated Processing of Nrf1 Is Essential for Coordinate Induction of All Proteasome Subunits and P97." Curr Biol 24, no. 14 (Jul 21 2014): 1573-83.

Stoppini, L., P. A. Buchs, and D. Muller. "A Simple Method for Organotypic Cultures of Nervous Tissue." J Neurosci Methods 37, no. 2 (Apr 1991): 173-82.

Woods, G. F., W. C. Oh, L. C. Boudewyn, S. K. Mikula, and K. Zito. "Loss of Psd-95 Enrichment Is Not a Prerequisite for Spine Retraction." J Neurosci 31, no. 34 (Aug 24 2011): 12129-38.

Woods, G., and K. Zito. "Preparation of Gene Gun Bullets and Biolistic Transfection of Neurons in Slice Culture." J Vis Exp , no. 12 (Feb 13 2008).

Xie, Y., and A. Varshavsky. "Rpn4 Is a Ligand, Substrate, and Transcriptional Regulator of the 26s Proteasome: A Negative Feedback Circuit." Proc Natl Acad Sci U S A 98, no. 6 (Mar 13 2001): 3056-61.

Yedidi, R. S., P. Wendler, and C. Enenkel. "Aaa-Atpases in Protein Degradation." Front Mol Biosci 4 (2017): 42.

Zhang, Y., and B. D. Manning. "Mtorc1 Signaling Activates Nrf1 to Increase Cellular Proteasome Levels." Cell Cycle 14, no. 13 (2015): 2011-7.

Zhang, Y., J. Nicholatos, J. R. Dreier, S. J. Ricoult, S. B. Widenmaier, G. S. Hotamisligil, D. J. Kwiatkowski, and B. D. Manning. "Coordinated Regulation of Protein Synthesis and Degradation by Mtorc1." Nature 513, no. 7518 (Sep 18 2014): 440- 3.

F. Figures and Figure Legends

Figure 2.1: Proteasome protein degradation and ATP hydrolysis kinetics are significantly modulated in Rpt6 S120A and S120D knock-in mice . (A) Western blot analysis of proteasomes purified using Ubl/UIM method, and verification of phospho- Rpt6 antibody specificity to WT Rpt6 with expected cross reaction with S120D Rpt6, (B) native gel fluorescent activity assay and tandem western blot with two bands representing singly and doubly capped proteasomes, (C) fluorescent peptidase activity assay; S120A has significantly decreased activity (P<0.001, N=12), and S120D shows increased activity (P=0.028, N=12), and ANOVA suggest significant difference within the group (F=13.96, P<0.001), (D) malachite green ATPase assay; S120A displays significantly lower activity (P=0.006, N=12), while S120D shows increased activity (P=0.012, N=12), and the one way ANOVA suggests significant difference within the group (F=15.4, P<0.001).

Figure 2.2: Basal synaptic transmission and LTP are unaltered in Rpt6 S120A and S120D knock-in mice. (A) , Input/output curve depicting increased field EPSP responses to increased current amplitude in acute hippocampal slices from wildtype and S120D mice (n = 11 slices per genotype). (B) , Input/output curve in slices from wildtype and S120A mice (n = 15 slices per genotype). (C) , Two successive stimuli with short separation delivered to the Schaffer collateral of acute hippocampal slices leads to enhancement of fEPSP amplitude (paired pulse facilitation) in wildtype and S120D mice (n = 10 slices). (D) , Paired pulse facilitation in wildtype and S120A slices (n = 16 slices). (E) , Delivery of 4 1-second trains (20 s apart) of 100 Hz stimulation causes potentiation of fEPSP amplitude in acute hippocampal slices from wildtype and S120D mice (n = 9, 10 slices). (F) , LTP induction in response to 4 1-second 100 Hz trains in wildtype and S120A slices (n = 6, 9 slices). (G) , Delivery of a single 1-second 100 Hz train induces short-lasting post-tetanic potentiation followed by a small long-lasting potentiation of fEPSP amplitude (n = 5, 6 slices). (H) , LTP induction in response to a single 1-second 100 Hz train in slices from wildtype and S120A animals (n = 6, 7 slices). All graphs represent mean ± SEM.

Figure 2.3a: Context discrimination and Fear Memory were unaffected in Rpt6 S120A and S120D knock-in mice. (A) S120A mice exhibited normal shock reactivity. The data depict activity for S120A (n=3) and WT (n=7), and for S120D (n=4) and WT (n=6) mice during the presentation of the 2s shock on days when mice were placed in context A+. There were no significant between group differences (p=0.37), and (p=0.15), respectively. (B) Baseline freezing did not differ between S120A and WT mice, nor between S120D and WT mice. Freezing was assessed during the initial 4 min in the chambers each day. There were no between group differences (p=0.77), and (p=0.75), respectively. (C) Context discrimination was not impaired in S120A mice, nor S120D mice. Context discrimination was assessed using a difference score. The difference in percent time freezing in Context A+ (shock context) and Context A- (no shock context) on consecutive days is depicted. No differences were observed (p=0.07) and (p=0.73), respectively.

Figure 2.3b: Immediate, context, and cued memory were unaffected in Rpt6 S120A and S120D knock-in mice. (A) Shock reactivity was significantly lower in S120D mice. The data depict average activity for S120D (n=21) and WT (n=13) groups during training for both the 2 min baseline and during the 2s shock. Baseline activity did not differ between S120D and WT mice (p=0.27) but S120D mice exhibited significantly lower shock reactivity (ANOVA, F(1,34)=10.48, p<0.005). (B) Immediate memory was not impaired in S120D mice. Immediate memory was measured as the average percent time freezing during the last 5 min of training. Percent time during the immediate memory test is depicted. There were no differences between S120D mice and WT mice (p=0.95). (C) S120D mice had normal contextual fear memory. 24 h post-training mice were placed back into the training context and freezing was assessed across the 5 min test. There were no significant differences between S120D and WT mice (p=0.77). (D) Cued fear memory did not differ between S120D mice and WT mice. 24 h after the context test, mice were placed in a novel context and three tones were presented. There were no between group differences (p=0.28). (E) Shock reactivity was attenuated in S120A mice. Average activity during the initial 2 min baseline and average activity during the 2s shock is depicted for S120A (n=16) and WT (n=8) mice. Baseline activity did not differ between groups (p=0.62) but shock reactivity was significantly lower in S120A mice (ANOVA, F(1,22)=4.33, p<0.05). (F) Immediate memory did not differ between groups. Immediate memory was measured as the average percent time freezing during the last 5 min of training. S120A mice did not exhibit impaired immediate memory (p=0.27). (G) Contextual fear memory was not disrupted in S120A mice. Contextual fear memory was assessed 24 h post training, in the same context as training. Freezing did not differ between S120A and WT mice (p=0.73). (H) Cued fear memory did not differ between S120A and WT mice. 24 h after the assessment of contextual memory. No differences were observed between S120A and WT mice (p=0.21).

Figure 2.4: Spine density is not altered in S120 knock-in mice . (A) CA1 pyramidal neuron in coronal section of trascardially perfused mouse filled by targeted microinjection of Alexa Fluor 594 Hydrazide. Scale Bar: 10µm (B) Representative proximal dendritic segments of wild type S120A, homozygous S120A, wild type S120D and homozygous S120D adult mice (P21-P182). Scale Bar: 3µm (C) Spine density quantified by counting the individual number of spine heads on 20 micron segments of secondary dendritic branches in 63X confocal stacks (WT S120A n = 56 segments from 12 cells; Hom S120A n = 71 segments from 14 cells; WT S120D n = 51 segments from 18 cells; Hom S120D n = 26 segments from 10 cells)

Figure 2.5: New spine generation was not altered in Rpt6 S120A mouse neurons . (A) Images of dendrites from EGFP-expressing CA1 pyramidal neurons at 7-9 DIV before and after the addition of vehicle or bicuculline (30 μM) at t = 0 (black arrow). Yellow arrowheads indicate new spines. (B) Measure of basal spine density (spines/µm) between WT and S120A neurons. No difference was observed between WT and Rpt6 S120A knock-in (P = 0.839, n = 6 for WT and n = 7 for Hom S120A). (C) Elevated neural activity in response to bicuculline increased the rate of new spine formation in WT and S120A KI mice. (D) Increase in rate of new spine formation in response to bicuculline- induced elevated neuronal activity, normalized to vehicle controls. No significant difference was observed between WT and Hom S120A neurons (P = 0.891, n = 6 for both WT and Hom S120A)

Figure 2.S1. Generation of Rpt6 S120D Knockin (KI) mice . (A) , Schematic of targeting strategy. Genomic DNA structure of psmc5 (Rpt6 ) region is shown. The targeting vector was designed such that the long homology arm (LA) extends 5.98kb 3’ to the first point mutation (asterisk; AG ‰ GA, S120D; AG ‰ GC, S120A) in exon 5. The FRT flanked Neo resistance cassette was inserted 463 bp 5’ to the point mutation. The short homology arm (SA) extends 2.79 kb 5’ to the FRT flanked Neo cassette. (B) , Tail genomic DNA was analyzed by PCR screening for genotyping and to verify deletion of the Neo cassette. (C), Electropherograms confirming the presence of the mutation in homozygous Rpt6 S120D mutant male mice . (D), Representative images of Nissl stained fixed whole brain coronal sections of 60 day old mice.

10 Figure 2.S2: Performance on the elevated plus maze was not impaired in S120 knock-in mice . (A) Total distance traveled was assessed. There were no differences between S120D and WT mice (p=0.87). (B) Time spent in each arm of the elevated plus maze did not differ between S120D (n=14) and WT (n=7) mice. Percent time spent in the open and closed arms is depicted for each group. There were no differences observed (p>0.10). (C) Distance traveled was calculated for the open and closed arms separately. There were no between group differences for distance travelled in either open or closed arms (p>0.30). (D) Time spent in each arm of the elevated plus maze did not differ between S120A (n=13) and WT (n=8) mice. Percent time spent in the open and closed arms is depicted for each group. There were no differences observed (p>0.10).

Table S2.1: Proteasome interacting proteins. Mass spec analysis of purified proteasomes from WT, S120A and S120D model mice. Sample proteins with differential spectral hits are shown for each genotype.

ACKNOWLEDGEMENTS

Chapter 2, in part is currently being prepared for submission for publication of material. Gonzales FR, Scudder SL, Howell KK, Dozier LE, Anagnostaras SG, and

Patrick GN. The dissertation/thesis author in the primary investigator and author of this material.

Chapter III: Proteasome phosphorylation regulates cocaine-induced sensitization

Frankie R. Gonzales* 2† , Kristin K. Howell* 1, Lara E. Dozier 2, Stephan G. Anagnostaras 1 and Gentry N. Patrick 2

Affiliations: 1Molecular Cognition Laboratory, Department of Psychology, University of

California San Diego, 9500 Gilman Drive, LA Jolla, CA 92093-0109; 2Section of

Neurobiology, Division of Biological Sciences, University of California San Diego, 9500

Gilman Drive, La Jolla, CA 92093-0347

* F.R.G. and K.K.H. contributed equally to this work.

† To whom correspondence should be addressed: Frankie R Gonzales, Division of

Chemistry and Biochemistry, University of California San Diego, 9500 Gilman Drive, La

Jolla, CA 92093-0347, [email protected], 707-486-9317

A. Abstract Repeated exposure to cocaine produces structural and functional modifications at synapses from neurons in several brain regions including the nucleus accumbens. These changes are thought to underlie cocaine-induced sensitization. The ubiquitin proteasome system plays a crucial role in the remodeling of synapses and has recently been implicated in addiction-related behavior. The ATPase Rpt6 subunit of the 26S proteasome is phosphorylated by Ca 2+ /calmodulin-dependent protein kinases II alpha at ser120 which is thought to regulate proteasome activity and distribution in neurons. Here, we demonstrate that Rpt6 phosphorylation is involved in cocaine-induced locomotor sensitization. Cocaine concomitantly increases proteasome activity and Rpt6 S120 phosphorylation in cultured neurons and in various brain regions of wild type mice including the nucleus accumbens and prefrontal cortex. In contrast, cocaine does not increase proteasome activity in Rpt6 phospho-mimetic (ser120Asp) mice. Strikingly, we found a complete absence of cocaine-induced locomotor sensitization in the Rpt6 ser120Asp mice. Together, these findings suggest a critical role for Rpt6 phosphorylation and proteasome function in the regulation cocaine-induced behavioral plasticity.

B. Introduction The persistent neural and behavioral adaptations characteristic of addiction can render an addict permanently susceptible to relapse even years after cessation of drug use.

The enduring nature of these modifications suggests the involvement of memory or memory-like neuronal remodeling (Robinson 1997, Russo, Dietz et al. 2010, Howell,

Monk et al. 2014). Behavioral sensitization is an increase in the sensitivity to a drug following repeated administration, characterized by enhanced locomotor activity, dopamine release, and the rewarding value of the drug (Robinson 2008). Sensitization is argued to mediate the transition from ordinary goal-seeking behavior to compulsive behavior as the drug progressively elicits a much stronger dopamine response than natural reinforcers. The involvement of protein synthesis is well demonstrated in traditional and addiction-related memories (Schafe 2000, Hernandez 2002), however the importance of protein degradation in memory and addiction-related plasticity and behavior has only recently been studied.

