Modulation of Receptor Protein Tyrosine

MODULATION OF RECEPTOR PROTEIN TYROSINE PHOSPHATASE SIGMA ENHANCES PROTEASE ACTIVITY TO RELIEVE CHONDROITIN SULFATE PROTEOGLYCAN INHIBITION OF PERIPHERAL AXONS AND OLIGODENDROCYTES

AMANDA PHUONG TRAN

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Dissertation Advisor: Jerry Silver, Ph.D.

Department of Neurosciences

CASE WESTERN RESERVE UNIVERSITY

August, 2018

CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the dissertation of

AMANDA PHUONG TRAN

Candidate for the degree of Doctor of Philosophy*.

Committee Chair

Heather Brohier, Ph.D.

Committee Member

Evan Deneris, Ph.D.

Committee Member

Yu-Shang Lee, Ph.D.

Committee Member

Jerry Silver, Ph.D.

Date of Defense

May 11, 2018

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

contained therein. DEDICATION

I am no island entire of itself. To tell my story, I must begin with my dad, Sum V Tran, to whom this work is dedicated. His story began on a piece of another continent, in Vietnam where he began life as a son of refugee parents fleeing the Japanese invasion of South

Eastern China. He grew up in poverty and dropped out of grade school to support his family, but he was a proud man who believed in the power of his self-determination, fueled by acrid grit boiling in his belly. At a Chinese-owned detergent factory, he found a rising foothold by which he envisioned a more than comfortable life for his young Vietnamese wife, my mom, and two infant children. Then the Vietnamese Communist Party overthrew the Republic of Vietnam.

It was one thing overturning a republic and upending every facet of Vietnamese life. It was another to rename his hometown of Saigon to its conqueror, Ho Chi Minh. My dad has always lectured us on the value of adaptability. Of this trait being man’s most stubbornly denied, but most necessary means of survival. But my dad began to truly know fear when his boss was jailed for being a wealthy Chinese factory owner. And when my dad awoke one morning to have discovered that the Vietnamese Communist Party had replaced their own currency with a newly “liberated” unit now worth 1/500 of its previous state-mandated value, he took his remaining savings and my mother’s baubles, converted them to indelible gold, and planned his escape.

My dad fled the port of “Saigon” in the dead of night with Communist soldiers shooting at his dingy shared with some fifty other passengers. Off the coast of Malaysia, they purposefully sank their dingy to ensure rescue by the Malaysian coast guard. I have always thought it was a miracle they were not attacked by South East Asian pirates – a fate endured by his best friend who lost his wife and young son. My dad was interned in a

Malaysian refugee camp for two years never knowing when he would see his wife, his children, or what country would mercifully grace him with a fair chance to prove himself anew. He had heard that country was the United States of America, whose western coast had previously exploited the cheap labor of his South Eastern Chinese kinsmen to work on the continental railroad and then ceremoniously barred their family and friends to enter until The Chinese Exclusion Act of 1862 was supplanted with The Immigrant Act of 1965.

It was this act by Congress and America’s involvement in the Vietnam War that my dad owes his prized American citizenship.

Grateful for this second chance, my dad arrived in the San Francisco Bay Area with little resources and toiled in Oakland’s Chinatown before sponsoring my mom and older sisters for American citizenship. They arrived by means of air travel instead of dingy. He continued to toil until he bought a house in East Oakland, then the Oakland Hills, and then an even more prosperous suburb past the Caldecott Tunnel, which cleaved the 1% from the

99% in the East Bay. I was born in that one-bedroom house in East Oakland. We lived across a liquor store that sold stale moon pies and drippy Otter Pops for some 50 cents. My little sister was born in a cute ranch-style house next to a meticulously manicured park in the Oakland Hills. We both grew up in a wealthy, mainly white suburb past the Oakland Hills where I met my future husband. In many ways, our story mirrors many other Asian,

African, Middle Eastern, and Latinx American refugee experiences. Many Americans will nod their heads and reiterate the vitality of the “American Dream.” Others will try to twist our narrative to claim that Asian Americans are “model minorities” to their own political ends. Neither one of these narrow narratives fully encapsulates the rich experience of our

Chinese-Vietnamese American-ness. These myths flatten our stories. They are revisionist and serve, not us, but the people who retell them.

To understand our story is to acknowledge that grit corrodes overtime. I imagine that my dad was full of…grit and vinegar, driven to surmount the world in his youth. Upon the defense of this thesis, my dad will be 65 years old. At this point he has had wriggled out of the grasp of Chinese and Vietnamese Communism and outlasted decades of diabetes. He has survived a liver transplant after enduring liver cancer. He has tolerated radiation therapy from nasopharyngeal cancer. And a couple economic crashes here and there. I have observed that it is this silent, steely version of grit that has driven him through these crises.

It is this same grit that drives young men and women to denounce their ancestral homes to make their fortune in alien lands. Grit is what compels them to leverage their homes to garner access past the Oakland Hills. But grit is abrasive. Allowed to ferment, it can corrode our entrails leaving fleshy wounds we can only hope will close overtime. It boils and bubbles over to escape in bile-tasting belches on our tongues. When we swallow it down, sealing our lips to prevent its escape it can disintegrate our innards to leave bitter ashes behind. To understand our story is to acknowledge the sacrifices demanded by this spirited demon. To understand my own is to acknowledge the sacrifices my dad has committed driven by this fiery force.

My success – in love with my wonderfully supportive husband, Jason Hartman, in joy which I share with my best friend and little sister, Jeannie Tran, and ultimately in the completion of this work, is forever entwined with the sacrifices made by my dad. I often think about these sacrifices and regrets he has offered at the feet of Grit and Hope. And I think I will spend the rest of my life reconciling any of my future success to his despite his pitfalls.

So it is to him, Sum V Tran, that I wholeheartedly dedicate this work. TABLE OF CONTENTS

List of Figures ...... iv

Acknowledgements ...... vii

List of Abbreviations ...... x

Abstract ...... xv

I. General Introduction ...... 1 1.1 Discovery and Effects of Inhibitory Chondroitin Sulfate Proteoglycans in Spinal Cord Injury ...... 4 1.1.1 Discovery of Receptor Protein Tyrosine Phosphatase Sigma as the Cognate Receptor of Chondroitin Sulfate Proteoglycans and its Role in Axon Inhibition ...... 8 1.2 Multiple Sclerosis and the Inhibitory Effects of Chondroitin Sulfate Proteoglycans to Functional Recovery ...... 14 1.2.1 Oligodendrocyte Progenitor Cell Dysregulation in Multiple Sclerosis and Contributions of Chondroitin Sulfate Proteoglycan and Protein Tyrosine Phosphatase Sigma to Remyelination Failure ...... 18 1.3 Relief of Chondroitin Sulfate Proteoglycan-Mediated Inhibition through Degradation by Proteases ...... 22 1.3.1 Protease-Mediated Digestion of Chondroitin Sulfate Proteoglycans by Neurons ...... 28 1.3.2 Proteases and Oligodendrocyte Homeostasis ...... 33 1.4 Peptide Modulation of Receptor Protein Tyrosine Phosphatase Sigma and Subsequent Protease Secretion to Relieve Chondroitin Sulfate Proteoglycan-Mediated Inhibition ...... 39 i

II. Modulation of Protein Tyrosine Phosphatase Receptor Sigma Enhances Cathepsin B Release in Dorsal Root Ganglion Neurons to Relieve CSPG Inhibition and enhance Axon Outgrowth ...... 42 2.1 Abstract ...... 43 2.2 Introduction ...... 45 2.3 Methods ...... 48 2.4 Results ...... 59 2.5 Discussion ...... 69 2.6 Figures ...... 74

III. Modulation of Protein Tyrosine Phosphatase Receptor Sigma Enhances MMP-2 Release in Oligodendrocytes to Relieve CSPG Inhibition and Restore Oligodendrocyte Homeostasis and Remyelination ...... 89 3.1 Abstract ...... 90 3.2 Introduction ...... 91 3.3 Methods ...... 94 3.4 Results ...... 106 3.5 Discussion ...... 120 3.6 Figures ...... 124

IV. General Discussion ...... 160 4.1.1 Summary and Discussion of Modulation of Protein Tyrosine Phosphatase Receptor Sigma Leading to Enhanced Cathepsin B Activity in Peripheral Axons ...... 161 4.1.2 Summary and Discussion of Modulation of Protein Tyrosine Phosphatase Receptor Sigma Leading to Enhanced MMP-2 Activity in Oligodendrocyte Progenitor Cells ...... 171 4.2 Study Implications ...... 177

4.2.1 Protein Tyrosine Phosphatase Receptor Sigma as a Switch in Plasticity ...... 180 4.3 Study Limitations ...... 185 4.3.1 Limitations of Modulation of Protein Tyrosine Phosphatase Receptor Sigma in Peripheral Axons ...... 190 4.3.2 Limitations of Modulation of Protein Tyrosine Phosphatase Receptor Sigma in Oligodendrocytes ...... 193 4.4 Future Directions ...... 197 4.4.1 The Role of Protease Activity in Protein Tyrosine Phosphatase Sigma as a Switch for Plasticity ...... 197 4.4.2 Identifying the Vesicular Body Linked to Protein Tyrosine Phosphatase Sigma Regulated Protease Release ...... 198 4.4.3 Protein Tyrosine Phosphatase Sigma, Chondroitin Sulfate Proteoglycans, and the Regulation of Autophagy ...... 200 4.5 Concluding Remarks ...... 206

Bibliography ...... 209

iii

LIST OF FIGURES

Chapter - Figure Number Page II - Figure 1 ISP treatment enhances glycosaminoglycan-chondroitin sulfate proteoglycans (GAG CSPG) degradation by neurons...... 74 II - Figure 2 ISP promotes GAG chain degradation in coverslip-bound aggrecan/laminin spot assays...... 76 II - Figure 3 ISP promotes GAG chain degradation through increasing protease activity...... 78 II - Figure 4 ISP promotes secretion of Cathepsin B (CatB)...... 80 II - Figure 5 Pretreatment of aggrecan spots with ISP-treated conditioned media (CM) or recombinant Cathepsin B (rCatB) enhances axon crossings through the aggrecan gradient...... 82 II - Figure 6 Overexpression of Cystatin B (CSTBo/e) decreases ISP-treated DRG axon crossings through CSPGs...... 84 II - Figure 7 Serotonergic axons express Cathepsin B (CatB)...... 86 II - Figure 8 Genetic loss of PTPσ correlates with Cathepsin B (CatB) activity and CSPG degradation...... 87 III - Figure 1 ISP promotes functional and histological recovery in EAE mouse model ...... 124 III - Figure 2

ISP promotes remyelination in the spinal cord of lysolecithin (LPC)- demyelinated mice...... 126 III - Figure 3 ISP accelerates remyelination in LPC treated organotypic cerebellar cultures...... 128 III - Figure 4 ISP decreases chondroitin sulfate proteoglycan (CSPG) load in both EAE and LPC models...... 130 III - Figure 5 ISP increases CSPG-degrading protease activity...... 132 III - Figure 6 ISP increases MMP-2 secretion and activity...... 134 III - Figure 7 ISP-induced MMP-2 activity increases OPC migration and remyelination through CSPG disinhibition...... 136 III - Figure 8 ISP promotes myelin repair through increasing MMP-2 expression in LPC-induced demyelination model of mice...... 138 III – Supplementary Figure 1 ISP accelerates remyelination in LPC treated organotypic cerebellar cultures...... 140 III - Supplementary Figure 2 PTPσ expression is enhanced following EAE and LPC...... 142 III - Supplementary Figure 3 ISP increases CSPG clearing as remyelination occurs...... 144 III - Supplementary Figure 4 ISP modulates inflammation in EAE models of mice...... 146 III - Supplementary Figure 5 ISP promotes OPC recruitment and survival on CSPGs...... 148 III - Supplementary Figure 6 ISP enhances OPC process outgrowth and maturation...... 150 III – Supplementary Figure 7

ISP-treated conditioned media protease array...... 152 III - Supplementary Figure 8 ISP enhances CS56 degradation in a dose-dependent manner...... 153 III - Supplementary Figure 9 Protease inhibitors attenuate ISP-induced CSPG degradation and subsequent CSPG-OPC disinhibition...... 155 III - Supplementary Figure 10 shRNA knock down of MMP-2 decreases OPC maturation and migration on CSPGs to limit remyelination in cerebellar slices...... 157 III - Supplementary Figure 11 MMP-2 mediates ISP-induced remyelination in LPC-demyelinated mouse model...... 159

ACKNOWLEDGMENTS

My sincere appreciation and greatest gratitude must be due to my mentor, Dr. Jerry Silver. I have worked in multiple labs across several university campuses and have never had any mentor wholeheartedly express his support as generously, nor as exuberantly, as Jerry.

Jerry was the first scientist – no, adult – who has asserted that I am smart. That I am capable of making a contribution. That I am - capable. I have long realized how lucky I am to be able to train in his lab and to be even a tiny part of his enthusiastic advancement of spinal cord injury research. Ramon y Cajal once lamented how “idle [it is] to dispute with old men. Their opinions, like their cranial sutures, are ossified.” This, however, has not been my experience in the Silver lab. I have seen in a plethora of examples where Jerry has cheerily championed new techniques, new ideas, and new researchers in their advancement towards understanding spinal cord injury. Jerry defends the new. New ideas and new scientists like me. I am forever grateful for his encouragement, mentorship, optimism, and warm support.

Many thanks and exuberantly jejune Amanda-style clapping must also be extended to members of the Silver lab. I am so grateful to be part of this “lean and mean” club. The work delineated in this thesis owes its foundation to all of Brad Lang’s previous work which itself is dependent on more than thirty years of collaborative chondroitin sulfate proteoglycan research begun in the Silver lab. It’s been really fun standing on the shoulders of these crafty pioneers. Thank you also, Brad and Sarah, for your cheerful support and encouragement. I am exceedingly lucky to have made a lifelong friend in Philippa Warren,

vii a thoughtful, inventive, and meticulous scientist to whom I owe a great deal of emotional support and of whose work I will always be the biggest fan. I am so happy that we were able to team up to write a monstrous review – it is one of my proudest accomplishments and would have never been possible without all your hard work. Many thanks for Marc

DePaul who always had a kind smile and encouraging words despite his plethora of animal work. Thanks to Jared Cregg for his invaluable insight and critical eye. I am happy to have been able to collaborate with Angela Nord whose cheery demeanor always brightened my day. My thanks to Teresa Evans for all her kind words and support. To Adrianna Milton, welcome to the club! Special thanks to Bruce Yu for his indispensible tissue work. And to the amazing undergraduates I have had the immense luck to mentor during my tenure as graduate student: Sapna Sundar, Meigen Yu, and Chandrika Sanapala. Thank you for all your hard work towards my thesis. I am so proud of your accomplishments and feel extremely lucky to have been allowed to be part of your life at Case in my own tiny way.

I must also acknowledge the help from Dr. Yan Yang and Dr. Fucheng Luo. I am grateful for their expertise and help with the oligodendrocyte work.

Special thanks to Dr. Barbara Bedogni whose familiarity with proteases and kind words were invaluable in the early stages of this thesis. Thanks also to the members of the committee, Drs. Heather Broihier, Evan Deneris, and Yu-Shang Lee for their valuable input.

And because I am longwinded and enjoy repetition, I would like to extend another thanks to my family. To my dad, Sum Tran, I owe my traits of perseverance and quirky

viii adaptability. To my best friend and little sister, Jeannie, I thank for all the unconditional and uplifting love. To my dad- and mom in-law, Tom and Sherry Hartman, my vigorous thanks and love for their understanding and unwavering support.

My most animated appreciation, thanks, and fondness is reserved for my husband, Jason

Hartman, without whom this thesis would have never been possible. No, really - this is not some facile, clichéd sentiment. Without Jason I would not have been able to apply to graduate school. I would not have been able to move 2500 miles away from home. I would not have been able to become a scientist. I would have had to choose between dream and filial piety instead. Thank you, Jason. Thank you for averting me from the choice of either self or family. To you, I owe my proverbial cake and being able to eat it, too. Without you I would have been stuck at home - stagnant and much, much less.

LIST OF ABBREVIATIONS

5-HT – 5-Hydroxy Tryptophan/ serotonin

AAV – Adeno-Associated Virus

ADAMTS – A Disintegrin and Metalloprotease with Thrombospondin motifs

Akt – Protein Kinase B

AMPA – α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid Receptor

Arg 1 – Arginase 1

ATP – Adenosine Triphosphate

BACE1 – beta-secretase 1

BBB/BSCB – Blood Brain Barrier/Blood-Spinal Cord Barrier

BLBP – Brain Lipid-Binding Protein

CCL2 – C-C Motif Ligand 2

Cdc42 – Cell division control protein 42 homolog

CGRP – Calcitonin Gene-Related Peptide

CNS – Central Nervous System

CNTF – Ciliary Neurotrophic Factor

COS7 – Ceropithecus aethiops kidney cell line 7

CS-56 – Chondroitin Sulfate 56

CS-D/E – Chondroitin Sulfate-D/E

CS-GAG – Chondroitin Sulfate Glycosaminoglycan

CSPG – Chondroitin Sulfate Proteoglycan

DRG – Dorsal Root Ganglion e/GFP – enhanced Green Fluorescent Protein

EAE – Experimental Autoimmine Encephalomyelitis

ECM – Extracellular Matrix

EGF – Epidermal Growth Factor

ERK – Extracellular Receptor Kinase

FasL- Fas Ligand

FGF2 – Fibroblast Growth Factor 2

GABA – Gamma-Aminobutryic Acid

GFAP – Glia Fibrillary Acidic Protein

HSPG – Heparan Sulfate Proteoglyan

Iba1 – Ionized Calcium-Binding Adapter Molecule 1

IFNγ – Interferon gamma

IL - /1β / 10 – Interleukin 1beta/10

ILP – Intracellular LAR Peptide

ISP – Intracellular Sigma Peptide

LAR – Leukocyte Common Antigen-Related Receptor

LC3B –Microtubule-Associated Protein Light Chain 3

LPC - Lysophosphatidylcholine

LPS – lipopolysaccharide

MAPK – Mitogen Activated Protein Kinase

MBP – Myelin Basic Protein

MMP – Metalloprotease

MOG – Myelin Oligodendrocyte Glycoprotein mRNA – messenger Ribonucleic Acid mTOR – mammalian Target of Rapamycin

NF-L – Neurofilament Light Chain

NG2 – Proteoglycan Neuron Glia Antigen 2

NGF – Nerve Growth Factor

NGL-3 – Netrin-G Ligand-3

NMDA – N-methyl-D-aspartate Receptor

NT-3 – Neurotrophin-3

xii

O4 – Oligodendrocyte 4

Olig2 – Oligodendrocyte transcription factor 2

P13K – Phosphoinositide 3-Kinase

PKC – Protein Kinase C

PLCγ – Phospholipase C gamma

PLP – Myelin Proteolipid Protein

PSD-95 – Postsynaptic Density 95

PTP δ/ µ / σ – Protein Tyrosine Phosphatase Receptor Delta/ Mu/ Sigma

RACK – Receptor for Activated Kinase C

Rho GTPase– Rho Guanosine Triphosphate

RhoA – Ras homolog member A

ROCK – Rho-associated Protein Kinase

SMAD – portmanteau of c. elegans gene Sma and drosophila gene MAD denotes protein transducers of transforming growth factor beta superfamily

TAT – Transactivator of Transcription

TGFβ – Transforming Growth Factor beta

THP-1 – human acute monocytic leukemia cell line

TIMP-1 – Tissue Inhibitor of Metalloprotease 1

xiii

TNFα – Tumor Necrosing Factor alpha tPA – tissue Plasminogen Activator

TrkC – Tropomyosin Receptor Kinase C

Tuj1 – Beta Tubulin Class III uPA – urokinase-type Plasminogen Activator

xiv

Modulation of Receptor Protein Tyrosine Phosphatase Sigma Enhances Protease Activity to Relieve Chondroitin Sulfate Proteoglycan Inhibition of Peripheral Axons and Oligodendrocytes

Abstract

Amanda Phuong Tran

Chondroitin sulfate proteoglycans (CSPGs) have been well known to inhibit axon regeneration following many models of CNS injury where they inhibit axons resulting in regeneration failure. Also notable is the increasingly important role upregulated CSPGs play in neurodegenerative diseases including multiple sclerosis where CSPG-laden plaques inhibit oligodendrocyte progenitor cell function and ultimately remyelination. Identified as the cognate receptor of CSPGs, manipulation of receptor protein tyrosine phosphatase receptor sigma (PTPσ) has been shown to successfully ameliorate the inhibitory effects of

CSPGs in both axonal outgrowth and oligodendrocyte progenitor cell maturation and function. Recently, the Silver lab has synthesized a peptide, called Intracellular Sigma

Peptide (ISP), specific to the regulatory domain of PTPσ and have found that systemic treatment of ISP to spinal cord contused rats restored coordinated locomotor, sensorimotor, and urinary function. Restoration of function following spinal cord injury was, in part, due to enhanced serotonergic sprouting; however, the exact cellular mechanisms underlying

PTPσ modulation and enhanced sprouting has been heretofore unknown. Work outlined

xv here proposes a novel mechanism by which manipulation of PTPσ relieves CSPG inhibition in both peripheral axons and oligodendrocyte progenitor cells through enhanced protease activity and subsequent CSPG degradation. Through various biochemical, in vitro cultures of dorsal root ganglion axons and oligodendrocyte progenitor cells, this work characterizes enhanced protease secretion and activity by both peripheral axons and oligodendrocyte progenitor cells through ISP modulation of PTPσ. In peripheral axons,

Cathepsin B secretion and activity is enhanced by ISP treatment, which enhances axonal outgrowth past a high CSPG gradient. In oligodendrocyte progenitor cells, ISP treatment enhances MMP-2 activity, which is important for increasing oligodendrocyte progenitor cell mobility past a high CSPG barrier. ISP treatment additionally relieves CSPG inhibition in a mouse model of multiple sclerosis to improve remyelination and ultimately locomotor function. Thus, separate studies characterizing ISP treatment of peripheral axons and oligodendrocyte progenitor cells underline the potent inhibitory effects of CSPGs in different models of CNS injuries. This work links protease activity to PTPσ modulation for the first time and highlights how CSPG disinhibition could occur in peripheral axons and oligodendrocyte progenitor cells through enhanced protease activity.

xvi

I. GENERAL INTRODUCTION

An underappreciated aspect of neuroscience is the brain and spinal cord extracellular space, specifically the extracellular matrix (ECM) that enwraps glia and neurons and supports their homeostasis, development, migration, differentiation, and regulates their plasticity.