The ubiquitin proteasome system (UPS) plays a major role in the development, maintenance and remodeling of synaptic connections (Mabb and Ehlers 2010, Hamilton and Zito 2013). The 26S proteasome, which degrades ubiquitinated proteins, is a large multi-subunit energy-dependent protease formed by the co-assembly of a 20S proteasome catalytic core and 19S cap regulatory particle (RP) (Hershko and Ciechanover 1998). We and others recently found a novel form of regulation for the 26S proteasome involving

CaMKIIα, a key kinase involved in neuronal plasticity underlying learned behaviors.

CaMKIIα phosphorylates the ATPase 19S (RP) subunit of the 26S proteasome, Rpt6, at serine 120 (S120) in an activity-dependent fashion to control the activity and distribution

48 of proteasomes in neurons (Djakovic, Schwarz et al. 2009, Bingol, Wang et al. 2010,

Djakovic, Marquez-Lona et al. 2012). We have previously shown that Rpt6 is involved in the regulation of synaptic strength and activity-dependent generation of new spines

(Djakovic, Marquez-Lona et al. 2012, Hamilton, Oh et al. 2012). Here, we utilize recently generated Rpt6 S120D phospho-mimetic knock in (KI) mice (serine 120 to aspartic acid) to examine the importance of Rpt6 phosphorylation on cocaine-induced locomotor sensitization. We found that cocaine increases Rpt6 S120 phosphorylation in cultured neurons, the nucleus accumbens (NAc), and the prefrontal cortex (PFC) of wild type

(WT) mice. Furthermore, repeated cocaine administration elevated proteasomal activity in both the NAc and PFC in WT mice but not in Rpt6 S120D KI mutant mice. Cocaine- induced locomotor sensitization was completely absent in the Rpt6 S120D KI mice.

Together, these findings implicate a critical role for Rpt6 phosphorylation and proteasome function in the regulation of cocaine-induced behavioral sensitization.

C. Materials and methods Generation of S120D Knockin mice . We generated Rpt6 phospho-mimetic (ser120 to aspartic acid; S120D) KI mice (iTL; www.genetargeting.com). The strategy for generating the KI mice is described in Figure 2. The targeting vector was linearized and transfected by electroporation into BA1 (C57Bl/6 x 129/SvEv) (Hybrid) embryonic stem cells. Selected clones were expanded for southern blot analysis to identify recombinant ES clones (data not shown).

The ES clones were microinjected into C57BL/6 blastocysts. After germline transmission, the

Neo cassette was removed by mating to C57BL/6 FLP mice. Tail DNA was analyzed by PCR to identify heterozygous mice and verify deletion of the Neo cassette. Mutant heterozygous mice were backcrossed to C57BL/6. By visual inspection, Rpt6 S120D homozygous mutants

(confirmed by PCR and sequencing) obtained by crossing heterozygous mutants, displayed

49 normal body size, feeding, and mating behaviors. The intercross of heterozygotes resulted in production of wild-type, heterozygous, and homozygous offspring at the expected 1:2:1

Mendelian ratio. All procedures were approved by the UCSD IACUC and compliant with the

NRC Guide.

Antibodies and Reagents . Mouse mAb Rpt6 (Enzo mAb P45-110, BML-PW9265-

0025), mAb 20S (Enzo mAb MCP231, BML-PW8195-0025), tubulin (Sigma T9026), mAb CaMKIIα Abcam ab2725), tyrosine hydroxylase (Millipore AB152), MAP2

(Abcam AB5392) and custom rabbit pAb phospho-specific Rpt6 S120 (Djakovic,

Marquez-Lona et al. 2012) antibodies were used. N-succinyl-Leu-Leu-Val-Tyr-7-amido-

4-methylcoumarin (Suc-LLVY-AMC) substrate was used (BACHEM).

Methamphetamine HCl (Sigma M8750), and atomoxetine HCl (TCI Chemicals A2357) were dissolved in water and citalopram HBr (Enzo BMC-NS112) was dissolved in ethanol. Cocaine HCl (Sigma) administered during behavioral assessment was dissolved in physiological saline to a dose of 15 mg/kg (salt weight) and administered i.p. (10 ml/kg).

Neuronal cultures. High Density Rat dissociated cortical neurons from postnatal day 1 pups of either sex were plated onto poly D-lysine-coated 6-well plastic dishes at

~500,000 cells per well (cortical cultures) and were maintained in B27 supplemented

Neurobasal media (Invitrogen) until ≥ 14 d in vitro (DIV), as previously described

(Djakovic, Schwarz et al. 2009).

Proteasome activity assays. Proteasome activity was measured as previously described (Kisselev and Goldberg 2005) with slight modifications. Briefly, cultured neurons were incubated for 24 h in either plain media (control), or media containing

Cocaine (1 μm, 5 μm), methamphetamine (500 µM), citalopram (25 µM) or atomoxetine

(25 µM). Cultured neurons and brain tissues were then lysed or dounced, respectively, in

Affinity Purification Buffer (APB) ( 25 mM Hepes–KOH, pH 7.4, 10% glycerol, 5 mM

MgCl2, 1 mM ATP, and 1 mM DTT) (Besche and Goldberg 2012), and lysates were cleared by centrifugation. Equal amounts of protein were used as inputs for the peptidase assay. The chymotrypsin-like activity of 26S proteasomes was monitored over time using

Suc-LLVY-AMC substrate. Assays were run in triplicate at 37 ⁰C using a microplate fluorimeter (HTS7000 Plus, Perkin-Elmer) with excitation and emission filters of 360nm and 465nm, respectively. Kinetics data was taken every 60 sec for 2 h using 96-well microplates (Costar). The data was averaged and plotted to find the kinetic rates of the chymotrypsin-like activity.

Western blot analysis. Equal or increasing amounts of protein lysates from dissociated cortical neurons as well as brain lysates were resolved by SDS-PAGE and probed with primary antibodies for total Rpt6, phospho-Rpt6 S120, CaMKIIα, and tyrosine hydroxylase. Resulting blots were digitized and band intensities quantitated using NIH ImageJ. For quantification of total phospho-Rpt6 S120 levels, band intensities in each condition were normalized to total Rpt6 band mean intensity from the same sample. Experimenters were blinded to condition during data collection and analysis.

qPCR. Total RNA was extracted from dissected samples or cultured neurons using Hybrid-R RNA purification kit (GeneAll Biotechnology). Purified RNA samples were reverse transcribed by using the SuperScript-IV First-strand cDNA synthesis kit

(Invitrogen). qRT-PCR was performed by using TaqMan Gene Expression Assay Kit

(Applied Biosystems). All TaqMan probes were purchased from Applied Biosystems

51 and are as follows: TH (Mm00447557_m1) and glyceraldhyde-3-phosphate dehydrogenase (GAPDH; Mm99999915_g1). Target amplification was performed by using ViiATM 7 Real-Time PCR System (Applied Biosystems). The relative mRNA expression levels were calculated via a comparative threshold cycle (Ct) method using

GAPDH as an internal control: DCt = Ct (gene of interest) – Ct (GAPDH). The gene expression fold change was normalized to the control sample and then was calculated as

2-DDCt (Schmittgen and Livak 2008)

Behavioral Sensitization. 40 adult Rpt6 S120D and WT mice were used for behavioral experiments. All animals were group housed. The vivarium was maintained on a 14:10 light:dark schedule. Mice had unrestricted access to food and water. All mice were handled for five days prior to behavioral assessment. Mice were then habituated to testing chambers during two, 1h sessions (30 min per side). Testing was conducted during the light phase (Howell, Monk et al. 2014). Two-sided polycarbonate chambers, each side distinct in appearance and floor, bisected by an opaque acrylic wall were paired with saline or drug (counterbalanced). Mice were divided into four groups: S120D-

Cocaine, WT-Cocaine, S120D-Saline, and WT-Saline. Sensitization was induced during four sessions of cocaine administration (as in Figure 1C or 3A). All animals first received an injection of saline (10 ml/kg) and were placed in the saline-paired side of the chamber for 15 min. Mice were removed, given an injection of either cocaine (S120D-Cocaine,

WT-Cocaine groups) or saline (S120D-Saline, WT-Saline groups) and immediately placed in the drug-paired side of the chamber for 15 min.

Histology . 24h post-sensitization, mice were given a final injection of cocaine (S120D-

Cocaine, WT-Cocaine) or saline (S120D-Saline, WT-Saline) in their home cage. 30 minutes later,

mice were anesthetized (isoflurane) and decapitated for fresh tissue collection. For whole brain

experiments, brains were flash frozen in liquid nitrogen and stored at -80°C. For experiments

targeting the NAc and PFC, brains were removed and frozen on dry ice. 1mm coronal sections

were obtained using an acrylic matrix (A/P: 2.10-1.10; Braintree Scientific). 1mm punches

(Miltex) were then taken from the NAc (M/L: ±0.5, D/V: 2.0) and PFC (M/L: ±1.25, D/V: 4.25),

flash frozen in liquid nitrogen and stored at -80°C.

Data Analysis . For western blot experiments statistical significance was determined

using unpaired Student t tests (Prism). Behavioral data were entered into a multivariate ANOVA

(SPSS). For proteasome activity assays, ANOVA was performed, or Mann-Whitney u test (for

non-parametric data). If a significant omnibus comparison or group x time/session interaction was

achieved, post-hoc comparisons were made using Fisher’s protected least significant difference

(PLSD). For all experiments, significance was set at a level of p≤.05.

D. Results 1. Cocaine increases Rpt6 S120 phosphorylation and proteasome activity .

It is known that cocaine activates CaMKIIα through T286 phosphorylation

(Anderson 2008). Since we discovered that CaMKIIα phosphorylates Rpt6 at ser120

which increases proteasome activity (Djakovic, Schwarz et al. 2009, Djakovic, Marquez-

Lona et al. 2012), we wondered whether cocaine could regulate Rpt6 S120 phosphorylation and proteasome activity. We examined Rpt6 S120 phosphorylation and proteasome activity in lysates from cultured neurons treated with increasing amounts of

cocaine. Measuring proteasome peptidase activity, we found that treatment with 5µM

cocaine significantly increased proteasome activity (ANOVA F(2,22)=11.81, P=0.001;

unpaired Student’s t test: t(16)=5.455, P<0.001) (Figure 1A). While we observe an

increase in Rpt6 phosphorylation after 1µM cocaine treatment (Figure 1B,C), we only

53 observe a significant increase in proteasome activity after 5µM treatment. To determine if monaminergic systems were involved in this effect, we utilized a second dopamine re- uptake inhibitor, methamphetamine (which also inhibits serotonin and norepinephrine re- uptake, yet does not interact with sodium ion channels), a serotonin-specific re-uptake inhibitor (citalopram) as well as a norepinephrine re-uptake inhibitor (atomoxetine) to treat cultured neurons. We found that treatment with methamphetamine (500µM) produced an increase in proteasome activity, similar to that of cocaine (ANOVA F(3,28)

= 4.204, P = 0.0142; unpaired Student’s t test: t(18) = 4.34, P < 0.001), however neither citalopram (25µM) (t(16) = 0.3462, P = 0.734) nor atomoxetine (25µM) (t(15) = 0.8767,

P = 0.395) had any significant effect (Figure 1D). Additionally, we observe significant expression of tyrosine hydroxylase in our dissociated neuron cultures, both via western blot (Figure 1E) as well as by qPCR (Figure 1F), which suggest the presence of dopaminergic systems. We additionally evaluated whether cocaine would produce similar effects on Rpt6 S120 phosphorylation and proteasome activity when administered to mice (i.p.). We treated mice with either cocaine or saline for 6 days (Figure 1G). Rpt6

S120 phosphorylation and proteasome activity was significantly increased in WT whole brains from mice treated with cocaine compared to saline (Mann-Whitney u test: U = 0, P

= 0.029) (Figure 1H,I,J). Taken together, we show that cocaine concomitantly increases

Rpt6 phosphorylation and proteasome activity.

2. Generation of Rpt6 S120D Knockin (KI) mice .

To begin to assess the biological relevance of Rpt6 (PSMC5) S120 phosphorylation, we generated genetically modified mice encoding a serine 120 to aspartic acid phospho-mimetic mutant of the Rpt6 protein (Figure 2A), which can be

54 distinguished from the wild type Rpt6 gene by PCR ‐based genotyping (Figure 2B and C).

Rpt6 S120D homozygous KI mice expressed similar levels of Rpt6 protein as wild ‐type

(+ / +) mice. By visual inspection, Rpt6 S120D homozygous mutants were indistinguishable from wild type littermates and displayed normal body size, feeding, and mating behaviors. We found that our custom pS120 phospho-specific antibody, which does not recognize Rpt6 S120A mutant protein (Djakovic, Marquez-Lona et al. 2012) slightly cross reacts with the Rpt6 S120D which is likely due to the mimetic nature of the negative charge (Figure 2D). Gross brain anatomy in Rpt6 S120D homozygous mice was comparable to wild-type (Figure 2E).

3. Rpt6 S120 phosphorylation and peptidase activity is increased in NAc and PFC in cocaine treated wild type mouse brains, but not in S120D mutant mice .

Major biochemical and structural changes in both the NAc as well as the PFC are associated with cocaine sensitization (Robinson 1997). To determine if cocaine increases

Rpt6 phosphorylation and proteasome activity specifically in the NAc and PFC, mice were administered five cocaine (15 mg/kg, i.p.) or saline treatments across 6 days (Figure

3A) and then lysates were extracted from tissue punched from the NAc and PFC of both

WT and Rpt6 phospho-mimetic S120D mice (Figure 3B). Western blot analysis and peptidase assays were performed. In WT mice, cocaine administration increased Rpt6 phosphorylation in the NAc in comparison to treatment with saline (Figure 3E). This increase is also correlated with a significant enhancement in proteasome activity in the

NAc (U = 0, P = 0.029) (Figure 3C). Interestingly, however, cocaine-induced increases in proteasome activity was not observed in the NAc of S120D mice (U = 7, P = 0.886)

(Figure 3D). A similar trend was observed in PFC. Cocaine increased Rpt6 S120

55 phosphorylation in PFC in cocaine- treated WT mice (Figure 3H), which correlated with increased peptidase activity (U = 0, P = 0.029) (Figure 3F). As in the NAc, this increase was occluded in the PFC of S120D mutated mice, as no difference was observed between mice administered cocaine or saline (U = 8, P > 0.999) (Figure 3F). Taken together, we show that cocaine concomitantly increases Rpt6 phosphorylation and proteasome activity specifically in the NAc and PFC and that the cocaine-induced increase in proteasome activity is occluded in the Rpt6 S120D mutant.

4. Behavioral sensitization is completely absent in Rpt6 S120D KI mutant mice .

Locomotor sensitization to cocaine (15 mg/kg i.p.) was assessed in four groups of mice; S120D mice administered cocaine and saline (S120D-Cocaine (n=11) and S120D-

Saline (n=5), respectively) and WT littermates administered cocaine and saline (WT-

Cocaine (n=18) and WT-Saline (n=6), respectively). Baseline activity (prior to any injections; habituation (H), Figure 4A) did not differ between groups [F(3,36)= .037, p=0.99)]. Differences emerged during subsequent sessions [Figure 4A, F(3,36)=5.79, p<.005]. While locomotor activity increased in WT-Cocaine mice across repeated injections, activity levels of S120D-Cocaine mice remained constant (Figure 4A, Fisher’s

PLSD, p=0.007). Further, locomotor activity did not differ between S120D-Cocaine mice and groups administered saline (S120-Cocaine/ S120D-Sal, p=0.59; S120D-Cocaine/

WT-Saline, p=0.36). There were no between group differences during the first session of cocaine (or saline) administration [Figure 4B, F(3,36)=.778, p=0.51]. Importantly, there was no difference in the acute (Day 1) response to cocaine between S120D and WT mice

(p=0.76). Session 4 locomotor activity was elevated in WT-Cocaine mice compared to all other groups [Figure 4C, F(3,36)= 4.57, p<0.01; S120D-Cocaine/ WT-Cocaine p= 0.01].

Interestingly, locomotor activity did not differ between S120D-Cocaine mice and either saline group (S120D-Cocaine/ S120D-Saline, p=0.64; S120D-Cocaine/ WT-Saline, p=0.67). Sensitization was assessed as a difference between session 1 (acute) and session

4 (sensitized) activity (Figure 4D). There were significant group differences [F(3,36)=

3.70, p<.0.05]. S120D-Cocaine mice did not sensitize, whereas WT-Cocaine mice demonstrated robust sensitization (Figure 4D; p=0.008). Again, there were no activity differences between S120D-Cocaine mice and saline control groups (S120D-Cocaine/

S120D-Saline, p=0.87; S120D-Cocaine/ WT-Saline, p=0.79).

E. Discussion In order to assess the importance of proteasome-dependent protein degradation in addiction-related behavior we generated a novel line of mutant mice with a single point mutation on a single subunit of the 26S proteasome. Phosphorylation of the 19S ATPase subunit Rpt6 at S120 has been shown to be regulated by CaMKII in an activity- dependent manner, and increases in Rpt6 S120 phosphorylation correlated with increased proteasome activity (Figure 1 and 3; (Djakovic, Marquez-Lona et al. 2012)). Here, we demonstrate that Rpt6 phosphorylation at S120 is critical for cocaine-induced sensitization, a prominent addiction-related behavior. In cell culture, we observe a cocaine-induced significant increase in proteasome phosphorylation as well as chymotrypsin-like activity. Furthermore, we observed similar effects treating with methamphetamine, and not SERT-specific nor NET-specific inhibitors produced a similar effect. Together with TH immunoreactivity observed via western blot, as well as verification of mRNA expression via qPCR, we can conclude that this cocaine-induced response involves the dopaminergic system. In Rpt6 S120D mutant mice, Rpt6 S120

57 phosphorylation is locked in the phospho-mimetic state, behavioral sensitization to cocaine is completely absent, indicating the importance of dynamic regulation of Rpt6

S120 phosphorylation.

When treated with multiple administrations of cocaine, locomotor activity of

S120D mice did not differ locomotor activity from mice receiving saline. It is important to note that while sensitization was blocked in these mice, the acute response to cocaine did not differ from WT animals (Figure 4B). Thus, we are able to conclude that there is a disruption in nonassociative memory rather than simply an impaired response to cocaine.

Our results tend to support an interactionist view of memory and addiction

(Volkow, Fowler et al. 2002, Carmack, Kim et al. 2013). According to this model there are distinguishable associative and nonassociative components of addiction, and the molecular mechanisms and neural substrates underlying these processes may overlap with those involved in canonical forms of memory (Anagnostaras and Robinson 1996,

Anagnostaras 2002, Russo, Dietz et al. 2010). Here, we demonstrate that behavioral changes reflecting the nonassociative component of addiction, e.g. sensitization, that result from repeated drug administration are affected by changes to cellular plasticity mediated by the ubiquitin proteasome system.

Other recent studies have begun to investigate the importance of the ubiquitin proteasome system in addiction, specifically the effects of protein degradation inhibition during the development and expression of conditioned place preference (a task that models drug-seeking) and sensitization (Massaly 2013, Ren 2013). Development, but not expression of morphine-induced CPP was impaired by administration of a UPS inhibitor

(Massaly 2013). Similarly, treatment with a proteasome inhibitor during the induction of

58 sensitization to morphine produced impairment (Massaly 2013). Ren and colleagues

(2013) investigated the effects of co-administration of a protein synthesis inhibitor and inhibitor of the UPS on cocaine-induced conditioned place preference. Co-treatment with a UPS inhibitor reversed the memory impairments typically produced by the administration of the protein synthesis inhibitor alone (Ren 2013). Understanding the role of protein degradation during addiction is a young area of investigation and many questions remain. The present study advances our current understanding by utilizing the first discrete mouse-model of altered proteasome function. This is important, as many of the previous studies have relied upon the use of inhibitors, which have been shown to be toxic to cells (Reaney 2006).

We also found that the increase in proteasomal activity in the PFC and NAc typically induced by cocaine administration was absent in S120D KI mice, indicating that the effects of cocaine are occluded since we found an overall increase in proteasome activity from intact affinity purified 26S proteasomes from total brain homogenates in

S120D KI mice (Gonzales and Patrick; data not shown). The PFC and NAc have previously been implicated in behavioral sensitization (Robinson 1997, Thomas 2001,

Steketee 2003). Enduring morphological alterations were found in NAc and PFC neurons following amphetamine-induced sensitization including an increase in the length of dendrites, density of spines, and number of branched spines (Robinson 1997). These results highlight the long-lasting adaptations to synaptic connectivity that result from repeated experience with drugs of abuse. Specifically how these morphological changes contribute to addiction is unknown. Interestingly, the UPS has previously been shown to be implicated in spine stability and recently Rpt6 phosphorylation and proteasome

59 function have been shown to regulate the formation of new dendritic spines (Hamilton,

Oh et al. 2012). One intriguing possibility that may account for the present findings is that activity-dependent Rpt6 phosphorylation at serine 120 contributes to the structural changes that occur following repeated psychostimulant administration. Potentially, constitutively active phosphorylation at this site may interfere with the stimulant-induced morphological changes, thus preventing sensitization.

F. Conclusions Surprisingly, we observed no locomotor sensitization of the Rpt6 S120D phospho-mimetic animals to cocaine, indicating this sensitization is occluded by the inability to modulate proteasome activity through its phosphorylation. Our results show that regulation of the proteasome itself to be important in cocaine-induced sensitization.

However, future understanding of downstream UPS targets involved in these paradigms will be essential to better understand the molecular mechanisms underlying cocaine sensitization.

G. References Anagnostaras, S., Schallert, T, Robinson, TE (2002). "Memory Processes Governing Amphetamine-induced psychomotor sensitization." Neuropsychopharmacology 26: 703-715.

Anagnostaras, S. G. and T. E. Robinson (1996). "Sensitization to the psychomotor stimulant effects of amphetamine: Modulation by associative learning." Behavioral Neuroscience 110(6): 1397-1414.

Anderson, S., Famous, KR, Sadri-Vakili, G, Kumaresan, V, Schmidt, HD, Bass, CE, Terwilliger, EF, Cha, JHJ, Pierce, RC ( 2008). "CaMKII: a biochemical bridge linking accumbens dopamine and glutamate systems in cocaine seeking." Nature Neuroscience 11: 344-353.

Besche, H. C. and A. L. Goldberg (2012). "Affinity purification of mammalian 26S proteasomes using an ubiquitin-like domain." Methods Mol Biol 832: 423-432.

Bingol, B., C. F. Wang, D. Arnott, D. Cheng, J. Peng and M. Sheng (2010). "Autophosphorylated CaMKIIalpha acts as a scaffold to recruit proteasomes to dendritic spines." Cell 140(4): 567-578.

Carmack, S. A., J. S. Kim, J. R. Sage, A. W. Thomas, K. N. Skillicorn and S. G. Anagnostaras (2013). "The competitive NMDA receptor antagonist CPP disrupts cocaine-induced conditioned place preference, but spares behavioral sensitization." Behav Brain Res 239: 155-163.

Djakovic, S. N., E. M. Marquez-Lona, S. K. Jakawich, R. Wright, C. Chu, M. A. Sutton and G. N. Patrick (2012). "Phosphorylation of Rpt6 regulates synaptic strength in hippocampal neurons." J Neurosci 32(15): 5126-5131.

Djakovic, S. N., L. A. Schwarz, B. Barylko, G. N. DeMartino and G. N. Patrick (2009). "Regulation of the proteasome by neuronal activity and calcium/calmodulin- dependent protein kinase II." J Biol Chem 284(39): 26655-26665.

Hamilton, A. M., W. C. Oh, H. Vega-Ramirez, I. S. Stein, J. W. Hell, G. N. Patrick and K. Zito (2012). "Activity-dependent growth of new dendritic spines is regulated by the proteasome." Neuron 74(6): 1023-1030.

Hamilton, A. M. and K. Zito (2013). "Breaking it down: the ubiquitin proteasome system in neuronal morphogenesis." Neural Plast 2013: 196848.

Hernandez, P., Sadeghian, K, Kelley, AE (2002). "Early consolidation of instrumental learning requires protein synthesis in the nucleus accumbens." Nature Neuroscience 5: 1327-1331.

Hershko, A. and A. Ciechanover (1998). "The ubiquitin system." Annu Rev Biochem 67: 425-479.

Howell, K. K., B. R. Monk, S. A. Carmack, O. D. Mrowczynski, R. E. Clark and S. G. Anagnostaras (2014). "Inhibition of PKC disrupts addiction-related memory." Frontiers in Behavioral Neuroscience 8.

Kisselev, A. F. and A. L. Goldberg (2005). "Monitoring activity and inhibition of 26S proteasomes with fluorogenic peptide substrates." Methods Enzymol 398: 364- 378.

Mabb, A. M. and M. D. Ehlers (2010). "Ubiquitination in postsynaptic function and plasticity." Annu Rev Cell Dev Biol 26: 179-210.

Massaly, N., Dahan, L, Baudonnat, M, Hovnanian, C, Rekik, K, Solinas, M, David, V, Pech, S, Zajac, JM, Roullet, P, Mouledous, L, Frances, B (2013). "Involvement of

protein degradation by the ubiquitin proteasome system in opiate addictive behaviors." Neuropsychopharmacology 38: 596-604.

Reaney, S., Johnston, LC, Langston, WJ, Di Monte, DA (2006). "Comparison of the neurotoxic effects of proteasomal inhibitors in primary mesencephalic cultures." Exp Neurol 202: 434-440.

Ren, Z., Liu, MM, Xue, YX, Ding, ZB, Xue, LF, Zhai, SD, Lu, L (2013). "A critical role for protein degradation in the nucleus accumbens core in cocaine reward memory." Neuropsychopharmacology 38: 778-790.

Robinson, T., Berridge, KC (2008). "The incentive sensitization theory of addiciton: some current issues." Philosophical Transactions of the Royal Society 363: 3137- 3146.

Robinson, T., Kolb, B (1997). "Persistent structural modifications in nucleus accumbens and prefrontal cortex neurons produced by previous experience with amphetamine." Journal of Neuroscience 17: 8491-8497.

Russo, S. J., D. M. Dietz, D. Dumitriu, J. H. Morrison, R. C. Malenka and E. J. Nestler (2010). "The addicted synapse: mechanisms of synaptic and structural plasticity in nucleus accumbens." Trends Neurosci 33(6): 267-276.

Schafe, G., LeDoux, JE (2000). "Memory consolidation of auditory pavlovian fear conditioning requires protein synthesis and protein kinase A in the amygdala." Journal of Neuroscience 20: RC96.

Schmittgen, T. D. and K. J. Livak (2008). "Analyzing real-time PCR data by the comparative C(T) method." Nat Protoc 3(6): 1101-1108.

Steketee, J. (2003). "Neurotransmitter systems of the medial prefrontal cortex: potential role in sensitization to pscyhostimulants." Brain Research Reviews 41: 203-228.

Thomas, M., Beurrier, C, Bonci, A, Malenka, RC (2001). "Long-term depression in the nucleus accumbens: a neural correlate of behavioral sensitization to cocaine." Nature Neuroscience 4: 1217-1223.