Chondroitin sulfate proteoglycans (CSPG) are macromolecules found in the ECM composed of a core protein with varying chains of chondroitin sulfate glycosaminoglycan

(CS-GAG) attached through a tetrasaccharide linkage added during post-translational processing (Kjellen and Lindahl 1991). These CS-GAGs are themselves composed of up to multiple repeating disaccharide units of glucuronic acid and N-acetylgalactosamine bound together through their O-linkages to serine residues (Kwok et al. 2011).

These repeating sugar moieties are especially important because they are what distinguishes the 16 different types of CSPGs present in the CNS. Specifically, each disaccharide unit can be differentially sulfated which ultimately confers specific binding patterns of CS-GAGs with other molecules or proteins (Gama et al. 2006). For example, a

CS-E pattern of sulfation, consisting of carbon-4 and carbon-6 sulfation of the repeating disaccharide chains, is upregulated following injury to the CNS and leads to increased binding to specific receptors of the leukocyte common antigen-related receptor (LAR) family including protein tyrosine phosphatase receptor sigma (PTPσ) to ultimately inhibit axonal outgrowth downstream (Miyata et al. 2012). Thus, it is the quantity of the CS-

GAGs and subsequent sulfation of these disaccharide chains that imparts the axon inhibitory properties of CSPGs. Adding to the diversity of CSPGs, the core protein of the 1

CSPG can also bestow a range of functionality by determining how the CSPG may be incorporated in to the ECM.

Members of the lectican group of CSPGs comprising brevican, neurocan, versican, and aggrecan are the most abundant CSPGs found in the ECM of the CNS (Viapiano and

Matthews 2006). Of particular note, aggrecan, which has the most CS-GAGs of all the lecticans and thus provides the greatest inhibition to axons, is confined to ECM structures called perineuronal nets that enwrap the somata of select neurons (Morawski et al. 2012).

For this reason, and the fact that it is easily and consistently extracted from bovine cartilage, aggrecan is most often utilized CSPG in in vitro axon regeneration studies.

The axon-inhibitory, and ultimately, plasticity-limiting properties of CSPGs have long been appreciated in mammalian CNS development. CSPGs are deposited in strategic crossroads or placed as barriers during embryogenesis to aid in axon guidance (M.T. Wilson and

Snow 2000; J.M. Brooks et al. 2013). The emergence of perineuronal nets laden with

CSPGs also closes the postnatal critical period, a time when neural plasticity is heightened due to emerging projections that are becoming pruned, refined, and optimized (Takesian and Hensch 2013). In contrast, during adulthood the presence of perineuronal nets preserves circuit stability by preventing large-scale plasticity, with the exception of plasticity responsible for learning or memory formation, as CSPGs finely regulate the formation, maintenance, and activity of synapses (Gogolla et al. 2009; S. Yang et al.

2015). Depleting the CS-GAGs in perineuronal nets can therefore reverse its plasticity- limiting properties. A well-known example of this is the injection of the bacterially derived

CS-GAG cleaving enzyme, chondroitinase ABC, into the visual cortex of adult rats to restore ocular dominance plasticity (Pizzorusso 2009).

In the context of CNS diseases and injury, chondroitinase ABC depletion of CS-GAGs has been well known to increase axon outgrowth following spinal cord injury (Bradbury et al.

2002) to enhance plasticity in adulthood even after injury when CSPGs are upregulated following inflammatory processes in not only the perineuronal nets around the lesion

(Alilain et al. 2011), but also in structures formed following injury called the glial scar

(Cregg et al. 2014). However, recent studies are beginning to highlight a burgeoning awareness of the role CSPGs play not only in limiting plasticity during adulthood, but also in neurological disorders including epilepsy (Kurazono et al. 2001; Okamoto et al. 2003), stroke (Carmichael et al. 2005; Deguchi et al. 2005), Alzheimer’s Disease (DeWitt et al.

1993; Végh et al. 2014; Fawcett 2015), and Multiple Sclerosis (Sobel and Ahmed 2001;

H. Mohan et al. 2010; Lau et al. 2012). In these disease models, CSPG increase is positively correlated with disease progression and plays a role in inhibiting restoration of function.

This thesis delineates work in enhancing axon outgrowth and oligodendrocyte progenitor cell health and differentiation following models of enhanced CSPG deposition in spinal cord injury and Multiple Sclerosis respectively through modulation of the CSPG receptor, protein tyrosine phosphatase sigma (PTPσ). Using a peptide designed to target the intracellular regulatory domain of PTPσ, we have found that axon outgrowth and

3 oligodendrocyte progenitor cell differentiation can be enhanced through degradation of inhibitory CSPGs through activation of proteases.

1.1 Discovery and Effects of Inhibitory Chondroitin Sulfate Proteoglycans in

Spinal Cord Injury

The inhibitory effects of CSPGs in the adult CNS were first observed through its robust impediment of neurite growth in vitro (Carbonetto et al. 1983; Snow et al. 1990; McKeon et al. 1995). While it has long been observed that adult neurons possess a reduced intrinsic capacity to grow, in vitro studies such as those performed in Carbonetto or Snow et al. altered the composition of the in vitro substrate to specifically test the effects of CSPGs and revealed the importance of the extrinsic effects of the ECM on neuronal outgrowth. For example, adult sensory dorsal root ganglion (DRG) neurons plated on a stripe assay of alternating stripes of permissive laminin or inhibitory aggrecan CSPG illustrated how axons preferred to travel along laminin while turning away to avoid CSPGs (Snow et al.

1990). Later work pioneered in the Silver Lab involving the transplantation of adult DRGs into a small crush lesion of the spinal cord discovered that these regenerating axons abruptly halted once they reached the glial scar (Davies et al. 1999). Immunostaining of the injured spinal cord revealed that axons formed bulbous dystrophic ends in areas of increasing CS-GAG deposition radiating from the penumbra of the lesion core. This seminal study from Davies et al. therefore broached the importance of CSPG deposited in the glial scar as a biochemical impediment to axon regeneration.

The glial scar formed following injury to the spinal cord is a complex structure orchestrated by multiple cell types that acutely seals off inflammation, and thus, further injury to the spinal cord. Inhibiting the inflammatory processes responsible for gliosis and immune cell activation through systemic glucocorticoid treatments have proven to be detrimental in human cases of spinal cord injury (Fehlings et al. 2014; Evaniew et al. 2016) and even resulted in morbidity in some rodent studies (Pereira et al. 2009; Marcon et al. 2010).

However, the glial scar ultimately acts as a chronic barrier to axon regeneration and the restoration of function. Many of the studies characterizing spinal cord injury and the subsequent development of the glial scar have been performed in rats or mice with crush lesions or contusive injuries of the spinal cord resulting from force impelled from a controlled piston onto an exposed area of the spinal cord.

Physical injury to the cord itself initiates glial reactivity (Sofroniew and Vinters 2009;

Bardehle et al. 2013), but the surge of inflammation is initiated through the release of chemoattractive alarmins from necrosing, and to a lesser extent, apoptosing cells at the lesion (Kigerl et al. 2009; Gadani, Walsh, Smirnov, et al. 2015). Also prevalent in this stage of injury is the immediate torsion and damage to axons and subsequent Wallerian degeneration (Crowe et al. 1997). Ultimately, the release of alarmins initiates an inflammatory cascade beginning with the activation of resident microglia and infiltration of systemic immune cells (Gadani, Walsh, Lukens, et al. 2015) through the compromised blood-spinal cord barrier (BSCB), which themselves contribute to inflammation through the secretion of pro-inflammatory factors once they become activated by the pro- inflammatory environment (Kawabata et al. 2010; A.R. Taylor et al. 2014). Consequently,

5 reactivity of many different cell types including ependymal cells (Barnabé-Heider et al.

2010), pericytes, fibroblasts (Göritz et al. 2011), NG2+ oligodendrocytes progenitor cells

(Filous et al. 2014; Rodriguez et al. 2014), and notably astrocytes (Davies et al. 1997;

Cregg et al. 2014) occur. As a result, the glial scar forms beginning with macrophages and fibroblasts sealed in the lesion core by a palisading formation of reactive astrocytes. Thus, the re-establishment of a glia limitans made through astrocyte-fibroblast cell contact through Ephrin-B2 signaling (Bundesen et al. 2003) forms to separate the inflammatory fibrotic lesion core from the lesion penumbra with NG2+ oligodendrocytes progenitor cells and microglia intermittently dispersed throughout the penumbra.

Along the lesion penumbra, reactive astrocytes and oligodendrocyte progenitor cells markedly deposit CSPGs (McKeon et al. 1991; McKeon et al. 1995; Asher et al. 2000;

Asher et al. 2002). In part, CSPG deposition is upregulated through the infusion of systemically circulating blood and related factors that leak through the damaged BSCB.

One consequence of this is the activation of proinflammatory factor tumor growth factor

(TGF)-β, which can promote the local upregulation of CSPGs through SMAD (Schachtrup et al. 2007) or P13K, mTOR, and EGF signaling (Garcia et al. 2016). In addition to TGFβ, other inflammatory factors can also increase the mRNA of CSPG-producing enzymes in glial activated by injury (Properzi et al. 2005; Gris et al. 2007). Reactive astrocytes, for example, predominantly deposit CSPGs such as neurocan, versican, NG2, and Tenascin-C

(McKeon et al. 1995; Tang et al. 2003; Andrews et al. 2012). Oligodendrocyte progenitor cells additionally cleave and deposit NG2 into the ECM through cell-surface metalloproteases (MMP) (Asher et al. 2005).

While the glial scar and its accompanying CSPGs persist chronically, tissue remodeling near and within the lesion site is a dynamic process. Some astrocyte-derived CSPGs such as aggrecan and brevican are deposited initially, but have been shown to decrease overtime as the glial scar matures; however, other CSPGs such as versican have been known to persist chronically (Andrews et al. 2012). The surrounding perineuronal nets outside of the lesion site additionally see an upregulation of CSPGs (Massey et al. 2006; Alilain et al.

2011). These structures found predominantly in the ventral motor pools along the spinal cord provide yet another impediment to the re-establishment of sprouting neural circuitry that may be helpful in functional recovery. Thus, CSPGs form a chemorepellant gradient radiating from the glial scar that serves to inhibit any approaching axons (Davies et al.

1999).

In response to an increasing gradient of CSPGs, approaching axons form swollen and distorted end balls (Y. Li and Raisman 1995; Tom et al. 2004). This was first observed by

Ramon y Cajal as “sterile clubs” in the injured spinal cords of adult cats (Cajal 1959).

Since Cajal, studies of axonal growth cones have refuted the idea of these dystrophic end balls as sterile bulbs. Instead, these dystrophic growth cones are dynamic structures entrapped by their surrounding ECM. A useful tool to study this phenomenon has been the development of the spot assay coupled with real-time or time-lapse microscopy. First developed by Tom et al., the spot assay takes advantage of the differing drying properties

(Yunker et al. 2011) of aggrecan and laminin to create a reciprocal gradient in a coffee-ring like effect (Tom et al. 2004). Thus, plated neurons would be attracted to the high

7 concentration of permissive laminin in the center of the spot assay, but as the axons grow outward, its growth cone is confronted with an increasing gradient of inhibitory aggrecan as seen in the pattern of CSPG deposition of the glial scar. Combined with time-lapse imaging, Tom et al. observed that these growth cones actively extended and retracted their finger-like filopodia and lamellipodium in “sheet-like veils” without any net movement forward. Adding to this dynamism is the constant blebbing of cycling vesicles in a subset of particularly swollen dystrophic growth cones. Indeed, swollen, dystrophic growth cones have been observed in axons following spinal cord injury in rats (Davies et al. 1999; Tom et al. 2004; Filous et al. 2014) and even after 40 years following a human case of spinal cord injury (Ruschel et al. 2015). Inhibition of axonal regeneration by CSPGs therefore occurs at the level of the growth cone mediated by the complex machinery of signaling partners found at the membrane.