Volkow, N. D., J. S. Fowler, G. J. Wang and R. Z. Goldstein (2002). "Role of dopamine, the frontal cortex and memory circuits in drug addiction: insight from imaging studies." Neurobiol Learn Mem 78(3): 610-624.

11 Figure 3.1a: Cocaine increases Rpt6 S120 phosphorylation and proteasome activity via the dopaminergic pathway. (A) , Dissociated cortical neurons (DIV >17) were treated with vehicle or cocaine (1 and 5uM, 24h) and chymotrypsin-like proteasome activity was measured in lysates with the fluorogenic substrate Suc-LLVY-AMC. Graph depicts rate substrate cleavage (mean ± S.E. fluorescence) normalized to control-treated neurons. Cocaine treatment (5uM, 24h) significantly increases proteasome activity (*, ANOVA: F(2,22) = 11.81, P = 0.001; unpaired Student’s t test: p<0.05, n = 9). (B) , Representative western blot of lysates in (A) probed with phosphor-Rpt6 pS120, total Rpt6, total CaMKIIα, and 20S core antibodies. (C) , Quantification of phosphor-Rpt6 blot in (B). Cocaine increases Rpt6 S120 phosphorylation in a dose dependent-manner. (D) , Dissociated cortical neurons (DIV >17) were treated with vehicle, methamphetamine (500uM, 24h), SERT inhibitor citalopram (25µM, 24h) or NET inhibitor atomoxetine (25µM, 24h) and chymotrypsin-like proteasome activity was measured in lysates with the fluorogenic substrate Suc-LLVY-AMC. Graph depicts rate substrate cleavage (mean ± S.E. fluorescence) normalized to control-treated neurons. Methamphetamine treatment significantly increases proteasome activity (*, ANOVA: F(3,28) = 4.204, P = 0.014; unpaired Student’s t test: p<0.001, n = 9), while neither citalopram (p = 0.73, n = 6) nor atomoxetine (p = 0.40, n = 6) had any significant effect. (E), Representative western blot of dissociated cortical neuron lysates (DIV=21) compared to whole brain lysate and probed with tyrosine hydroxylase antibodies (Millipore AB152) to show relative expression levels of TH in dissociated cortical neuron cultures.

12 Figure 3.1b: Cocaine increases Rpt6 S120 phosphorylation and proteasome activity via the dopaminergic pathway. (F) , qPCR of lysates from dissociated cortical neurons (DIV=21) and whole brain cortex show expression of tyrosine hydroxylase in both lysates, indicating monaminergic systems in dissociated cortical neuronal cultures. (G) , Schematic of drug administration protocol. After a habituation period on day 1 and 2, cocaine (15 mg/kg, i.p) or saline was administered on days 3 thru 7 and a final injection was delivered on day 8 immediately prior to brain extraction. (H) , Whole brain lysates were prepared and proteasome activity and Rpt6 phosphorylation were examined as above. Proteasome activity increased in brains of cocaine-injected animals relative to saline (*, p<0.05, Mann-Whitney u test; U = 0, P = 0.029, n = 4). (I) , Representative western blot of lysates in (H) probed with phospho Rpt6 pS120, total Rpt6 and 20S core antibodies shows that a concomitant increase in Rpt6 S120 phosphorylation occurs in cocaine-treated animals when compared to saline. (J) , Quantification of phosphor-Rpt6 blot in (I) shows a significant increase in Rpt6 phosphorylation in cocaine treated animals.

13 Figure 3.2: Generation of Rpt6 S120D Knockin (KI) mice. (A) , Schematic of targeting strategy. Genomic DNA structure of psmc5 (Rpt6 ) region is shown. The targeting vector was designed such that the long homology arm (LA) extends 5.98kb 3’ to the first point mutation (asterisk; AG ‰ GA, S120D) in exon 5. The FRT flanked Neo resistance cassette was inserted 463 bp 5’ to the point mutation. The short homology arm (SA) extends 2.79 kb 5’ to the FRT flanked Neo cassette. (B) , Tail genomic DNA was analyzed by PCR screening for genotyping and to verify deletion of the Neo cassette. (C) , Electropherograms confirming the presence of the mutation in homozygous Rpt6 S120D mutant male mice. (D) , Western blot analysis of affinity purified 26S proteasomes from Rpt6 S120D KI mice and wt littermates and probed with phospho Rpt6 pS120, total Rpt6 and 20S core antibodies. (E) , Representative images of Nissl stained fixed whole brain coronal sections of 60 day old mice.

14 Figure 3.3: Cocaine increases Rpt6 phosphorylation and proteasome activity in NAc and PFC in wildtype but not Rpt6 S120D KI mutant mice. (A), Schematic of drug administration protocol. After a habituation period on days 1 and 2, cocaine (15 mg/kg, i.p) or saline was administered on days 3-7 and a final injection was delivered on day 8 immediately prior to brain extraction. (B) , 1mm coronal sections were obtained using an acrylic matrix and tissue punches were taken from NAc and PFC. Depicted are the centers of individual punches. Lysates were prepared and chymotrypsin-like proteasome activity and Rpt6 phosphorylation was examined. (C) thru (E) (NAc) and (F) thru (H) (PFC), Cocaine increases proteasome activity in NAc and PFC of wildtype ( C,F ) but not Rpt6 S120D KI mutant mice ( D,G ), measured by rate of substrate cleavage normalized to saline-treated animals (*, p<0.05, Mann-Whitney u test; n=4: NAc: U=0 and 7, P=0.029 and 0.886 for wildtype and Rpt6 S120D cocaine to saline-treatments, respectively; and *, p < 0.05, Mann-Whitney u test; n=3: U=0 and 8, p=0.029 and 0.999 for wildtype and Rpt6 S120D cocaine to saline-treatments, respectively). ( E,H) , Representative western blot of lysates from wildtype NAc (C) and PFC (G) probed with phospho Rpt6 pS120, total Rpt6, and tubulin antibodies shows an increase in the NAc and increase in PFC for Rpt6 S120 phosphorylation in cocaine-treated animals compared to saline-treated controls.

15 Figure 3.4: Disruption of locomotor sensitization in S120D mice. (A) , S120D mice did not sensitize across 4 days of cocaine administration (n=18 WT-Cocaine, n=11 S120D- Cocaine, n=6 WT-Saline, n=5 S120D-Saline). The average distance traveled for each session (± SEM) is depicted. All mice showed a similar level of activity during habituation and day 1. However, during days 2-4, locomotor activity of WT-Cocaine mice was elevated compared to S120D-Cocaine mice and mice receiving saline (15 mg/kg, i.p; p<0.005). (B) , Locomotor activity during day 1. The data represent the distance traveled per minute of the session (± SEM). There were no differences between the four groups ( p=0.51). (C) , Locomotor activity during day 4. Distance traveled per minute of the session (± SEM) is depicted. While the response to cocaine was elevated in WT-Cocaine mice, S120D-Cocaine mice did not exhibit locomotor activity different from saline groups (WT-Cocaine, S120D-Cocaine Fisher’s PLSD p=0.01; S120D- Cocaine, WT-Saline, S120D-Saline Fisher’s PLSD p>0.5). (D) , Sensitization was measured as the difference in locomotor activity between day 4 (sensitized response) and day 1 (acute response). The average difference in distance traveled (± SEM) per group is shown. S120D-Cocaine mice did not develop sensitization and did not differ from mice that received saline (S120D-Cocaine/S120D-Saline, S120D-Cocaine/WT-Saline, p>0.7 WT-Cocaine mice displayed sensitized, elevated locomotor activity that differed from S120D-Cocaine mice ( p<0.01).

Acknowledgements

Chapter 3, in part is currently under review for publication of material. Gonzales

FR, Howell KK, Dozier LE, Anagnostaras SG, and Patrick GN. The dissertation/thesis author in the primary investigator and author of this material.

Chapter IV: Phosphorylation of the 19S regulatory particle ATPase subunit, Rpt6, modifies susceptibility to proteotoxic stress and protein aggregation

Esther Magdalena Marquez-Lona 1*a , Ana Lilia Torres-Machorro 2*b , Frankie R. Gonzales 1, Lorraine Pillus 2, Gentry N. Patrick 1*

1 Section of Neurobiology, Division of Biological Sciences, University of California San Diego, La Jolla, California, United States of America, 2 Section of Molecular Biology and UCSD Moores Cancer Center, Division of Biological Sciences, University of California San Diego, La Jolla, California, United States of America *These authors contributed equally to this work. Current address: University of Tennessee Health Science Center, Anatomy and Neurobiology, Department, Memphis, Tennessee, United States of America

A. Abstract

The ubiquitin proteasome system (UPS) is a highly conserved and tightly regulated biochemical pathway that degrades the majority of proteins in eukaryotic cells. Importantly, the UPS is responsible for counteracting altered protein homeostasis induced by a variety of proteotoxic stresses. We previously reported that Rpt6, the

ATPase subunit of the 19S regulatory particle (RP) of the 26S proteasome, is phosphorylated in mammalian neurons at serine 120 in response to neuronal activity.

Furthermore, we found that Rpt6 S120 phosphorylation, which regulates the activity and distribution of proteasomes in neurons, is relevant for proteasome-dependent synaptic remodeling and function. To better understand the role of proteasome phosphorylation, we have constructed models of altered Rpt6 phosphorylation in S. cerevisiae by introducing chromosomal point mutations that prevent or mimic phosphorylation at the conserved serine (S119). We find that mutants which prevent Rpt6 phosphorylation at this site (rpt6-S119A), had increased susceptibility to proteotoxic stress, displayed abnormal morphology and had reduced proteasome activity. Since impaired proteasome function has been linked to the aggregation of toxic proteins including the Huntington’s

disease (HD) related huntingtin (Htt) protein with expanded polyglutamine repeats, we evaluated the extent of Htt aggregation in our phosphor-dead (rpt6-S119A) and phosphomimetic (rpt6-S119D) mutants. We showed Htt103Q aggregate size to be significantly larger in rpt6-S119A mutants compared to wild-type or rpt6-S119D strains.

Further- more, we observed that phosphorylation of endogenous Rpt6 at S119 is increased in response to various stress conditions. Together, these data suggest that Rpt6 phosphorylation at S119 may play an

B. Introduction

The balance between synthesis and degradation is crucial for maintaining protein homeostasis. Many studies have shown that altered protein homeostasis occurs during normal aging and age-related disease [1–6]. In post-mitotic neurons, genetic alterations together with oxidative stress and other forms of cellular damage are thought to underlie many of the abnormalities that contribute to age-related and neurodegenerative diseases.

The inability to repair or remove damaged and aggregated proteins is a key feature of the pathogenesis of many diseases of the central nervous system [7–10]. Whereas ribosomes and chaperones ensure proper synthesis and folding of proteins, the ubiquitin proteasome system (UPS) and autophagosomal/ lysosomal pathways control the majority of cellular protein degradation.

The UPS is an evolutionarily conserved and tightly regulated biochemical pathway involved in protein degradation [2, 3, 11]. The selective degradation of proteins via the UPS involves three steps: recognition of the target protein via specific signals, modification of the target pro tein with ubiquitin chains, and delivery of the target protein

to the 26S proteasome, a multi- subunit protein complex that degrades the ubiquitinated proteins [3, 12]. The 26S proteasome is a large energy-dependent protease consisting of a proteolytic 20S core particle (CP) and a 19S regulatory particle (RP). A base and a lid form the RP of the proteasome. A heterohexameric ring formed by six ATPases (Rpt1-6) constitutes part of the base, which makes direct contact with the proteolytic CP. The RP is necessary for interaction and unfolding of ubiquiti nated substrates and gating of the CP to ensure proper degradation [13–15].

Accumulation of aberrant or misfolded proteins occurs in many progressive neurodegener ative disorders [8–10, 16, 17]. Alzheimer’s disease (AD), Huntington’s disease (HD), Parkin son’s disease (PD) and other clinically distinct neurodegenerative disorders are considered ’proteinopathies’ of the nervous system. However, it remains unresolved whether neuronal dysfunction and disease occur as a result of toxic protein aggregation and pathogenic activities or whether protein aggregation interferes with the degradation of other proteins [18, 19]. Although numerous reports indicate a decline in proteasome function in age-related and neurodegenerative diseases (reviewed in [20, 21]), a mechanistic understanding of how protea some activity decreases remains unclear.

We previously described the activity-dependent regulation of the 26S proteasome in mam malian neurons including the key step of phosphorylation of the 19S ATPase subunit, Rpt6, by Ca 2+/calmodulin-dependent protein kinase II α (CaMKIIα).

Additionally, we have reported a crucial role of Rpt6 phosphorylation on proteasome- dependent regulation of synaptic strength and synaptic remodeling [22–24]. S. cerevisiae has been successfully utilized as a model organism to explore aging research. Indeed, studies analyzing human protein orthologs, or expression of heterologous human disease-

associated proteins have revealed the value of yeast as model to study molecular mechanisms of disease [25–29]. We confirmed that Rpt6 is phosphorylated in yeast, and that this modification is increased upon stress. In an effort to further understand the functional significance of proteasome phosphorylation, we next generated discrete models of altered Rpt6 phosphorylation and 26S proteasome function by manipulating a single residue of a single proteasome subunit in S. cerevisiae. We introduced point mutations that prevent or mimic phosphorylation at the conserved serine (S119) of Rpt6. We found that mutants which prevent Rpt6 phosphorylation at this site (rpt6-S119A mutants) have reduced proteasome activity, increased susceptibility to proteotoxic stress and display abnormal mor phology. Since impaired proteasome function has been linked to the aggregation of toxic pro teins including the HD related huntingtin (Htt) protein with expanded polyglutamine (polyQ) repeats, we evaluated the extent of Htt aggregation in our phospho-dead (rpt6-S119A) and phosphomimetic (rpt6-S119D ) mutants. We showed

Htt103Q aggregate size to be significantly larger in rpt6-S119A mutants compared to wild-type or rpt6-S119D mutants. Furthermore, rpt6-S119A mutants expressing Htt103Q displayed altered senescence profiles compared to wild-type or rpt6-S119D mutants.

Together, these data suggest that dynamic phosphorylation of Rpt6 at S119 in yeast

(S120 in mammals) may play an important, conserved function in the ability of the proteasome to counteract proteotoxic stress and protein aggregate formation.

C. Materials and methods

Yeast strains and knock-in strategy

We constructed an N-terminal 9xMyc-tagged RPT6 strain as previously described

[30]. Since RPT6 is essential, chromosomal gene replacement was performed in a wild- type diploid strain (LPY15916, see S1 Table). To direct integration, a PCR product was prepared using oligonucleotides oLP1751 and 1752 (S3 Table) and pLP1956 (POM20) plasmid DNA as a template (S2 Table). The strain generated was LPY17011. To allow expression of the tagged protein, the kanMX marker was removed by Cre recombinase expression, induced from plasmid pLP192 (Life Technologies pBS39; [31]) by growth in

1% galactose. Myc-Rpt6 expression was confirmed by protein immunoblotting. The myc-RPT6 heterozygous diploid (LPY17208) strain was sporulated and dissected to obtain LPY18040, the haploid myc-RPT6 strain (S1 Table).

To construct the site-specific substitution mutants, an rpt6Δ::kanMX strain

(LPY16270) was dissected from the yeast heterozygous diploid collection [32] using a covering plasmid containing wild-type RPT6 (pLP2636). The kanMX marker of

LPY16270 was replaced with a natMX cassette using pLP1630/p4339 (kindly provided by the C. Boone laboratory). ApaI digested pLP2858 and pLP2859 were used to replace rpt6Δ::natMX in LPY16270 with the rpt6-S119A and rpt6-S119D mutant versions of 5’-

RPT6-kanMX-N9xMyc-RPT6-RPT6-3’. Cre recombinase was induced as above using pLP196 (Life Technologies), and cells without the covering plasmid were recovered by selecting for growth on 5-FOA. The integrated mutants were verified by sequencing and immunoblot and subsequently backcrossed to wild-type to generate LPY19188

(rpt6Δ::9myc-rpt6-S119A) and LPY19193 (rpt6Δ::9myc-rpt6-S119D) (S1 Table).

Plasmids

The wild-type RPT6 gene was subcloned from a genomic tiling library [33] into pRS316 (pLP126) using the 1.8 kb Eco RI-KpnI genomic fragment (pLP2636). A PCR product containing the 5’-RPT6-kanMX-N9xMyc-RPT6-RPT6-3’ construct from

LPY17011 was cloned into TOPO (Life Technologies), generating pLP2855 (S2 Table).

The construct was amplified by PCR sewing using oLP1707 + oLP883 for product one, and oLP1708 + oLP236 for product two (S3 Table). Both products were mixed and amplified with oLP1707 and 1708. The construct was verified by sequencing. The RPT6 mutations S119A and S119D were introduced into pLP2855 by site directed mutagenesis using oligonucleotides listed in S3 Table and verified by sequencing. The resulting plasmids were pLP2858 ( rpt6-S119A ) and pLP2859 ( rpt6-S119D ) (S2 Table).

Plasmids containing the GFP-tagged N-terminal portion of the human Huntingtin protein were from Addgene (#1177, 1179 and 1180; [27]). Plasmids containing mHtt

(Htt25Q, 72Q and 103Q) were transformed into RPT6 mutant strains (LPY6496,

LPY18040, LPY19188 and LPY19193) by standard methods [34].

Immunoprecipitations and immunoblots

200 ml cultures were grown to A600 0.8, collected by centrifugation and washed with 1X PBS. Pellets were lysed in 250 μL of cold IP lysis buffer (50mM Tris-HCl pH7.5, 100mM NaCl, 1mM EDTA, 0.1% Triton X-100, 10% Glycerol supplemented with Sigma COMPLETE Protease Inhibitor cocktail and 1X NME) by bead beating.

Cleared lysates were incubated overnight with 25mg of mAb anti-Myc. Protein A-

Sepharose (Sigma-Aldrich) previously equilibrated in IP buffer was added and samples

were rotated for 4 hours at 4˚C. Beads were washed 3 times with IP wash buffer (50mM

Tris-HCl pH 7.5, 100mM NaCl and 1mM EDTA). Samples were boiled for 10 min in sample buffer and resolved on 10% SDS-PAGE. Proteins were transferred to a 0.2μm nitrocellulose membrane in a Bio-Rad Trans-Blot SD semi-dry transfer box, blocked in

5% milk in TBS-Tween and immunoblotted for either anti-Myc 1:4,000 (Santa Cruz, c-

40; Myc1-9E10.2 hybridoma: B lymphocyte cells from Spleen: CRL-1729), pAb anti- phospho Rpt6 1:4,000 [23], β- tubulin 1:20,000 (described in [35]) or mouse anti-GFP

1:5,000 (NeuroMab, 75–131) and anti-20S antibodies (Enzo, BML-PW8195-0025).

Growth assays

Dilution assays were performed as previously described [36] and represented five-fold serial dilutions from A600 0.5 units of cells. Images were captured after 2–5 days of growth. Stress plates tested included canavanine 1μg/ml (in arg- media), ethanol

8% (in synthetic complete (SC) media), formamide 2.5% (SC), CdCl2 20μM (SC) and

YPD (Yeast extract-Peptone-Dextrose medium) at 37˚C. Camptothecin sensitivity was assayed using 20μg/ml in DMSO, added to YPD plates buffered with 100 mM potassium phosphate to pH 7.5 [37].

Viability assays

Assays were performed as previously described [38]. Freshly transformed strains

(with pLP3012, pLP3013 or pLP3014) were grown to saturation and diluted to A600 0.2 in ura- selection medium. Duplicates of 400 cells were plated on ura- plates from the same starting cultures after 1, 5, 7, 10, 12 and 15 days of growth at 30˚C. Cells were counted before plating using a hemocytometer. After 3 days at 30˚C, colony forming

units (CFU) were counted from three independent experiments. Results were normalized to vector transformed myc-RPT6 -wild-type strain.

Flow cytometry

As described in [39], cells were ethanol fixed (A600 0.8) during logarithmic growth. A total of 30,000 propidium iodide-stained cells of each strain were analyzed by flow cytometry (BD Accuri C6) after sonication.

Bud profiling

Morphology was assessed by light microscopy for 2500–3000 cells per mutant strain from three independent experiments. Either single cells or cells with one bud were registered as normal. Two or more fused cells, cells with two or more buds or increased size cells when compared to wild type were registered as abnormal. Statistical analysis consisted of average and one-way ANOVA p < 0.1.

Native-PAGE peptidase and 26S fluorogenic peptidase activity assays

Strains were grown to both A600 5.0 and 0.6, spun down and freeze-dried with liquid nitrogen as previously described [40]. Briefly, cells were rehydrated using lysis buffer (25mM HEPES-- KOH pH 7.4, 5mM MgCl2, 10% Glycerol, 1mM DTT, 2mM

ATP), and ultracentrifuged at 100,000xg 30 min to obtain whole cell lysates. Native

PAGE was performed as previously described [41]. 3–12% gradient gels run at constant voltage for 4 hr at 4˚C were soaked in developing buffer (50mM Tris-HCl pH 7.4, 5mM

MgCl2, 0.5mM EDTA, 1mM ATP) containing 50μM Suc-LLVY-AMC substrate.

Stimulation of CP gate opening was performed using 0.02% SDS for 30 min. All images were obtained using Protein Simple FluorChem E imaging system. Gels were then transferred to nitrocellulose and probed with anti-myc and anti-20S antibodies.

Chymotrypsin-like activity of 26S proteasomes from whole cell lysates was monitored over time (20 sec/ 2 hr) using Suc-LLVY-AMC substrate (Enzo) as previously described

[42]. The peptidase assay was run in triplicate using a normalized protein concentration monitored on a Perkin-Elmer HTS7000 BioAssay Microplate reader. Kinetic rates from each mutant strain were plotted from the averaged triplicate data.

D. Results

1. Rpt6 is phosphorylated in S. cerevisiae in response to stress

Several subunits of the proteasome have been reported to be phosphorylated in vivo [43, 44], however, relatively few studies have addressed how phosphorylation may regulate proteasome function. Phosphorylation might regulate proteasome subunit assembly, disassembly, or distinct proteasome activities, such as peptidase activity,

ATPase activity, ubiquitin binding, and/ or deubiquitinating activities. Further, phosphorylation might regulate the cellular distribution and trafficking of proteasomes and association with proteasome-interacting proteins. In mammalian neurons, we found that Rpt6 is phosphorylated at serine 120 (S120) in a neuronal activity-dependent fashion

[22]. It is clear that Rpt6 S120 phosphorylation increases proteasome activity and promotes the redistribution of proteasomes to synapses [22, 23, 45], although further mechanistic details remain under study.

In S. cerevisiae RPT6 has been previously studied with conditional alleles in which variable phenotypes were observed, depending on the strength of each mutation

[46, 47]. These studies were important for defining Rpt6 function in RP’s base assembly and its interaction with the core particle, which is indispensable for 26S proteasome

assembly [48]. Similar mutants of other RP ATPases had distinct phenotypes [49], showing specialized roles in protein degradation, chromatin regulation, gene expression and DNA damage repair, among other cellular processes [50–55]. To date however, little is known about the role of phosphorylation in the regulation of RP ATPase subunits.

The sequence of proteasomal subunits is conserved from yeast to humans. The sequence identities of the RP base ATPase subunits range from 66%-76%, with Rpt6

74% identical to its human ortholog [56]. Furthermore, the sequence surrounding S120 in mammals and S119 in yeast is 81% identical (Fig 1A). To address the functional relevance of Rpt6 phosphorylation on proteasome function in vivo , we performed studies in yeast. Two strains with chromosomally integrated mutations were constructed. The first substituted Ser119 of Rpt6 with alanine ( rpt6 -S119A ) creating a non-modifiable phospho-dead residue; the second mutant strain was generated with an aspartic acid

(rpt6 -S119D ) substitution as a phosphomimetic residue (S1 Table). Because the carboxy- terminus of Rpt6 contributes to proteasome assembly by establishing direct contacts with the CP [40, 57] and with protein chaperone Rpn14 [58], we constructed an N-terminal

Myc-epitope tagged Rpt6 to facilitate immuno-precipitation while retaining C-terminal function. Wild-type ( myc-RPT6 ) and both mutants ( myc-rpt6-S119A and -S119D ) were expressed from the endogenous chromosomal locus at similar levels (Fig 1B). To determine if Rpt6 was phosphorylated at S119 in yeast, Myc-Rpt6 was immunoprecipitated using a mouse anti-Myc antibody, and probed with an anti-phospho

S120 Rpt6 pAb (pS120) by immunoblotting. We observed little signal to no signal for the myc-rpt6-S119A mutant (Fig 1C). In contrast, strong immunoreactivity was detected for

the Myc-Rpt6 protein (Fig 1C) indicating that Rpt6 is phosphorylated in yeast at S119

(Fig 1C).

16 Figure 4.1: The rpt6-S119A mutants had increased susceptibility to proteotoxic stress and displayed abnormal morphology. (A) Alignment of Rpt6 sequences from rat, human and yeast indicate conservation of the S120 phosphorylation site. Note the presence of a serine at position 117 in yeast. This residue may contribute to some cross- immunoreactivity for the pS120 antibody in the myc-rpt6-S119A strain (see below). (B) Qualitative analysis of myc-tagged Rpt6 (myc-Rpt6) expression. Immunoblot depicts similar levels of expression in strains tested. (C) Representative immunoblot of myc-RPT6 and myc-rpt6-S119A lysates immunoprecipitated with α-myc antibodies and resolved on SDS-PAGE. The anti-Rpt6 pS120 antibody recognizes wild-type myc- Rpt6, whereas little to no signal is observed for the myc-rpt6-S119A mutant, suggesting that Rpt6 is phosphorylated in yeast. (D) Increased sensitivity to proteotoxic stress is observed in the myc-rpt6-S119A mutant. The fitness of the myc-rpt6-S119D strain was comparable to myc- RPT6. The myc-rpt6-S119A strain was sensitive to elevated temperature, ethanol and canavanine. Arg- medium serves as a control for canavanine, which is a toxic analog of arginine. (E) Phosphorylation of endogenous Rpt6 is increased upon stress. Protein lysates from cells grown under normal and stress conditions (37˚ and canavanine (1ug/ml;2h)) were probed with the anti-pS120 antibody. (F) Representative micrographs of myc-RPT6 strains during log phase indicate abnormal morphology in the myc-rpt6-S119A mutant. (G) Quantitative analysis of morphological studies in (F). Graphs denote 3 independent experiments. Abnormal morphology is increased to 10% of the total population in myc-rpt6- S119A.