1.1.1 Discovery of Receptor Protein Tyrosine Phosphatase Sigma as the Cognate

Receptor of Chondroitin Sulfate Proteoglycans and its Role in Axon

Inhibition

PTPσ was first cloned by the Schlessinger group in 1993 from rat brains where it was found to be notably enriched in the CNS (Yan et al. 1993). The Tremblay group then generated a constitutive knockout mouse after their own cloning experiments where they named the same receptor PTP NU3 (Wagner et al. 1994) and found major endocrine- related abnormalities that they hypothesized may be due to dysregulation of the adhesiveness of various cell types (Elchebly et al. 1999). However, the discovery that

PTPσ is the cognate receptor for the robust inhibition of axonal outgrowth by CSPGs, and thus an important regulator of growth cone and subsequent axonal plasticity, was not made until 2009 by Shen et al. Furthermore, PTPσ has been found to bind to CS-D and CS-E sulfation patterns (Brown et al. 2012) which are highly upregulated following spinal cord injury (Dickendesher et al. 2012; Yi et al. 2014), suggesting that PTPσ would be an appealing candidate in which to target axonal regeneration following CNS injury.

PTPσ belongs to the type IIA subfamily of receptor protein tyrosine phosphate called LAR, which includes two other members: LAR itself and PTPδ. Based on in situ hybridization studies by Schaapveld et al., LAR appears to be highly expressed in the heart, lung, gut, nasal cavity, and DRGs of the developing mouse embryo while only lowly expressed in the

CNS. PTPδ mRNA expression seems to be moderately expressed in the CNS with enrichment in the prefrontal cortex (Schaapveld et al. 1998). In contrast, PTPσ is highly expressed throughout the CNS, nasal cavity, and DRGs and additionally plays an important role in axon guidance during development (Schaapveld et al. 1998; Shen 2014).

Like other receptor-like PTP subtypes, members of the LAR subfamily are transmembrane receptors with two intracellular domains including the phosphatase-active and inactive D1 and D2 domains respectively (Andersen et al. 2001; Chagnon et al. 2004). The extracellular domain consists of varying four to eight fibronectin type III repeats closest to the transmembrane domain, followed by three immunoglobulin domains (Chagnon et al.

2004; Y. Xu and G.J. Fisher 2012). PTPσ have four to eight repeating fibronectin type III repeats depending on the two alternatively spliced isoforms (Pulido et al. 1995).

Importantly, it is the immunogluboulin domain in the N-terminal of the receptor that confers binding to the ligands of PTPσ: axon inhibitory CSPGs and axon permissive heparan sulfate proteoglycans (HSPG) (Coles et al. 2011). Thus, PTPσ is a bifunctional receptor bestowing both axonal inhibition and permissive axonal outgrowth depending on the immediate ECM context.

As PTPσ is highly expressed throughout the CNS (Yan et al. 1993; Schaapveld et al.

1998), and importantly, in the growth cones of axons, they are uniquely positioned to mediate axon growth dynamics (Lang et al. 2015). The axon-growth permissive properties of HSPGs have been well known in vitro (Dow et al. 1994; Kinnunen et al. 1998). Like

CSPGs, HSPGs also contain GAG chains bound to a protein core; however, they differ in the sulfation pattern of these sugar moieties. But how is PTPσ able to confer both permissive and inhibitory axon activity?

PTPσ was first observed to bind in high affinity to HSPGs in chick retinal axons (Aricescu et al. 2002) where it plays a major role in axon guidance and outgrowth (Aricescu et al.

2002; Rashid-Doubell et al. 2002). The application of heparinase, which enzymatically cleaves the GAG chains of HSPGs, abolished the ability of PTPσ to bind with HSPGs indicating that the GAG domain is what confers receptor binding (Aricescu et al. 2002).

Shen et al. then discovered that PTPσ additionally binds to the CS-GAG domain of CSPGs with an approximate affinity as HSPGs (Shen et al. 2009). In fact, it is PTPσ that effects the axon-inhibitory qualities of CSPGs. This was illustrated by a set of experiments involving DRG neurons cultured from wild type or constitutive PTPσ knockout animals

(Shen et al. 2009). Wild type DRG neurons were predictably inhibited with stunted axonal growth when grown on CSGPs; however PTPσ knockout DRGs were able to extend long axons similarly to wild type DRGs grown on poly-l-lysine. PTPσ knockout DRGs were additionally able to extend past the aggrecan gradient in spot assays and axons were able to extend farther distances than wild type axons in a model of dorsal column spinal cord crush. Furthermore, chondroitinase ABC digestion of the CS-GAGs of CSPGs were able to distrupt PTPσ binding with CSPGs in vivo.

Efforts from Coles et al. further elucidated the kinematics of PTPσ and HSPG/CSPG binding. Using a series of in silico and oligomerization assays, work from this seminal paper proposed that the GAG domains of HSPGs promoted aggregation or clustering of

PTPσ on the membrane of growth cones while the CS-GAG domain of CSPGs encouraged

PTPσ monomerization. It has been proposed that the uneven distribution of sulfated moieties of the GAG domain of HSPGs may be what promotes PTPσ clustering (Coles et al. 2011). The conformation of PTPσ in either oligomers or monomers on the growth cone membrane may be important to their subsequent signaling capabilities in that studies of other receptor tyrosine phosphatases, for example, have shown that oligomerization of other members of this family of receptors leads to inhibition of their phosphatase activity

(Bilwes et al. 1996; Wallace et al. 1998). The binding of HSPG or CSPG ligands may therefore serve as either an “off” or “on” switch of their intracellular phosphatase activity respectively (Shen 2014).

Indeed, these results have been recently validated from the Tremblay group who generated a split luciferase assay to simultaneously monitor PTPσ dimerization and phosphatase activity in living cells and found that HSPGs induced dimerization of PTPσ while CSPGs inhibited dimerization to allow enhanced phosphatase activity (Wu et al. 2017).

Furthermore, the uneven distribution of “islands” of clustered or monomeric PTPσ, depending on the context of the ECM, may mean that cellular phosphatase activity also becomes irregularly distributed creating a spatio-temporal regulation of signaling. Of course, this calls into question which downstream, secondary effector proteins may be targeted by PTPσ phosphatase activity. Although the mechanisms by which PTPσ mediates axon dynamics are still largely unknown, recent studies have pinpointed that the phosphorylation state of ERK may be involved (Kaplan et al. 2015; Lang et al. 2015), perhaps via the Rho/Rock pathway (Dyck et al. 2015).

While the mechanisms underlying CSPG/PTPσ signaling will require further elucidation, the relief of CSPG inhibition on axons in vivo through PTPσ perturbation is well known.

For example, PTPσ knockout mice show enhanced axon regeneration in a variety of models of CNS injury including different models of spinal cord injury including crush and contusive injuries (Shen et al. 2009; Fry et al. 2010), optic nerve lesion (Sapieha et al.

2005), and also in peripheral nerve injuries including sciatic nerve crush (McLean et al.

2002) and facial nerve crush (Thompson et al. 2003). Finally, results from our lab have illustrated the efficacy of peptide modulation of PTPσ in a variety of injury models as well including serotonergic sprouting in contusive spinal cord injury (Lang et al. 2015), sympathetic nerve re-innervation of a CSPG-enriched zone following myocardial infarction

(Gardner and Habecker 2013), and functional recovery following spinal root avulsion (H.

Li et al. 2015). Thus, treatments modulating PTPσ uniquely target the convergence of extrinsic, such as CSPGs upregulated following spinal cord injury, and intrinsic factors negatively affecting axon regeneration.

As PTPσ exhibits up to 60-80% sequence homology to its sister receptors, notably LAR, it is no surprise that LAR confers similar axon growth regulation as PTPσ (Tremblay 2009).

Indeed LAR also binds to HSPGs (Fox and Zinn 2005), to CSPGs in a dose-dependent manner (D. Fisher et al. 2011), and to itself perhaps in a clustering manner similar to PTPσ

(T. Yang et al. 2005). Constitutive LAR knockout mice have additionally been shown to exhibit enhanced serotonergic and cortical spinal tract axon collateral sprouting following

CNS injury (D. Fisher et al. 2011; B. Xu et al. 2015). The Li group has also developed a mimetic wedge peptide targeting the regulatory D1 intracellular domain of LAR and found some improvements following spinal cord injury (D. Fisher et al. 2011). While in vitro experiments using the Li group’s LAR peptide have been replicated successfully (Lang et al. 2015), successful replication of the LAR peptide in in vivo models of CNS injury have yet to be seen. Nonetheless, LAR is an interesting receptor more widely known for its role in synaptogenesis in drosophila (Clandinin et al. 2001; Stryker and Johnson 2007;

Hofmeyer and Treisman 2009) where downstream mechanisms including affecting RhoA,

Akt (D. Fisher et al. 2011; Dyck et al. 2015), Trk, PKC, MAPK, and PLCγ signaling (T.

Yang et al. 2005) have been more extensively studied than PTPσ. Perhaps future studies of

LAR will also help elucidate mechanisms underlying PTPσ signaling as well as its burgeoning roles in synaptogenesis (Um and J. Ko 2013).

It is also important to note that while PTPσ-targeting strategies have indeed underlined the salience of this receptor to mediating the axon-inhibitory effects of CSPGs, other receptors have also been known to bind to CSPG with subsequent convergent Rho/ROCK signaling, among which include semaphorin 3A, an axon guidance receptor well known for its axon repulsive effects during development (De Wit et al. 2005), and members of the Nogo family of receptors, specifically NgR1 and NgR3 (Dickendesher et al. 2012). However,

Nogo receptors are most well known for their axon inhibitory effects through myelin- associated and oligodendrocyte myelin glycoprotein binding (Schwab and Strittmatter

2014). Nonetheless, it is clear that CSPGs play a robust and negative role in axon regeneration, particularly mediated through PTPσ binding, following several CNS injury models.

1.2 Multiple Sclerosis and the Inhibitory Effects of Chondroitin Sulfate

Proteoglycans to Functional Recovery

While the exact etiology of autoreactivity is unknown, environmental and genetic factors converge to mediate Multiple Sclerosis, a progressively neurodegenerative disease initiated by myelin-specific CD4+ T cells that invade the CNS to cause multifocal demyelination, particularly in the white matter areas of the brain and spinal cord (Dendrou et al. 2015).

Myelin, produced from mature oligodendrocytes, acts as insulation enveloping axons to allow for saltatory conduction and to support axonal homeostasis; thus, its destruction results in neurodegeneration and functional failure over time. Neurodegeneration is

14 especially prevalent in the chronic progressive phase of the disease as seen through profuse damage of denuded axons (Criste et al. 2014). The hallmark of Multiple Sclerosis includes a relapsing-remitting phase whereby autoimmune-mediated flares result in demyelinated lesions that may repair initially. However, the progressive phase of the disease is noted by the eventual failure of remyelination and expansion of demyelinated areas chronically.