The rpt6-S119A phospho-dead mutant is sensitive to conditions that induce protein misfolding Proteasome function is required for maintaining protein homeostasis and cell survival during protein stress. To characterize the importance of Rpt6 phosphorylation on proteasome function, we evaluated growth of the RPT6 phospho- mutant strains under proteotoxic stress. As shown in Fig 1D, the growth and fitness of myc-RPT6 and the phosphomimetic mutant, myc-rpt6-S119D, was comparable to wild- type RPT6 (See also S1 Fig). We did, however, find that the expression of the phosphomimetic rpt6-S120D on a 2μ plasmid in the RPT6 null background promoted greater resistance to proteotoxic stress (ethanol 8%) when compared to RPT6 or rpt6-

S120A (also expressed on a 2μ plasmid in the RPT6 null background) (S1 Fig). In contrast, the chromosomal myc-rpt6-S119A phospho-dead mutant was sensitive to growth at

37˚C, to ethanol and to canavanine. The sensitivity phenotypes of the myc-rpt6-

S119A mutant are similar to those previously found in strains with mutations in the

ATPase domain of Rpt6 [40, 59]. Since growth was comparable between myc-RPT6 and myc-rpt6-S119D, we suggest that phosphorylation of Rpt6 at S119 is important for maintaining proteasome function during cellular protein stress. In addition, when myc- rpt6-S119A was evaluated as a heterozygote, no sensitivity to proteotoxic stress was observed (S2B Fig), demonstrating that the S119A mutation in RPT6 was recessive.

The myc-rpt6-S119A mutant proved to be sensitive to other agents that disrupt protein structure, including formamide and cadmium chloride (S2A Fig). Furthermore, this mutant was sensitive to the DNA damage inducing agent camptothecin (CPT), an

inhibitor of topoisomerase I (S2A Fig). Collectively, these data indicate that alanine substitution at S119, preventing its posttranslational modification, results in growth defects in stress conditions similar to those of null alleles of non-essential proteasome subunits and to strains carrying thermosensitive alleles of CP or RP subunits [48, 50, 54,

55, 60].

To confirm the biological relevance of the findings above, we evaluated phosphorylation of myc-Rpt6-S119 in response to stress. When cells were grown at either high temperatures or in the presence of proteotoxic stress agents, we observed an increase in S119 phosphorylation relative to cells grown under normal conditions (Fig 1E). This suggests that phosphorylation of Rpt6 at S119 may have a role in regulating proteasome function as a response to stress mechanisms.

Because of established roles for protein turnover in cell cycle regulation [61, 62] the cell cycle profiles of both rpt6 phospho-mutants were also analyzed. As shown in S3

Fig, a defect in progression through the G2/M phase of the cell cycle was observed in all myc-RPT6 strains tested, which was in agreement with previous reports [46]. This shows that the N-terminal

Myc-epitope may modestly interfere with Rpt6 function in the G2/M transition.

This effect appears independent of any effects on proteasome function in proteotoxic and cellular stress conditions, and may thus point to a previously unsuspected role for the N- terminus. Notably, the delay in the G2/M transition was independent of S119 mutations

(S3 Fig). Typically, a G2/ M block is characterized by a high percentage of budded cells.

All three myc-RPT6, myc-rpt6-S119A and myc-rpt6-S119D strains had between 40–55% budded cells (Fig 1F). However, we observed a significant increase in the multi-budded

phenotype in myc-rpt6-S119A mutants (Fig 1G) suggesting additional defects in the control of cell division. Together, these data indicate that the myc-rpt6-S119D mutant strain had similar growth phenotypes to the myc-RPT6 strain, whereas the myc-rpt6-

S119A mutant was sensitive to proteotoxic and other cellular stressors and had an exacerbated budding defect.

2. Decreased proteasome activity in the myc-rpt6 S119A mutant

It was important to determine if proteasome function was altered in the myc-rpt6-

S119A mutant. We first evaluated proteasome activity from lysates of wild-type and mutant strains by native gel electrophoresis followed by an in-gel activity assay with the fluorogenic proteasome substrate Suc-LLVY-AMC. The myc-rpt6-S119A mutant had significantly decreased proteasome activity associated with both singly and doubly capped proteasomes (20S CP associated with one and two 19S RP, respectively), although the total level of proteasomes was comparable (Fig 2A and 2B). We additionally monitored the chymotrypsin-like activity of 26S proteasomes in whole cell lysates by

Suc-LLVY-AMC cleavage over time. We found that proteasome activity was significantly decreased in lysates of myc-rpt6-S119A mutant when compared to myc-

RPT6 or myc-rpt6-S119D mutant strains (Fig 2C). We did not see enhanced proteasome activity in myc-rpt6-S119D mutant strain compared to myc-RPT6. Thus, preventing Rpt6 phosphorylation at S119 decreased proteasome function and increased susceptibility to proteotoxic stress.

3. Altered Rpt6 phosphorylation exacerbates Htt polyQ aggregation and proteotoxicity

Since the cloning of the Huntington’s disease (HD) gene in 1993, transgenic models of HD have been constructed in multiple model organisms including yeast, nematodes, flies and mice [63]. S. cerevisiae has proven to be an excellent organism in which to study molecular mechanistic aspects of human neurodegenerative disease [28,

64]. In particular, HD which is caused by the expansion of a tri-nucleotide CAG repeat

(poly-Q expansion) at the N-terminus of exon 1 of mutant

Huntingtin (Htt), has been well-characterized [27, 65, 66]. The poly-Q expansion leads to misfolding of Htt and formation of mutant Htt-containing protein aggregates [27, 64].

Several studies have shown that proteasome activity is decreased in the affected brain regions of HD patients [67], whereas in yeast, expression of a mutant Htt fragment

(Htt72Q or higher) causes aggregate formation, transcriptional dysregulation, cellular toxicity, perturbations in kynurenine pathway metabolites, increased reactive oxygen species (ROS), mitochondrial dysfunction, and defects in endocytosis and apoptotic events [27, 64–66]. Importantly, some models of HD recapitulate the relationship of mutant Htt-containing protein aggregates and decreased proteasome activity.

To determine if altered Rpt6 phosphorylation influences mutant Htt-induced defects in growth and the extent of protein aggregate formation, we transformed myc-

RPT6, myc-rpt6-S119A and myc-rpt6-S119D strains with plasmids to induce expression of a fragment of Htt with differing polyQ lengths (Htt25Q, Htt72Q and Htt103Q) (Fig

3A). The growth of myc- RPT6 and myc-rpt6-S119D strains expressing all mutant Htt polyQ variants upon challenge with canavanine was comparable to cells transformed with the vector control (Fig 3B). There was a slight decrease in growth with increasing length

of the polyQ repeat length (Fig 3B). In contrast, the myc-rpt6-S119A strain displayed significantly reduced growth with increasing polyQ repeat length (Fig 3B). We subsequently evaluated protein aggregate formation of mutant GFP-Htt in all RPT6 strains. As shown in Fig 3C, expression of Htt25Q was evenly distributed throughout the cell, whereas Htt72Q formed small protein aggregates in all strains, as previously described in a wild-type strain [27]. Htt103Q expression in the myc-rpt6-S119A mutant formed considerably larger mHtt aggregates compared to myc-RPT6 and myc-rpt6-

S1119D (Fig 3C and 3D). Together these data demonstrate that clearance of mutant

Htt103Q protein aggregates is compromised in the myc-rpt6-S119A mutant with decreased proteasome function.

Figure 4.2: Proteasome activity was decreased in the myc-rpt6-S119A mutant. (A) Lysates from strains in stationary phase (A600 >5.0) were subjected to native gel electrophoresis followed by an in-gel activity assay with the fluorogenic proteasome substrate Suc-LLVY- AMC. The myc-rpt6-S119A mutant displays lower activity by native in-gel analysis (before and after 20S gate opening by SDS). (B) Gels from (A) were transferred to nitrocellulose and probed with anti-myc and anti-20S antibodies. Immunoblotting indicated similar levels of proteasomes. Shown are representative in-gel and immunoblot analysis from 3 or more independent experiments. RP = Regulatory Particle, CP = Core Particle. (C) The chymotrypsin-like activity of 26S proteasomes in whole cell lysates of stationary phase cells (A 600 >5.0) was monitored by Suc-LLVY-AMC cleavage over time on microplate reader (ex. 360nm and em. 465nm). The mean ± SEM rate of Suc-LLVY-AMC cleavage is decreased in the myc-rpt6-S119A mutant compared to the WT and myc-rpt6-S119D strains. *p < 0.05, one way ANOVA. Data are from two biological samples (n = 2) with triplicate analysis.

4. Compromised senescence in myc-rpt6-S119A cells expressing Htt103Q

Yeast stationary phase growth or chronological life span (CLS) assays measure the survival of cell populations in post-mitotic or non-dividing phases. These metrics appear similar to those for mammalian cells such as neurons that do not divide or that have long non-mitotic rest phases [68, 69]. Because we found differences in protein aggregate size among RPT6 strains, we sought to simulate conditions analogous to the neuronal non-dividing state by analyzing how cell survival upon extended chronological age might be affected. The viability of all myc-

18 Figure 4.3: Altered Rpt6 phosphorylation and proteasome function exacerbated Htt polyQ aggregation and proteotoxicity. (A) Representative immunoblot (anti- GFP) of GFP-tagged (Htt) with different polyQ repeats (Htt25Q, Htt72Q and Htt103Q). Expression is similar in all myc-Rpt6 strains. (B) Expression of Htt72Q and Htt103Q increased the sensitivity of myc-rpt6-S119A to canavanine. Strains transformed with URA3 marked vector, Htt25, Htt72 and Htt103 plasmids were plated on canavanine to induce proteotoxic stress. The arg- ura- plates are controls. (C) Representative images of RPT6 mutants transformed with Htt25Q, Htt72Q and Htt103Q constructs. Larger GFP-Htt aggregates are observed in the myc-rpt6-S119A mutant compared to WT and the myc-rpt6-S119D mutant. (D) Quantification of aggregate size (total GFP signal within a cell) is presented for Htt103Q expression. Graphs depict mean aggregate size ± SEM (arbitrary units). **p < 0.01, one way ANOVA. n = 45 to 60 individual GFP- aggregates (cells) measured over 2 independent experiments.

RPT6 strains expressing vector control or mutant Htt103Q was measured to assess the influence of mutant Htt protein aggregation and senescence profiles. We grew all transformed strains in glucose-containing medium for up to 15 continuous days, plating on fresh medium after 1, 5, 7, 10, 12 and 15 days of growth. The colony forming units

(CFU), corresponding to cells able to exit senescence and resume growth were quantified after three days and the results were normalized to myc-RPT6 cultures transformed with vector (plated after one day of growth). The relative number of colonies indicated the survival of the strains, where the CFU number is proportional to the fitness of the strains after varying times in stationary phase. Vector-transformed myc-RPT6, myc-rpt6-S119D and myc-rpt6-S119A strains had a similar loss in viability after 5 days of growth, also comparable to that obtained for Rpt6 mutants transformed with Htt- 25Q (S4 Fig).

However, when GFP-Htt103Q was expressed, the initial viability of all tested strains after one day of growth was 30% lower compared to vector transformants. We propose that this phenotype could reflect a problem in establishing a senescent state, delaying cell death as the cells age. This phenotype is stronger in the myc-rpt6-S119A mutant (Fig

4B), suggesting that normal proteasomal function may be relevant in establishing a cell’s death programming during chronological aging.

E. Discussion

The UPS, which controls the majority of protein degradation in eukaryotic cells, is particularly important for counteracting altered protein homeostasis observed in age- related and neurodegenerative diseases. Indeed, the inability to rid cells of damaged

proteins is believed to be key to the pathogenesis of many neurodegenerative diseases including AD, PD and HD. In some cases these proteins form heavily ubiquitinated protein aggregates. One hypothesis is that an increased proteotoxic load impairs proteasome function, thereby making cells more susceptible to stress. Alternatively, the degradation of non-toxic proteins may become impaired, which then alters normal cellular function. Historically, proteasome inhibitors have been used to understand how proteasome dysfunction contributes to disease pathogenesis. However, these have a limited utility because they are toxic to cells and cause death by apoptosis, perhaps due to cell cycle arrest [42, 70–73]. Several recent studies have reported the phosphorylation of multiple proteasome subunits. We and others have shown that the ATPase RP subunit,

Rpt6, is phosphorylated at serine 120 by the CaMKIIα plasticity kinase in the mammalian

CNS [22, 45]. We reported that Rpt6 phosphorylation is important for synaptic strength and proteasome-dependent remodeling of dendritic spines [23, 24]. Yet, the physiological significance of how proteasome phosphorylation regulates the rate and selectivity of substrate degradation remains an important question that is the topic of ongoing investigation.

The budding yeast S. cerevisiae has been used successfully to model neurodegenerative diseases, not only to understand the molecular mechanisms involved but also to provide insight into potential therapeutic approaches for these diseases [74]. In this study, we created phospho-dead and phospho-mimetic mutants of Rpt6 serine 119 in yeast (serine 120 in mammals) and report phosphorylation of the 26S proteasome to be an important cis-regulatory posttranslational modification for the proteasome in counteracting proteotoxic stress. Using phosphospecific Rpt6 S120 antibodies we showed

that Rpt6 is phosphorylated at serine 119. This posttranslational modification is increased upon temperature and proteotoxic stress, under- scoring a possible role in regulation of

Rpt6 and proteasome function. In addition, rpt6-S119A mutants that prevent Rpt6 phosphorylation at this site have reduced proteasome activity.

19 Figure 4.4: Senescent myc-rpt6-S119A cells expressing Htt103Q had a delayed loss of viability relative to myc-RPT6 upon chronological aging . Strains were plated at 30˚C to assess viability after 1, 5, 7, 10, 12 and 15 days of growth in liquid medium. Graphs are the average of three independent experiments performed with duplicate samples. Values were plotted relative to myc-RPT6 transformed with vector. Comparison of all strains transformed with p416 control vector (A) or p416-Htt103Q (B) . Graph for all strains expressing Htt25Q is shown in S4 Fig.