Whereas the etiology of Multiple Sclerosis is based on autoimmune reactivity, it shares similar inflammatory-based dysregulation at the cellular level as spinal cord injury. Like spinal cord injury, much of the damage seen in Multiple Sclerosis is initiated and further exasperated by inflammation. Infiltration of systemic immune cells through the

BSCB/BBB ignites an inflammatory cascade that includes further breakdown of the BSCB, activation of microglia and invading macrophages in multifocal plaques wherein hypoxia, and gliosis of notably astroctyes and oligodendrocyte progenitor cells occur (Mahad et al.

2015). Thus, the glial scar and deposition of CSPG in demyelinated plaques also occur in

Multiple Sclerosis (H. Mohan et al. 2010). The culmination of these events ultimately results in activation and death of oligodendrocyte progenitor cells and their subsequent inability to differentiate into a mature myelinating form, additional mature oligodendrocyte death, and demyelination in these randomly dispersed inflammatory plaques.

To begin, microglia and infiltrating leukocytes that have differentiated into macrophages within the CNS shift toward an M1-like proinflammatory response. M1-polarized macrophages have been known to activate oligodendrocyte progenitor cells and normally induce their proliferation while inhibiting their differentiation (Miller et al. 2007; D.L.

Taylor et al. 2010). On the other end of the spectrum, M2-like microglia normally promote

15 oligodendrocyte progenitor cell survival and differentiation through GSK3β and Akt signaling (G. Wang et al. 2015) to therefore encourage remyelination. In the context of

Multiple Sclerosis, M1 macrophages highly activated by the proinflammatory environment have been shown to induce oligodendrocyte apoptosis. Specifically, the conditioned media of M1-induced microglia, such as those activated in vitro by tumor necrosing factor (TNF)-

α, interferon (IFN)-γ, or lipopolysaccarhide, injected in vivo have been shown to reduce oligodendrocyte progenitor cell proliferation globally (D.L. Taylor et al. 2010). M1 microglia have additionally been shown to themselves induce demyelination through the secretion of proinflammatory factors such as TNFα (di Penta et al. 2013).

The release of proinflammatory factors from activated immune cells further induces damage to the CNS through hypoxic injury. In itself, this causes ionic imbalance, demyelination, axonal injury, and mitochondrial damage through the release of reactive oxygen species, reactive nitrogen species, and other factors that may cause further oxidative damage to axons and oligodendrocyte progenitor cells (Mahad et al. 2015). More over, inflammation within the CNS induces astrogliosis and glial scarring in demyelinated plaques (Moreno et al. 2013; Hostenbach et al. 2014). Normally, astrocytes support oligodendrocyte progenitor cells metabolically and encourage their differentiation through the secretion of factors such as CNTF (Robinson et al. 1998; Nash et al. 2010; Nash et al.

2011). Reactive astrocytes such as those galvanized by spinal cord injury and Multiple

Sclerosis, however, are detrimental to oligodendrocyte progenitor cell differentiation and further contribute to neurodegeneration by restricting remyelination (Rosen et al. 1989).

Neonatal oligodendrocyte progenitor cells transplanted into an astrocyte-free area of

16 demyelination are additionally able to differentiate and remyelinate significantly better

(Blakemore et al. 2003). One of the reasons for this inhibition of oligodendrocyte progenitor cell differentiation by reactive astrocytes is due to the deposition of oligodendrocyte progenitor cell-inhibiting CSPGs.

Following lysophosphatidylcholine/lysolecithin (LPC)-mediated demyelination injected into the mouse spinal cord, reactive astrocytes deposit CSPGs including versican, aggrecan, and neurocan (Fuller et al. 2007; Lau et al. 2012; Keough et al. 2016). Reactive astrocytes have also been known to secrete Tenascin-C (Nash et al. 2011), a glycoprotein known to prevent formation of myelin in a RhoA-dependent manner (Czopka et al. 2009). Indeed, demyelinated plaques formed in Multiple Sclerosis are replete with CSPGs, and in the LPC model of focal demyelination, must be cleared before remyelination may occur (Lau et al.

2012). In a mouse model of inflammatory demyelination, experimental autoimmune encephalomyelitis (EAE), gene transcripts of lectican CSPGs including brevican, neurocan, versican, and especially aggrecan were greatly upregulated in active compared to inactive demyelinating lesions (H. Mohan et al. 2010). Human cases of active Multiple Sclerosis plaques additionally display enrichment of versican, aggrecan, and neurocan along the edges of the lesion in association with reactive astrocytes; however, non-active repairing lesions showed decreased CSPG load in contrast to increased myelin (Sobel and Ahmed

2001). Reducing the formation of CSPGs such as injection of xyloside (Lau et al. 2012), which inhibits CSPG synthesis, or fluorosamine, which prevents some sulfation of the CS-

GAG chains, promotes remyelination in LPC and even the reduces the proliferation of T cells to ameliorate EAE disease progression (Keough et al. 2016).

Together with oligodendrocyte progenitor cell death and inhibition of its differentiation, these events cause axon degeneration over time as classically visualized by early work from the Trapp group who saw transection and dystrophic-like bulges along the length of the axons in mouse models of Multiple Sclerosis (A. Chang et al. 2002). Denuded axons are especially vulnerable to energy deprivation resulting from the loss of mitochondria and the increase of ATP consumption, oxidative stress from gliosis, and damage occurring from the withdrawal of myelin-derived trophic support (Criste et al. 2014). Over time, the decreased capacity for remyelination in progressive stages of Multiple Sclerosis causes axonal degeneration and subsequent functional loss that may manifest itself as several debilitating symptoms such as ataxia, spasticity, visual system disturbances, autonomic dysregulation, loss of sensation, cognitive impairment, and impairment of respiratory function among many others.

1.2.1 Oligodendrocyte Progenitor Cell Dysregulation in Multiple Sclerosis and

Contributions of Chondroitin Sulfate Proteoglycan and Protein Tyrosine

Phosphatase Sigma to Remyelination Failure

Oligodendrocyte progenitor cells must mature through varying stages including pre- progenitor and immature stages before possessing the ability to myelinate in its mature form. Specific antibodies can be utilized to ascertain the different stages of oligodendrocyte development. To simplify, oligodendrocyte progenitor cells can be visualized using antibodies against NG2, the PDGFα receptor, oligodendrocyte (O)-4, and oligodendrocyte

18 lineage gene (Olig)-2. The mature oligodendrocyte may also exhibit overlapping expression with O4 but can be predominantly determined using antibodies against myelin variants including myelin basic protein (MBP), myelin proteolipid protein (PLP), and myelin oligodendrocyte glycoprotein (MOG) (Bergles and Richardson 2015). Spinal cord oligodendrocyte progenitor cells derive from the ventral area of the neural tube during development where they migrate dorsally to populate the rest of the cord (Baumann and

Pham-Dinh 2001). In the adult cord, satellite oligodendrocyte progenitor cells of the gray matter and myelinating oligodendrocyte progenitor cells of the white matter constantly proliferate to sustain a clutch of self-renewing progenitor cells as well as to generate mature oligodendrocytes. Interestingly, oligodendrocyte progenitor cells use blood vessels to migrate to their destination (Tsai et al. 2012) possibly through integrin receptors and laminin signaling. However, in the context of a pro-inflammatory environment as seen in

Multiple Sclerosis, oligodendrocyte progenitor cells have been observed to devolve into a neonatal phenotype that initially contributes to increased migration into demyelinated plaques during the relapsing-remitting phase (Franklin and ffrench-Constant 2008).

Overtime, however, this process is compromised and the burden of increased oligodendrocyte progenitor cell activation and inhibition of differentiation decreases the capacity of these cells to remyelinate and restore function in their mature forms.

Specifically, oligodendrocyte progenitor cell recruitment and differentiation into mature oligodendrocytes is a rate-limiting step for remyelination in Multiple Sclerosis (Boyd et al.

2013). Aside from myelinating and maintaining this insulation around axons, oligodendrocytes are additionally vital to neuronal survival as they secrete trophic support to sustain axons (Baumann and Pham-Dinh 2001; Oluich et al. 2012). Thus, efforts to

19 restore oligodendrocyte progenitor cell homeostasis specifically by improving survival and migration despite an inhibitory ECM and encouraging oligodendrocyte maturation are attractive targets to enhance remyelination and restore function following Multiple

Sclerosis.

Dysregulation of oligodendrocyte progenitor cell homeostasis in Multiple Sclerosis begins with increased oligodendrocyte progenitor cell death in the midst of a pro-inflammatory and inhibitory environment. The release of pro-inflammatory cytokines such as IFNγ,

FasL, TNFα, and reactive oxygen species in such an environment induces mitochondrial dysfunction and apoptosis in mature oligodendrocytes (Cudrici et al. 2006). While oligodendrocyte progenitor cells become activated, proliferate, and migrate following inflammation, their responses ultimately hinder remyelination. oligodendrocyte progenitor cells have been found in chronically demyelinated lesions especially in the penumbra of these regions where CSPGs are highly enriched. In chronic stages of Multiple Sclerosis, lesion-associated oligodendrocyte progenitor cells are additionally reduced and unable to differentiate (Wolswijk 1998; A. Chang et al. 2000; Kuhlmann et al. 2008). In human cases of Multiple Sclerosis, Lucchinetti et al. have observed that 30% of Multiple Sclerosis lesions are completely devoid of oligodendrocyte progenitor cells while 70% have an increase of oligodendrocyte progenitor cells that were unable to differentiate (Lucchinetti et al. 1999). Characterization of these lesion-inhabiting oligodendrocyte progenitor cells has further demonstrated that they are more neonatal in phenotype and produce proinflammatory factors such as IL-1β and CCL2 (Vela et al. 2002). Along with their resistance to differentiation, these activated oligodendrocyte progenitor cells have been

20 observed to self-proliferate and feed forward into the proinflammatory response to drive oxidative damage and further tissue injury in other models of CNS injury (Rodriguez et al.

2014). Many of these oligodendrocyte progenitor cells additionally express the CSPG

NG2, which may be cleaved and secreted in the ECM (Asher et al. 2005) to further contribute to an inhibitory CSPG-laden lesion environment. Thus, oligodendrocyte progenitor cells in CSPG-upregulated demyelinated plaques have been found to be proinflammatory and unable to differentiate into mature myelinating forms.

Recent work has begun to elucidate the role of PTPσ in the inhibition of oligodendrocyte progenitor cell homeostasis and differentiation. Siebert et al. in 2011 first reported that oligodendrocyte progenitor cells could be inhibited by CSPGs in vitro. Specifically, oligodendrocyte progenitor cells plated on low concentrations of CSPGs were found to possess fewer extending processes and membrane sheets as visualized by MBP. This was the first indication that CSPGs alone could alter oligodendrocyte progenitor cell morphology. However, chondroitinase ABC treatment of CSPGs restored MBP volume as well as arborization of process extension. In fact, the same group found that chondroitinase

ABC treatment following spinal cord injury was able to increase the migration of oligodendrocyte progenitor cells into the glial scar (Siebert and Osterhout 2011). Like neurons, oligodendrocyte progenitor cells therefore seemed to be inhibited by the CS-GAG domain of CSPGs suggesting that PTPσ could be a likely cognate receptor responsible for these CSPG-induced inhibited effects. Indeed, oligodendrocyte progenitor cells were found to express PTPσ soon after (Pendleton et al. 2013). Other groups have cultured oligodendrocytes on aggrecan and other CSPGs such as neurocan and NG2 to confirm that

MBP surface volume, oligodendrocyte progenitor cell differentiation, and remyelination of peripheral axons was decreased in the presence of CSPGs. In all cases, these phenotypes were rescued by chondroitinase ABC treatment (Pendleton et al. 2013; Karus et al. 2015).

These CSPG-mediated effects were propagated through PTPσ perhaps, in part, through

Rho and ROCK signaling as RNA-mediated knockdown of the receptor restored MBP area and process extension despite the presence of CSPGs (Pendleton et al. 2013; Dyck et al.

2015). Additionally, the Yong group has independently confirmed CSPG inhibition of oligodendrocyte progenitor cells including impaired oligodendrocyte progenitor cell differentiation by CSPGs that was rescued through chondroitinase ABC treatment (Lau et al. 2012). Subsequent work form the Yong group found that the reduction of CSPGs improved remyelination in an LPC-induced demyelination model (Lau et al. 2012) and

EAE model of inflammatory demyelination (Keough et al. 2016) showing proof of principle that amelioration of CSPGs may have therapeutic effects in models of Multiple

Sclerosis .

1.3 Relief of Chondroitin Sulfate Proteoglycan-Mediated Inhibition through

Degradation by Proteases

There are three over-arching strategies by which to ameliorate the inhibitory effects of

CSPGs in vivo. First, many groups have targeted the enzymes responsible for de novo

CSPG production following inflammatory events induced by injury using novel drugs that prevent the post-translational addition of the inhibitory sugar moieties onto the protein core through inhibition of the enzyme, xylosyltransferase (Schwartz 1977; Grimpe and J. Silver

2004). These enzyme-inhibiting drugs include xyloside (Rolls et al. 2008; Lau et al. 2012) and most recently, fluorosamine (Keough et al. 2016); however, there appears to be controversies regarding the efficacy of such a strategy especially in regard to timing after injury. Rolls et al. for example, has reported that CSPGs play an acute role in sequestering inflammatory elements shortly following spinal cord injury and that inhibition of this scarring process unleashes a more devastating inflammatory response in the cord (Rolls et al. 2008). Additionally, long-term systemic administration of CSPG-synthesizing enzyme inhibitors may wreak havoc on joints maintained by aggrecan-producing chondrocytes, which may preclude its use long-term in chronic injuries. The second method by which to mitigate CSPGs’ inhibitory effects involves targeting the signaling cascade CSPGs invoke including the use of Rho and ROCK kinase inhibitors, which has been reported to yield limited functional recovery (Lord-Fontaine et al. 2008; Forgione and Fehlings 2014). The

Silver lab has recently reported the synthesis of a peptide we call Intracellular Sigma

Peptide (ISP) as it targets the intracellular regulatory D1 domain of PTPσ (Lang et al.

2015). Results from this thesis additionally report that this peptide modulates PTPσ in a way that allows for protease release and activation that degrades CSPGs, which comprises the third strategy by which CSPG inhibition could be targeted.

Conventionally, CSPG degradation has been accomplished in vitro and in vivo by the bacterially derived chondroitinase ABC enzyme, which catalyzes the excision of chondroitin-4 and chondroitin-6 sulfate residues from the CS-GAG chain (Fawcett 2015).

While this strategy has yielded functional recovery following many models of injury, especially spinal cord injury (Bradbury et al. 2002; Alilain et al. 2011; Bradbury and

Carter 2011) the enzyme itself must be invasively administered through an injection into the cord which itself causes damage, and is limited by short term action as it is short-lived under physiological conditions (Crespo et al. 2007). However several endogenous families of proteases have been reported to degrade CSPGs, and if precisely regulated, could present an attractive strategy to specifically target inhibitory CSPGs well after injury. While many proteases are capable of CSPG degradation, expounded in the results of this thesis include upregulation of Cathepsin B by peripheral axons and metalloprotease (MMP)-2 by oligodendrocyte progenitor cells following ISP modulation of PTPσ.

Proteinases, or proteases, encompass a variety of enzymes defined by their ability to catalytically lyse proteins through the hydrolysis of their peptide bonds. Broadly, the families of proteases are classified into seven groups based on their catalytic residues, which include: serine, cysteine, threonine, aspartic, glutamic, metallo-, and asparagine proteases. Currently, at least 600 mammalian proteases have been identified through comparative genomic approaches (Puente et al. 2003; Rawlings et al. 2004) and have been implicated in a myriad of biological processes beyond what was proposed to be their primary roles in protein degradation and turn over in the gut (Turk et al. 2012). However the discovery of thrombin and trypsinogen activation in blood clotting broached the idea that proteases were in fact precisely regulated to encompass a greater range of signaling processes in the immune system, during development, cell homeostasis, apoptosis, wound healing, and tissue and finer ECM remodeling (Davie and Ratnoff 1964; Turk et al. 2012).

Indeed, protease activity is tightly regulated at every level to prevent disastrous off-target effects including perturbed cell signaling or damage to other proteins or sub-cellular structures. Specifically, protease production at the transcription and post-translation processing levels, protease activation, and inhibition by specific endogenous inhibitors are some of the convenient points where protease regulation can be enforced (Barrett 2002;

Bai and Pfaff 2011). To begin, proteases are translated and processed through the endoplasmic reticulum as catalytically inactive pro-forms of the enzyme called zymogens.

Zymogens are not fully allowed to be activated until they are transported to their subcellular localization where their intended substrates are found (Barrett 2002). While some zymogens may autoactivate through self-cleavage or through low pH activation, most zymogens are activated through cleavage of their pro-peptide domains through other previously activated proteases (Olson and Joyce 2015). In addition to a catalytic-activation step, some proteases such as those in the metalloprotease family also require metallic ion cofactors such as zinc to become catalytically active (McFarlane 2003). The requirement for specific cofactors or additional pH requirements further adds another level of control that forces proteases to be dependent on their subcellular localization (Barrett 2002).

However, it is important to think of these regulatory steps as occurring in a network instead of a one-two step cascade as more than one protease is responsible for activating another.

Adding to this complication is that each protease often has more than one substrate including other proteases. Thus, in regards to the regulation of proteases, it is essential to consider the timing and the specific localization of protease activity in addition to their specific upstream activators and substrate availability.

In contrast to the normally tightly controlled temporal-spatial regulation of proteases, promiscuous and often rampant protease activation has been described in several models of

CNS injury and neurodegeneration. Inflammatory processes incurred by spinal cord injury, for example, initiates a protease storm whereby promiscuous protease activity is unleashed in a non-regulated way that instigates tissue damage and furthers the inflammatory cascade

(Noble et al. 2002; Veeravalli et al. 2012; Brkic et al. 2015). As proteases are ubiquitously expressed, upregulated protease transcription and secretion has been noted in a variety of different cell types including immune cells, oligodendrocyte progenitor cells, astrocytes, and even neurons (Brkic et al. 2015). For example, a wide range of pro-apoptotic proteases released following impact and subsequent cellular damage can further neuronal and glial death around the lesion epicenter (Veeravalli et al. 2012). Additionally, mRNA levels of

MMPs-3, 7, 9, 10, 11, 13, 19, and 20 have been reported to become upregulated as quickly as 24 hours after spinal cord injury to persist for almost a week after injury (Wells et al.

2003). Upregulated MMPs such as MMP-3 and 9 furthers tissue damage after injury through indiscriminate degradation of the components of the BSCB (Noble et al. 2002;

Yong et al. 2007) including proteins of the basal lamina that keeps the BSCB chronically leaky (Veeravalli et al. 2012). MMP-3, for example, has been shown to be secreted from endothelial cells after spinal cord injury to activate microglia, which contributed to oligodendrocyte apoptosis (J.Y. Lee et al. 2015). Long-term disruption of the BSCB further contributes to oxidative stress and a progressive, low-grade inflammatory response

(Rosell and Lo 2008), which results in continual leukocyte extravasation and infiltration

(Trivedi et al. 2006; Haoqian Zhang et al. 2011), demyelination (Starckx et al. 2003;

Agrawal et al. 2008), and may even play a role in chronic pain following spinal cord injury

(Ji et al. 2009; Miranpuri et al. 2016).

Likewise, unregulated protease activity following inflammation in models of Multiple

Sclerosis further disrupts the BSCB to exacerbate peripheral leukocyte infiltration (Cuzner et al. 1996; Leppert et al. 1996; Lindberg et al. 2001). In human cases of Multiple

Sclerosis, proteases such as MMP-9 have been observed in the cerebral spinal fluid of patients (Leppert et al. 1998; Correale and de los Milagros Bassani Molinas 2003) and upregulated MMP-7 and 9 have been correlated with peak progression of the disease

(Kieseier et al. 1998). Adding to increasing inflammation in Multiple Sclerosis, protease dysregulation additionally increases oxidative stress through their non-specific catabolism and have been shown to break down myelin (Brkic et al. 2015). Surely rampant protease activity plays a role in the development of demyelinated plaques and axonal destruction in

Multiple Sclerosis (Agrawal et al. 2008; Brkic et al. 2015).

While dysregulated protease release and activity clearly have potent and detrimental effects in the inflammatory milieu, greater emphasis must be placed on the specific actions of proteases that are highly regulated in a temporal-spatial manner to allow for cellular homeostasis. Emphasized here is the use of proteases well after pro-inflammatory events to degrade inhibitory CSPGs that have been upregulated during CNS insult. A variety of different proteases, for example, have been known to be specifically secreted by axons to enhance CSPG-restricted plasticity in adulthood. For oligodendrocyte progenitor cells,

27 growing evidence suggests that proteases are important for their survival, differentiation, and axon myelination.

1.3.1 Protease-Mediated Digestion of Chondroitin Sulfate Proteoglycans by

Neurons

Protease release by the growth cones of axons have been observed since the eighties by the

Nicholas Seeds group who discovered that dissociated neurons were capable of expressing

ECM-remodeling proteases (Krystosek and Seeds 1984). The Seeds group has since broached the vital role of proteases in neuronal migration during development (Friedman and Seeds 1994; Seeds et al. 1999), after peripheral nerve injury (Siconolfi and Seeds

2003), and in an activity-dependent manner that has been implicated in memory function and plasticity (Seeds et al. 1995). Since these initial findings, many groups have confirmed the role of precisely regulated protease activity in ECM remodeling to enhance neural plasticity (Mataga et al. 2004). Recent examples include works from Bijata et al. who have found that serotonergic stimulation increases MMP-9 activity to trigger dendritic spine remodeling (Bijata et al. 2017), and Pademsey et al. who found that activity dependent lysosomal release of cathepsin B regulates the long-term structural plasticity of dendritic spines (Padamsey et al. 2017).

Proteases additionally play a role in the ability of peripheral axons to increase axonal outgrowth past an inhibitory ECM. Different proteases, among them tissue plasminogen activator (tPA), have been implicated in increasing the axon outgrowth response following

28 a conditional injury, a process which transcriptionally upregulates growth-promoting factors in the periphery (Siconolfi and Seeds 2003; Minor et al. 2009). DRGs have additionally been observed to express MMP-2, which is transported specifically to the growth cone to increase neurite outgrowth (Zuo et al. 1998).

Along with peripheral injury, the use of proteases as another strategy to remove CSPGs following injury has also been widely studied. Knockout of MMP-2, for example, resulted in fewer serotonergic fibers sprouting following spinal cord injury (Hsu et al. 2006).