In mammalian cell lines and primary neurons we found that the overexpression of

CaMKIIα induced a dramatic increase in the chymotryptic peptidase activity of the proteasome [22]. Furthermore, purified 26S proteasomes from Rpt6 S120A mutant mouse brains have both decreased chymotryptic activity in addition to decreased ATPase activity. CaMKIIα activity promotes the active sequestration of 26S proteasomes in dendritic spines [23, 45], however in this case, CaMKIIα kinase activity is dispensable which indicates that interactions of CaMKIIα with the 26S proteasomes can be sufficient to drive proteasomes into dendritic spines [45]. It is therefore plausible that in yeast the

S119A mutation alters a variety of 19S proteasome activities (e.g. assembly, gating of

20S CP, substrate translocation) as well as interactions with the many proteasome interacting proteins (PIPs) [57, 58, 75–81].

Although it is yet to be determined mechanistically how the S119A mutant alters

26S proteasome function, we find that it renders the cells more susceptible to various proteotoxic stresses. One might predict that the rpt6-S119D mutation would confer enhanced ability to counteract proteotoxic stress in comparison to the wild-type strain.

We found this to be the case when RPT6 null strains overexpressing rpt6-S119D were grown in 8% ethanol (S1 Fig). When rpt6-S119D was expressed from the endogenous locus, the susceptibility to proteotoxic stress was similar to wild type RPT6. Nonetheless, while the rpt6-S119A strain grew similar to wild type and rpt6-S119D in normal growth media, growth was significantly perturbed in stress-inducing conditions. As we demonstrated that the myc-Rpt6-S119A protein (the only cellular source of the Rpt6 protein in the null background) is clearly incorporated into functional proteasomes

allowing viability of the cells in normal growth media, we suggest that Rpt6 phosphorylation has a role in the cellular response to proteotoxic stress.

We also found protein aggregates of mutant Huntingtin protein with expanded

CAG repeats (mHtt103Q) to be significantly larger in rpt6-S119A mutants compared to wild-type or rpt6-S119D mutants. Intriguingly, Krobitsch and Lindquist did not find any effects on aggregate size in various partial loss-of-function mutations of the UPS, including mutations in genes encoding 20S (DOA3) and 26S (SEN3/RPN2) proteasome subunits [27]. In this work, we showed that Htt aggregates did not appear to be toxic for cells unless sensitized by proteotoxic stress, such as growth in canavanine. Furthermore,

Lin et al. (2013) showed that expressing Rpt6 S120A in mouse brain blocked the ability of a cAMP-elevating reagent to enhance proteasome activity, whereas the phosphomimetic Rpt6 mutant (Rpt6 S120D) increased proteasome activity, reduced HTT aggregates, and ameliorated motor impairment [82]. This further sup- ports the idea that

Rpt6 S120 phosphorylation is important for proteasome function under various stress conditions including protein aggregation.

We expected myc-rpt6-S120A mutants, which are significantly more sensitive to proteotoxic stress and have larger toxic Htt103Q aggregates (Figs 1 thru 3), to be more susceptible to chronological aging. However, viability was instead increased in myc-rpt6-

S120A mutant strain compared to myc-rpt6 and myc-rpt6-S119D strains (Fig 4B). In part, this is reminiscent of the findings in neurodegenerative disease models, where persistence of a non-functional cell state can exhaust brain resources, causing longer time in a senescent state, until apoptosis is activated [83–85]. This result will be of interest to pursue as the possibility of an adaptive response involving activation of chaperones or

autophagy remains. It is also possible that the senescence phenotypes observed here are distinct from the response to stress discussed above.

Together, our findings support the notion that Rpt6 phosphorylation at S119/S120 plays an important, conserved function in the ability of the proteasome to counteract proteotoxic stress and protein aggregate formation. Interestingly, the proteasome has been shown to be modified by other post-translational modifications [86–89, 89, 90]. Future studies will be important to help determine the mechanisms of how Rtp6 phosphorylation or other modifications alter 26S proteasome function in both normal and pathological states.

F. References

Matthews W, Driscoll J, Tanaka K, Ichihara A, Goldberg AL. Involvement of the proteasome in various degradative processes in mammalian cells. Proc Natl Acad Sci U S A. 1989; 86: 2597–2601. PMID: 2539595

Jentsch S. Ubiquitin-dependent protein degradation: a cellular perspective. Trends Cell Biol. 1992; 2: 98–103. PMID: 14732013

Ciechanover A. The ubiquitin-proteasome proteolytic pathway. Cell. 1994; 79: 13–21. PMID: 7923371

Muller S, Schwartz LM. Ubiquitin in homeostasis, development and disease. Bioessays. 1995; 17: 677–684. https://doi.org/10.1002/bies.950170804 PMID: 7661849

Stolzing A, Grune T. The proteasome and its function in the ageing process. Clin Exp Dermatol. 2001; 26: 566–572. PMID: 11696059

Goldberg AL, St John AC. Intracellular protein degradation in mammalian and bacterial cells: Part 2. Annu Rev Biochem. 1976; 45: 747–803. https://doi.org/10.1146/annurev.bi.45.070176.003531 PMID: 786161

Keller JN, Gee J, Ding Q. The proteasome in brain aging. Ageing Res Rev. 2002; 1: 279– 293. PMID: 12039443

Bossy-Wetzel E, Schwarzenbacher R, Lipton SA. Molecular pathways to neurodegeneration. Nat Med. 2004; 10 Suppl: S2–S9.

Nijholt DA, De KL, Elfrink HL, Hoozemans JJ, Scheper W. Removing protein aggregates: the role of proteolysis in neurodegeneration. Curr Med Chem. 2011; 18: 2459–2476. PMID: 21568912

Vilchez D, Saez I, Dillin A. The role of protein clearance mechanisms in organismal ageing and age- related diseases. Nat Commun. 2014; 5: 5659. https://doi.org/10.1038/ncomms6659 PMID: 25482515

Schwartz AL, Ciechanover A. Ubiquitin-mediated protein modification and degradation. Am J Respir Cell Mol Biol. 1992; 7: 463–468. https://doi.org/10.1165/ajrcmb/7.5.463 PMID: 1329865

Jana NR. Protein homeostasis and aging: role of ubiquitin protein ligases. Neurochem Int. 2012; 60: 443–447. https://doi.org/10.1016/j.neuint.2012.02.009 PMID: 22353631

Hershko A, Ciechanover A. The ubiquitin system. Annu Rev Biochem. 1998; 67: 425– 479. https://doi. org/10.1146/annurev.biochem.67.1.425 PMID: 9759494

Ciechanover A. Tracing the history of the ubiquitin proteolytic system: the pioneering article. Biochem Biophys Res Commun. 2009; 387: 1–10. https://doi.org/10.1016/j.bbrc.2009.06.065 PMID: 19539608

Pickart CM, Cohen RE. Proteasomes and their kin: proteases in the machine age. Nat Rev Mol Cell Biol. 2004; 5: 177–187. https://doi.org/10.1038/nrm1336 PMID: 14990998

Mayer RJ. From neurodegeneration to neurohomeostasis: the role of ubiquitin. Drug News Perspect. 2003; 16: 103–108. PMID: 12792671

Berke SJ, Paulson HL. Protein aggregation and the ubiquitin proteasome pathway: gaining the UPPer hand on neurodegeneration. Curr Opin Genet Dev. 2003; 13: 253–261. PMID: 12787787

Bence NF, Sampat RM, Kopito RR. Impairment of the ubiquitin-proteasome system by protein aggregation. Science. 2001; 292: 1552–1555. https://doi.org/10.1126/science.292.5521.1552 PMID: 11375494

Wang J, Wang CE, Orr A, Tydlacka S, Li SH, Li XJ. Impaired ubiquitin-proteasome system activity in the synapses of Huntington’s disease mice. J Cell Biol. 2008; 180: 1177–1189. https://doi.org/10.1083/ jcb.200709080 PMID: 18362179

Tanaka K, Matsuda N. Proteostasis and neurodegeneration: the roles of proteasomal degradation and autophagy. Biochim Biophys Acta. 2014; 1843: 197–204. https://doi.org/10.1016/j.bbamcr.2013.03. 012 PMID: 23523933

Dantuma NP, Bott LC. The ubiquitin-proteasome system in neurodegenerative diseases: precipitating factor, yet part of the solution. Front Mol Neurosci. 2014; 7: 70. https://doi.org/10.3389/fnmol.2014. 00070 PMID: 25132814

Djakovic SN, Schwarz LA, Barylko B, DeMartino GN, Patrick GN. Regulation of the proteasome by neuronal activity and calcium/calmodulin-dependent protein kinase II. J Biol Chem. 2009; 284: 26655– 26665. https://doi.org/10.1074/jbc.M109.021956 PMID: 19638347

Djakovic SN, Marquez-Lona EM, Jakawich SK, Wright R, Chu C, Sutton MA et al. Phosphorylation of Rpt6 regulates synaptic strength in hippocampal neurons. J Neurosci. 2012; 32: 5126–5131. https:// doi.org/10.1523/JNEUROSCI.4427- 11.2012 PMID: 22496558

Hamilton AM, Oh WC, Vega-Ramirez H, Stein IS, Hell JW, Patrick GN et al. Activity- dependent growth of new dendritic spines is regulated by the proteasome. Neuron. 2012; 74: 1023–1030. https://doi.org/ 10.1016/j.neuron.2012.04.031 PMID: 22726833

Braun RJ. Ubiquitin-dependent proteolysis in yeast cells expressing neurotoxic proteins. Front Mol Neu- rosci. 2015; 8: 8. https://doi.org/10.3389/fnmol.2015.00008 PMID: 25814926

Khurana V, Lindquist S. Modelling neurodegeneration in Saccharomyces cerevisiae: why cook with baker’s yeast? Nat Rev Neurosci. 2010; 11: 436–449. https://doi.org/10.1038/nrn2809 PMID: 20424620

Krobitsch S, Lindquist S. Aggregation of huntingtin in yeast varies with the length of the polyglutamine expansion and the expression of chaperone proteins. Proc Natl Acad Sci U S A. 2000; 97: 1589–1594. PMID: 10677504

Pereira C, Bessa C, Soares J, Leao M, Saraiva L. Contribution of yeast models to neurodegeneration research. J Biomed Biotechnol. 2012; 2012: 941232. https://doi.org/10.1155/2012/941232 PMID: 22910375

Hieter P, Boycott KM. Understanding rare disease pathogenesis: a grand challenge for model organisms. Genetics. 2014; 198: 443–445. https://doi.org/10.1534/genetics.114.170217 PMID: 25316782

Gauss R, Trautwein M, Sommer T, Spang A. New modules for the repeated internal and N-terminal epitope tagging of genes in Saccharomyces cerevisiae. Yeast. 2005; 22: 1–12. https://doi.org/10.1002/ yea.1187 PMID: 15565729

Sauer B. Recycling selectable markers in yeast. Biotechniques. 1994; 16: 1086–1088. PMID: 8074874

Winzeler EA, Shoemaker DD, Astromoff A, Liang H, Anderson K, Andre B et al. Functional characterization of the S. cerevisiae genome by gene deletion and parallel analysis. Science. 1999; 285: 901– 906. PMID: 10436161

Jones GM, Stalker J, Humphray S, West A, Cox T, Rogers J et al. A systematic library for comprehen- sive overexpression screens in Saccharomyces cerevisiae. Nat Methods. 2008; 5: 239–241. https:// doi.org/10.1038/nmeth.1181 PMID: 18246075

Elble R. A simple and efficient procedure for transformation of yeasts. Biotechniques. 1992; 13: 18–20. PMID: 1503765

Bond JF, Fridovich-Keil JL, Pillus L, Mulligan RC, Solomon F. A chicken-yeast chimeric beta-tubulin protein is incorporated into mouse microtubules in vivo. Cell. 1986; 44: 461–468. PMID: 3753663

Chang CS, Pillus L. Collaboration between the essential Esa1 acetyltransferase and the Rpd3 deacety- lase is mediated by H4K12 histone acetylation in Saccharomyces cerevisiae. Genetics. 2009; 183: 149–160. https://doi.org/10.1534/genetics.109.103846 PMID: 19596907

Nitiss J, Wang JC. DNA topoisomerase-targeting antitumor drugs can be studied in yeast. Proc Natl Acad Sci U S A. 1988; 85: 7501–7505. PMID: 2845409

Reifsnyder C, Lowell J, Clarke A, Pillus L. Yeast SAS silencing genes and human genes associated with AML and HIV-1 Tat interactions are homologous with acetyltransferases. Nat Genet. 1996; 14: 42–49. https://doi.org/10.1038/ng0996- 42 PMID: 8782818

Chang CS, Clarke A, Pillus L. Suppression analysis of esa1 mutants in Saccharomyces cerevisiae links NAB3 to transcriptional silencing and nucleolar functions. G3 (Bethesda). 2012; 2: 1223–1232.

Park S, Kim W, Tian G, Gygi SP, Finley D. Structural defects in the regulatory particle- core particle interface of the proteasome induce a novel proteasome stress response. J Biol Chem. 2011; 286: 36652– 36666. https://doi.org/10.1074/jbc.M111.285924 PMID: 21878652

Elsasser S, Schmidt M, Finley D. Characterization of the proteasome using native gel electrophoresis. Methods Enzymol. 2005; 398: 353–363. https://doi.org/10.1016/S0076-6879(05)98029-4 PMID: 16275342

Kisselev AF, Goldberg AL. Monitoring activity and inhibition of 26S proteasomes with fluorogenic peptide substrates. Methods Enzymol. 2005; 398: 364–378. https://doi.org/10.1016/S0076-6879(05) 98030-0 PMID: 16275343

Etlinger JD, Li SX, Guo GG, Li N. Phosphorylation and ubiquitination of the 26S proteasome complex. Enzyme Protein. 1993; 47: 325–329. PMID: 7697130

Rivett AJ, Bose S, Brooks P, Broadfoot KI. Regulation of proteasome complexes by gamma-interferon and phosphorylation. Biochimie. 2001; 83: 363–366. PMID: 11295498

Bingol B, Wang CF, Arnott D, Cheng D, Peng J, Sheng M. Autophosphorylated CaMKIIalpha acts as a scaffold to recruit proteasomes to dendritic spines. Cell. 2010; 140: 567–578. https://doi.org/10.1016/j. cell.2010.01.024 PMID: 20178748

Ghislain M, Udvardy A, Mann C. S. cerevisiae 26S protease mutants arrest cell division in G2/meta- phase. Nature. 1993; 366: 358–362. https://doi.org/10.1038/366358a0 PMID: 8247132

Xu Q, Singer RA, Johnston GC. Sug1 modulates yeast transcription activation by Cdc68. Mol Cell Biol. 1995; 15: 6025–6035. PMID: 7565755

Rubin DM, Glickman MH, Larsen CN, Dhruvakumar S, Finley D. Active site mutants in the six regulatory particle ATPases reveal multiple roles for ATP in the proteasome. EMBO J. 1998; 17: 4909–4919. https://doi.org/10.1093/emboj/17.17.4909 PMID: 9724628

Kim YC, DeMartino GN. C termini of proteasomal ATPases play nonequivalent roles in cellular assembly of mammalian 26 S proteasome. J Biol Chem. 2011; 286: 26652–26666. https://doi.org/10.1074/ jbc.M111.246793 PMID: 21628461

Glickman MH, Rubin DM, Fu H, Larsen CN, Coux O, Wefes I et al. Functional analysis of the proteasome regulatory particle. Mol Biol Rep. 1999; 26: 21–28. PMID: 10363642

Russell SJ, Reed SH, Huang W, Friedberg EC, Johnston SA. The 19S regulatory complex of the proteasome functions independently of proteolysis in nucleotide excision repair. Mol Cell. 1999; 3: 687–695. PMID: 10394357

Ezhkova E, Tansey WP. Proteasomal ATPases link ubiquitylation of histone H2B to methylation of histone H3. Mol Cell. 2004; 13: 435–442. PMID: 14967150

Lee D, Ezhkova E, Li B, Pattenden SG, Tansey WP, Workman JL. The proteasome regulatory particle alters the SAGA coactivator to enhance its interactions with transcriptional activators. Cell. 2005; 123: 423–436. https://doi.org/10.1016/j.cell.2005.08.015 PMID: 16269334

Chouduri AU, Tokumoto T, Dohra H, Ushimaru T, Yamada S. Functional and biochemical characterization of the 20S proteasome in a yeast temperature- sensitive mutant, rpt6-1. BMC Biochem. 2008; 9: 20. https://doi.org/10.1186/1471-2091-9-20 PMID: 18644121

Finley D, Ulrich HD, Sommer T, Kaiser P. The ubiquitin-proteasome system of Saccharomyces cerevisiae. Genetics. 2012; 192: 319–360. https://doi.org/10.1534/genetics.112.140467 PMID: 23028185

Glickman MH, Rubin DM, Fried VA, Finley D. The regulatory particle of the Saccharomyces cerevisiae proteasome. Mol Cell Biol. 1998; 18: 3149–3162. PMID: 9584156

Tomko RJ, Funakoshi M, Schneider K, Wang J, Hochstrasser M. Heterohexameric ring arrangement of the eukaryotic proteasomal ATPases: implications for proteasome structure and assembly. Mol Cell. 2010; 38: 393–403. https://doi.org/10.1016/j.molcel.2010.02.035 PMID: 20471945

Ehlinger A, Park S, Fahmy A, Lary JW, Cole JL, Finley D et al. Conformational dynamics of the Rpt6 ATPase in proteasome assembly and Rpn14 binding. Structure. 2013; 21: 753–765. https://doi.org/10. 1016/j.str.2013.02.021 PMID: 23562395

Park S, Roelofs J, Kim W, Robert J, Schmidt M, Gygi SP et al. Hexameric assembly of the proteasomal ATPases is templated through their C termini. Nature. 2009; 459: 866–870. https://doi.org/10.1038/ nature08065 PMID: 19412160

Heinemeyer W, Kleinschmidt JA, Saidowsky J, Escher C, Wolf DH. Proteinase yscE, the yeast proteasome/multicatalytic-multifunctional proteinase: mutants unravel its

100

function in stress induced proteolysis and uncover its necessity for cell survival. EMBO J. 1991; 10: 555–562. PMID: 2001673

Teixeira LK, Reed SI. Ubiquitin ligases and cell cycle control. Annu Rev Biochem. 2013; 82: 387–414. https://doi.org/10.1146/annurev-biochem-060410-105307 PMID: 23495935

Bassermann F, Eichner R, Pagano M. The ubiquitin proteasome system—implications for cell cycle control and the targeted treatment of cancer. Biochim Biophys Acta. 2014; 1843: 150–162. https://doi. org/10.1016/j.bbamcr.2013.02.028 PMID: 23466868

McGurk L, Berson A, Bonini N. Drosophila as an In Vivo Model for Human Neurodegenerative Disease. Genetics. 2015; 201: 377–402. https://doi.org/10.1534/genetics.115.179457 PMID: 26447127

Mason RP, Giorgini F. Modeling Huntington disease in yeast: perspectives and future directions. Prion. 2011; 5: 269–276. https://doi.org/10.4161/pri.18005 PMID: 22052350

Colby DW, Chu Y, Cassady JP, Duennwald M, Zazulak H, Webster JM et al. Potent inhibition of huntingtin aggregation and cytotoxicity by a disulfide bond-free single-domain intracellular antibody. Proc Natl Acad Sci U S A. 2004; 101: 17616–17621. https://doi.org/10.1073/pnas.0408134101 PMID: 15598740

Duennwald ML, Lindquist S. Impaired ERAD and ER stress are early and specific events in polyglutamine toxicity. Genes Dev. 2008; 22: 3308–3319. https://doi.org/10.1101/gad.1673408 PMID: 19015277

Petrucelli L, Dawson TM. Mechanism of neurodegenerative disease: role of the ubiquitin proteasome system. Ann Med. 2004; 36: 315–320. PMID: 15224658

Chen Q, Ding Q, Keller JN. The stationary phase model of aging in yeast for the study of oxidative stress and age-related neurodegeneration. Biogerontology. 2005; 6: 1– 13. https://doi.org/10.1007/s10522- 004-7379-6 PMID: 15834659

Kaeberlein M. Lessons on longevity from budding yeast. Nature. 2010; 464: 513–519. https://doi.org/ 10.1038/nature08981 PMID: 20336133

Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A, Lazarus DD et al. Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res. 1999; 59: 2615–2622. PMID: 10363983

101

Kumeda SI, Deguchi A, Toi M, Omura S, Umezawa K. Induction of G1 arrest and selective growth inhibition by lactacystin in human umbilical vein endothelial cells. Anticancer Res. 1999; 19: 3961–3968. PMID: 10628338

Machiels BM, Henfling ME, Gerards WL, Broers JL, Bloemendal H, Ramaekers FC et al. Detailed analysis of cell cycle kinetics upon proteasome inhibition. Cytometry. 1997; 28: 243–252. PMID: 9222110rnal.pone.0179893

Kisselev AF, Goldberg AL. Proteasome inhibitors: from research tools to drug candidates. Chem Biol. 2001; 8: 739–758. PMID: 11514224

Kachroo AH, Laurent JM, Yellman CM, Meyer AG, Wilke CO, Marcotte EM. Evolution. Systematic humanization of yeast genes reveals conserved functions and genetic modularity. Science. 2015; 348: 921–925. https://doi.org/10.1126/science.aaa0769 PMID: 25999509

Sokolova V, Li F, Polovin G, Park S. Proteasome Activation is Mediated via a Functional Switch of the Rpt6 C-terminal Tail Following Chaperone-dependent Assembly. Sci Rep. 2015; 5: 14909. https://doi. org/10.1038/srep14909 PMID: 26449534

Aufderheide A, Beck F, Stengel F, Hartwig M, Schweitzer A, Pfeifer G et al. Structural characterization of the interaction of Ubp6 with the 26S proteasome. Proc Natl Acad Sci U S A. 2015; 112: 8626–8631. https://doi.org/10.1073/pnas.1510449112 PMID: 26130806

Sledz P, Unverdorben P, Beck F, Pfeifer G, Schweitzer A, Forster F et al. Structure of the 26S proteasome with ATP-gammaS bound provides insights into the mechanism of nucleotide-dependent substrate translocation. Proc Natl Acad Sci U S A. 2013; 110: 7264–7269. https://doi.org/10.1073/pnas. 1305782110 PMID: 23589842

Kim YC, Li X, Thompson D, DeMartino GN. ATP binding by proteasomal ATPases regulates cellular assembly and substrate-induced functions of the 26 S proteasome. J Biol Chem. 2013; 288: 3334– 3345. https://doi.org/10.1074/jbc.M112.424788 PMID: 23212908

Park S, Li X, Kim HM, Singh CR, Tian G, Hoyt MA et al. Reconfiguration of the proteasome during chaperone-mediated assembly. Nature. 2013; 497: 512–516. https://doi.org/10.1038/nature12123 PMID: 23644457

Tian G, Park S, Lee MJ, Huck B, McAllister F, Hill CP et al. An asymmetric interface between the regulatory and core particles of the proteasome. Nat Struct Mol Biol. 2011; 18: 1259–1267. https://doi.org/10. 1038/nsmb.2147 PMID: 22037170

102

Xie Y, Varshavsky A. UFD4 lacking the proteasome-binding region catalyses ubiquitination but is impaired in proteolysis. Nat Cell Biol. 2002; 4: 1003–1007. https://doi.org/10.1038/ncb889 PMID: 12447385

Lin JT, Chang WC, Chen HM, Lai HL, Chen CY, Tao MH et al. Regulation of feedback between protein kinase A and the proteasome system worsens Huntington’s disease. Mol Cell Biol. 2013; 33: 1073– 1084. https://doi.org/10.1128/MCB.01434-12 PMID: 23275441

Moloney TC, Hyland R, O’Toole D, Paucard A, Kirik D, O’Doherty A et al. Heat shock protein 70 reduces alpha-synuclein-induced predegenerative neuronal dystrophy in the alpha-synuclein viral gene transfer rat model of Parkinson’s disease. CNS Neurosci Ther. 2014; 20: 50–58. https://doi.org/10.1111/cns. 12200 PMID: 24279716

Wirz KT, Keitel S, Swaab DF, Verhaagen J, Bossers K. Early molecular changes in Alzheimer disease: can we catch the disease in its presymptomatic phase? J Alzheimers Dis. 2014; 38: 719–740. https:// doi.org/10.3233/JAD-130920 PMID: 24072070

Rodriguez-Arellano JJ, Parpura V, Zorec R, Verkhratsky A. Astrocytes in physiological aging and Alz- heimer’s disease. Neuroscience. 2016; 323: 170–182. https://doi.org/10.1016/j.neuroscience.2015.01. 007 PMID: 25595973

Lin YY, Lu JY, Zhang J, Walter W, Dang W, Wan J et al. Protein acetylation microarray reveals that NuA4 controls key metabolic target regulating gluconeogenesis. Cell. 2009; 136: 1073–1084. https:// doi.org/10.1016/j.cell.2009.01.033 PMID: 19303850

Mitchell L, Huard S, Cotrut M, Pourhanifeh-Lemeri R, Steunou AL, Hamza A et al. mChIP-KAT-MS, a method to map protein interactions and acetylation sites for lysine acetyltransferases. Proc Natl Acad Sci U S A. 2013; 110: E1641–E1650. https://doi.org/10.1073/pnas.1218515110 PMID: 23572591

Kikuchi J, Iwafune Y, Akiyama T, Okayama A, Nakamura H, Arakawa N et al. Co- and post-translational modifications of the 26S proteasome in yeast. Proteomics. 2010; 10: 2769–2779. https://doi.org/10. 1002/pmic.200900283 PMID: 20486117

Besche HC, Sha Z, Kukushkin NV, Peth A, Hock EM, Kim W et al. Autoubiquitination of the 26S proteasome on Rpn13 regulates breakdown of ubiquitin conjugates. EMBO J. 2014; 33: 1159–1176. https:// doi.org/10.1002/embj.201386906 PMID: 24811749

103

Zhang F, Su K, Yang X, Bowe DB, Paterson AJ, Kudlow JE. O-GlcNAc modification is an endogenous inhibitor of the proteasome. Cell. 2003; 115: 715–725. PMID: 14675536

104

Acknowledgements

Chapter 4, in full, is a reprint of the material as it appears in Phosphorylation of the 19S regulatory particle ATPase subunit, Rpt6, modifies susceptibility to proteotoxic stress and protein aggregation, 2017. Marquez-Lona, EM, Torres-Machorro AL,

Gonzales FR, Pillus L, Patrick GN, PLOS ONE, 2017. The dissertation author was the

co-primary investigator and author of this material.