Aggrecanase, or distintegrin-like and metalloprotease with thrombospondin type 1 motif 4

(ADAMTS4) is one such protease that specifically targets not only the axon inhibitory CS-

GAG chains upregulated following spinal cord injury, but the chondroitinase ABC- resistant core protein and its sugar stubs as well (Schmalfeldt et al. 2000). In fact, in combination with chondroitinase ABC, ADAMTS4 has been shown to have a drastically pro-neurite outgrowth effect in vitro (Cua et al. 2013). Moreover, functional regeneration has been noted to increase following the upregulation of ADAMTS-4 through the application of interleukin-1α treatment after contusive spinal cord injury (Lemke et al.

2010). Besides ADAMTS-4, other proteases or protease activators such as urokinase plasminogen activator (uPA) (Seeds et al. 2011) or tPA (Minor et al. 2009) has been shown to promote axonal outgrowth following instances of spinal cord injury. Though these studies highlight the proof of principle concept that CSPG degradation by proteases could be mediated in a spatially and temporally specific manner to encourage axonal outgrowth following spinal cord injury while avoiding exasperation of the inflammatory

29 stage, there remains some limits to this strategy including how to specifically upregulate

CSPG-digesting proteases in axonal growth cones without invasive injections into the cord.

To this end, the work in this thesis has identified specific expression of cathepsin B in neurons following peptide modulation of PTPσ. In the spinal cord, cathepsin B is restricted to neurons although it is upregulated in immune cells acutely following spinal cord injury

(Ellis et al. 2005). That said, the potent endopeptidase activity of cathepsin B especially as it relates to degradation of CSPGs and its role in activating other CSPG-degrading proteases makes cathepsin B an attractive target to study not only in the context of spinal cord injury, but in other lysosomal-affected neurodegenerative disorders as well. Cathepsin

B is part of the cysteine protease superfamily that encompasses other cathepsins of which there are at least twenty variants labeled A-Z (Chen et al. 2017). First termed by Richard

Willstatter and Eugen Bamann to denote its kathepsein or “digestive” properties, cathepsins were first described for their proteolytic abilities derived from leukocytes activated by low pH (Verma et al. 2016). Since Willstatter and Bamann, we have come to understand that cathepsins are ubiquitiously expressed in all cell types within lysosomes. As cathepsins require a low pH environment to become activated, lysosomes provide the ideal subcellular structure in which to sequester potently degradative cathepsins as v-type ATP-powered H+ ion pumps maintain the lysosome environment around pH 5 compared to the pH 7.2 of the cytoplasm (Neha Aggarwal 2014). Thus, cathepsins are integral to the processes associated with lysosomes including protein metabolism, degradation, turnover, antigen presentation, growth factor receptor recycling, cell death, phagocytosis, and notably autophagy (Turk et al. 2012).

The role of cathepsins, especially cathepsin B, has been best studied in the role of the tumor microenvironment as it relates to cancer cell metastasis and migration (Olson and Joyce

2015). Increased cathepsin expression, for example, is highly correlated with poor prognosis in a variety of cancers including breast, lung, and nasal pharyngeal (Berdowska

2004; Jedeszko and Sloane 2004). Further research has uncloaked the role of cathepsin B in ECM degradation and the establishment of “invasion trails” upon which other migrating tumors use as a guide to expand the tumor environment (Olson and Joyce 2015). Inhibition of cathepsin B activity using the drug CA074 has additionally been reported to attenuate inflammatory breast cancer invasion through decreasing tumor cell migration across a more intact ECM (Victor et al. 2011). Cathepsin B plays a role in this process by cleaving cell- cell adhesion molecules (Sevenich and Joyce 2014) and importantly, through its secretion into the extracellular space, degrades inhibitory ECM proteins to allow for greater cell migration (Gocheva et al. 2010).

Cysteine cathepsins have been noted for their promiscuous substrate selectivity based on their specific endopeptidase activity on common amino acids such as glycine or phenylalanine (Biniossek et al. 2011). Cathepsin L, for example, has been identified to have 845 cleavage sites in the most comprehensive proteomic protease cleavage site to date

(Biniossek et al. 2011). Included in this range of substrates are most certainly inhibitory components of the ECM, as cathepsin B has been well known to be able to cleave aggrecan

(Fosang et al. 1992; Fosang et al. 1995; Mort et al. 1998). As cathepsins are able to degrade such a wide range of substrates, activate other proteases through cleavage of their

31 pro-peptide domains, and are able to activate themselves through autocleavage activity, cathepsins present an interesting “hub-like” status within the cell’s proteolytic network.

Cathepsin B, for example, has been noted to be able to activate a variety of other proteases such as MMP-2, 3, uPA, and even other cathepsins (Mason and Joyce 2011). Thus, cathepsin B represents a highly integrated node in this network by which a proteolytic cascade to concertedly remodel the ECM can take place.

However, because of this overlap or promiscuity in substrate specificity, cathepsin B must be tightly regulated to prevent rampant structural cell damage. One way cells have accomplished this task is through transcriptional regulation of the inhibitor of cathepsin B, stefin or cystatin B, instead of the protease itself (Neha Aggarwal 2014). Another is to regulate how and where cathepsin B may be secreted beyond its subcellular localization in the lysosome. For instance, cathepsin B has been observed in secretory vesicle-enriched fractions (Sendler et al. 2016) suggesting they may be sorted and excreted through the secretory autophagic signaling pathway, a process that has been well-known to sort and secrete lysosomal contents (Sasaki and Yoshida 2015). Moreover, cathepsin B has been known to localize in membrane microdomains including caveolae in tumor and endothelial cells in order to degrade the ECM (Cavallo-Medved et al. 2003; Cavallo-Medved et al.

2009). Cathepsin B exocytosis has been additionally observed in other cell types such as fibroblasts or endothelial cells in an annexin II regulated manner (Mohamed and Sloane

2006), and in neurons as well (Padamsey et al. 2017).

In neurons, because of its importance in normal homeostasis, cathepsin B has been found to play an integral role in many neurodegenerative diseases where lysosomal function is compromised such as Alzheimer’s Disease (Cataldo and Nixon 1990), amyotrophic lateral sclerosis (Kikuchi et al. 2003), Parkinson’s disease (McGlinchey and J.C. Lee 2015), and

Huntington’s Disease (Bahr et al. 2012). Cathepsin B activity, for example, has been shown to improve β-amylodiosis and improve learning and memory in models of mouse

Alzheimer’s Disease (C. Wang et al. 2012; Embury et al. 2017; Farizatto et al. 2017) while genetic knockout of cathepsin B leads to lysosomal dysfunction and enhanced amyloid-β and amyloid precursor protein load (Cermak et al. 2016). Furthermore, inhibition of cathepsins or suppressing lysosomal acidification caused dystrophic swellings along axons that have inhibitory consequences on axon transport and general homeostasis

(S. Lee et al. 2011). It is becoming clear that cathepsin B secretion is important in memory function as cathepsin B levels have been found to positively correlate with fitness and hippocampus-dependent memory function, perhaps due to its association with lysosomes and autophagy (Moon et al. 2016). Recently, Padamsey et al. has pinpointed the activity- dependent exocytosis of cathepsin B in ECM remodeling to enhance dendritic spine structural plasticity (Padamsey et al. 2017). In fact, the “hub-like” role of cathepsin B has been implicated to trigger MMP-9 (Padamsey et al. 2017) and MMP-2 activity (Tholen et al. 2014) in neurons to precisely remodel their surrounding ECM and enhance structural plasticity.

1.3.2 Proteases and Oligodendrocyte Homeostasis

Aside from the use of proteases to modulate cell-surface receptor signaling in oligodendrocyte progenitor cell differentiation and proliferation among other functions

(Bai and Pfaff 2011; Valente et al. 2015), little is known about the use of proteases by oligodendrocyte progenitor cells to specifically remodel their surrounding ECM.

Oligodendrocyte progenitor cells, however, have been known to secrete MMPs that shed their membrane-bound NG2 proteoglycan (Asher et al. 2005) in order to modulate AMPA receptor currents (Sakry et al. 2014). Recent studies on metalloproteases (MMPs) in oligodendrocyte progenitor cell cultures are beginning to illustrate the importance of these proteases and the content of their surrounding ECM outside of development. Specifically,

MMPs appear to play vital roles in oligodendrocyte progenitor cell maturation and homeostasis that ultimately culminate in their ability to remyelinate as mature oligodendrocytes.

MMPs, specifically MMP-1, were initially identified as collagen cleaving enzymes in 1962 by Gross and Lapiere (Gross and Lapiere 1962), but have since been identified in a myriad of cell signaling processes among them regulating cellular signaling cascades such as growth factor, cytokine, cell surface transmembrane and protein processing to effect cellular proliferation, death, motility, proliferation, and differentiation (Page-McCaw et al.

2007). Since the discovery of MMP-1, 23 other MMPs have been discovered in mice or 22 others in humans (Brkic et al. 2015). Much has been investigated about this particular family of metzincin, or zinc catalysis-dependent and calcium containing, endopeptidases through the use of single MMP knockout lines. Interestingly, all individual MMP knockout lines in the mouse are embryonically viable with subtle phenotypes probably owing to

34 either the dispensability of MMPs during development due to enzymatic redundancy or compensation by related MMP subtypes (Page-McCaw et al. 2007). Archetypal MMPs, for example, include three collagenases, two stromelysins, and four other MMPs with overlapping substrate specificity. Of particular interest in this thesis is the gelatinase subgroup of MMPs, which include MMP-2 and MMP-9.

Like most other proteases, MMPs are produced as catalytically inert zymogens, with a cysteine-bearing propeptide domain at the N-terminal that must be cleaved from the zinc- containing catalytic domain to be active (Geurts et al. 2012). Zymogens are transported from the endoplasmic reticulum Golgi complex packaged in extracellular vesicles and have been known to be secreted from cells to appear in the ECM or in body fluids (Shimoda and

Khokha 2013). While certain MMPs such as MT-MMP1 are membrane anchored due to their glycosylphosphatidylinositol anchor or transmembrane-1 domains, gelatinases are exclusively secreted and become catalytically stimulated by a network of activating proteases which may include cathepsin B, MT-MMP1, meprins, and other previously activated gelatinases (Bauvois 2012).

Activated secreted gelatinases have best studied for their ability to mobilize a variety of cancer cell lines including growth migration of mast cells (Yu and Stamenkovic 1999), metastatic melanomas (P.C. Brooks et al. 1996), and breast cancer cells (P.C. Brooks et al.

1998) among others. Elevated levels of plasma MMP-2 and MMP-9 have been found in human cases of breast, brain, ovarian, pancreatic, and other cancers making them biomarkers for especially aggressive tumor progression (Bauvois 2012). While integral

35 membrane protein cleavage of adhesion receptors such as CD44 or integrins have been implicated as gelatinase targets, cell migration is additionally enhanced through remodeling of the ECM to establish invasive trails by which other metastatic cells may follow (Page-

McCaw et al. 2007; Bauvois 2012). Indeed, a variety of ECM proteins have been implicated as substrates for both gelatinases including collagens, gelatin, and elastin

(Bauvois 2012). There also seems to be some substrate specificity, however, as MMP-2 has additionally been known to cleave tenascin, fibronectin (Bauvois 2011), and most importantly, NG2 (Asher et al. 2005) and other CSPGs (Zuo et al. 1998).

The ability of gelatinases to target CSPGs may be one important regulatory process that helps to sustain oligodendrocyte progenitor cell homeostasis once they have matured.

CSPGs, for example, have been shown to inhibit oligodendrocyte maturation and lineage progression (Karus et al. 2015). Indeed, a clutch of MMPs has been implicated in their vital roles for oligodendrocyte progenitor cell maintenance and homeostasis in part, perhaps, through the digestion of CSPG and other ECM elements. Both MMP-2 and 9 seem to be necessary for oligodendrocyte generation and commitment from human neural stem cells (Sypecka et al. 2017), and MMP-12 expression by oligodendrocyte progenitor cells seems to regulate their maturation and morphological differentiation as use of an inhibitory MMP-12 antibody prevented oligodendrocyte progenitor cells from expressing the mature marker MBP and from extending processes (Larsen and Yong 2004). Other studies have confirmed the role of MMP-9 in enhancing process outgrowth by oligodendrocytes (Uhm et al. 1998). Oh et al. demonstrated that oligodendrocyte progenitor cells express both gelatinases in vitro and are capable of in situ gelatinase

36 activity not only at the soma, but at the processes as well. The same study showed that

MMP-9 facilitates process outgrowth by oligodendrocytes as the use of an inhibitory

MMP-9 antibody significantly decreased process outgrowth and the ability of these oligodendrocyte progenitor cells to eventually form myelin during optic nerve development in an MMP9 null mouse line (Oh et al. 1999). Inhibiting these gelatinases not only impairs normal oligodendrocyte progenitor cell functioning, but has been shown to induce oligodendrocyte progenitor cell apoptosis as well since overexpression of TIMP-3, the endogenous inhibitor of MMP-2 and 9 (J. Hu et al. 2016) among others, mediates immature oligodendrocyte cell death (Y. Yang et al. 2011).

In addition to supporting oligodendrocyte progenitor cell differentiation, recent work has revealed the importance of MMPs in promoting oligodendrocyte progenitor cells’ ability to remyelinate after various demyelinating models. Conventionally, MMPs have been known to wreak havoc in demyelinating lesions in Multiple Sclerosis patients by initiating and sustaining inflammation (Trentini et al. 2016). However, recent work has begun to see a more nuanced role of MMPs in Multiple Sclerosis. In a cuprizone-mediated model of mouse corpus callosum demyelination, Skuljec et al. have highlighted how several MMPs are differentially up or down regulated across demyelinating and remyelinating phases in the brain (Skuljec et al. 2011). Gene expression of MMP3, for example, seems to slightly peak during demyelination, but then becomes greatly upregulated during remyelination. In contrast, MMP24 gene expression decreases steadily following demyelination and only begins to recover beginning around six weeks following cuprizone injection. Likewise, deletion of certain MMPs seems to delay remyelination. For instance, MMP-9 upregulation

37 by sildenafil (commonly known as Viagra) has been linked to promoting remyelination in cuprizone-demyelinating (de Santana Nunes et al. 2016) and LPC models (Larsen et al.

2003). Interestingly, MMP9 and 12 null mice show great delays of myelin formation during the development of the CNS (Larsen and Yong 2004) and MMP9 null mice show reduced numbers of mature oligodendrocytes and reduced remyelination after LPC injections partly through the ability of MMP-9 to cleave inhibitory NG2 (Larsen et al.

2003).

Recently, work from Pruvost et al. have additionally reported the ability of another metalloprotease from the ADAMTS family called ADAMTS-4, or aggrecanase, to enhance myelination. Characterization of the ADAMTS4 null mice revealed motor deficits in coordinated locomotion and rotarod tests as well as abnormal electrical activity in the somatosensory pathway due to abnormal myelin formation along the CSPG-rich periphery during development (Pruvost et al. 2017). While demyelinating studies were not performed by Pruvost et al., the importance of ADAMTS-4, which includes CSPGs as its main substrate, and other CSPG-degrading MMPs to myelination in the midst of CSPG-rich environments remains intriguing.

Thus the capability of MMPs in the ability of oligodendrocytes to remyelinate following

CNS insult may be due to a combination of several factors including sustaining the ability to maintain oligodendrocyte progenitor cell health and homeostasis and encouraging differentiation into mature, myelinating oligodendrocytes. CSPGs have been recently found

38 to impair both these processes in oligodendrocyte progenitor cells (Pendleton et al. 2013) and the expression of MMPs may encourage remyelination through its degradation as well.

1.4 Peptide Modulation of Receptor Protein Tyrosine Phosphatase Sigma and

Subsequent Protease Secretion to Relieve Chondroitin Sulfate Proteoglycan-

Mediated Inhibition

Given the potently inhibitory effects of CSPGs following a variety of CNS injuries and what was currently known about its cognate receptor, PTPσ appeared to be a promising therapeutic target to relieve CSPG inhibition following injury. Previous work from the

Brady-Kalnay lab had established a method by which to pharmacologically target a related

PTP receptor called PTPu by synthesizing a peptide inhibitor specific to the intracellular wedge domain located near the D1 catalytic domain (Oblander and Brady-Kalnay 2010).

This juxtamembrane domain of the receptor had previously been found to regulate phosphatase catalytic activity of the receptor (Bixby 2001) and peptide binding to the domain presumably confers phosphatase inhibition through steric hindrance of the site

(Hoffmann et al. 1997). Additionally, the peptide was synthesized with an HIV-derived

TAT sequence in the C terminus end to allow for cellular penetration, and peptide modulation of PTPu significantly altered neurite outgrowth in vitro as predicted. Using the same strategy, the Shu Xin Li group also synthesized a TAT-bound peptide to target LAR in order to enhance axon outgrowth in vitro (D. Fisher et al. 2011).

Recently, the Silver lab designed a TAT-linked peptide, called Intracellular Sigma Peptide

(ISP), against the wedge domain of PTPσ in a similar fashion (Lang et al. 2015). Indeed, peptide modulation of PTPσ significantly increased axon outgrowth past the inhibitory

CSPG gradient found in the in vitro spot assay in a dose-dependent manner. Subcutaneous administration of ISP moreover enhanced axon sprouting or regeneration and corresponding return of forelimb function in a model of spinal root avulsion where spinal roots are pulled away from the spinal cord (H. Li et al. 2015). Interestingly, systemic administration improved myocardial infarction where CSPGs are also upregulated by enhancing peripheral axon sprouting near the infarcted zone (Gardner et al. 2015).

Importantly, the seminal paper by Lang et al. characterized significant improvement in a variety of functions following moderate-to-severe contusive spinal cord injury at thoracic level 8 including enhanced sensorimotor, urinary, and coordinated locomotor functions.

Analysis of the sectioned spinal cord of ISP-treated rats further revealed significant sprouting of serotonergic fibers caudal to the thoracic level 8 injury site.

Recovery of coordinated locomotor functions was, in fact, dependent on these serotonergic sprouts as systemic administration of the serotonin antagonist drug, methysergide, ablated gains in coordinated walking scores. Further analysis of spinal cord sections caudal to the injury site revealed enhanced serotonergic expression especially along the ventral motor pools in ISP-treated animals. Additionally, when wisteria floribunda agglutinin (WFA) was used to visualize the GAG-CSPGs, GAGs were strikingly decreased in serotonergic- enriched regions in a pattern suggesting that serotonin staining fit in WFA-depleted areas in a “lock-and-key” manner. Adding to this observation, we had previously seen diminished

GAG staining in our in vitro spot assays and occasionally “shadows” of absent GAG-

CSPGs, similar to in vitro work performed by the Nicholas Seeds group in the 1980s, as

DRG axons negotiated their way past the CSPG spot barrier. These observations lead us to hypothesize that protease activity may be enhanced in neurons by ISP administration in vivo and in vitro and provided the rationale to systematically characterize protease activity by peptide modulation of PTPσ.

But what is the biological relevance of punctate protease release following PTPσ modulation? Neural plasticity or axon outgrowth following spinal cord injury is severely inhibited by upregulated CSPG in perineuronal nets and the glial scar. Similarly, demyelinated lesions seen in Multiple Sclerosis are also laden with CSPGs that have been shown to diminish oligodendrocyte progenitor cell health and their ability to differentiate into myelinating, mature oligodendrocytes. The work described here includes a variety of biochemical, in vitro, and ex vivo techniques in service of characterizing enhanced protease release following PTPσ modulation in both peripheral axons and oligodendrocytes. To this end, this work summarily presents a novel finding whereby peptide modulation of PTPσ enhances proteases release in both dorsal root ganglion neurons and oligodendrocyte progenitor cells that ultimately allows for relief of inhibition through degradation of CSPGs to enhance axon outgrowth and remyelination by oligodendrocytes.

II. MODULATION OF PROTEIN TYROSINE PHOSPHATASE

RECEPTOR SIGMA ENHANCES CATHEPSIN B RELEASE IN

DORSAL ROOT GANGLION NEURONS TO RELIEVE

CHONDROITIN SULFATE PROTEOGLYCAN INHIBITION AND

ENHANCE AXON OUTGROWTH

This chapter may also be found in full in the following publication:

Amanda Phuong Tran, Sapna Sundar, Meigen Yu, Bradley T. Lang, Jerry Silver.

Modulation of receptor protein tyrosine phosphatase sigma enhances chondroitin sulfate proteoglycan degradation through Cathepsin B secretion to enhance axon outgrowth.

Accepted Journal of Neuroscience, May 2018.

2.1 Abstract

Severed axon tips reform growth cones following spinal cord injury that fail to regenerate, in part, because they become embedded within an inhibitory extracellular matrix.

Chondroitin sulfate proteoglycans (CSPGs) are the major axon inhibitory matrix component that is increased within the lesion scar and in perineuronal nets around deafferented neurons. We have recently developed a novel peptide modulator (Intracellular

Sigma Peptide or ISP) of the cognate receptor of CSPGs, protein tyrosine phosphatase sigma (PTPσ), which has been shown to markedly improve sensorimotor function, micturition, and coordinated locomotor behavior in spinal cord contused rats. However, the mechanism(s) underlying how modulation of PTPσ mediates axon outgrowth through inhibitory CSPGs remain unclear. Here, we describe how ISP modulation of PTPσ induces enhanced protease Cathepsin B activity. Using dorsal root ganglion neurons from female

Sprague-Dawley rats cultured on an aggrecan/laminin spot assay and a combination of biochemical techniques, we provide evidence suggesting that modulation of PTPσ regulates secretion of proteases that, in turn, relieves CSPG inhibition through its digestion to allow axon migration though proteoglycan barriers. Understanding the mechanisms underlying

PTPσ modulation elucidates how axon regeneration is impaired by proteoglycans but can then be facilitated following injury.

Significance Statement

Following spinal cord injury, chondroitin sulfate proteoglycans (CSPGs) upregulate and potently inhibit axon regeneration and functional recovery. Protein tyrosine phosphatase

43 sigma (PTPσ) has been identified as a critical cognate receptor of CSPGs. We have previously characterized a synthetic peptide (ISP) that targets the regulatory intracellular domain of the receptor to allow axons to regenerate despite the presence of CSPGs. Here, we have found that one important mechanism by which peptide modulation of the receptor enhances axon outgrowth is through secretion of a protease, Cathepsin B, which enables digestion of CSPGs. This work links protease secretion to the CSPG receptor PTPσ for the first time with implications for understanding the molecular mechanisms underlying neural regeneration and plasticity.

2.2 Introduction

Chondroitin sulfate proteoglycans (CSPGs) comprise a large family of extracellular matrix glycoproteins with varying numbers and types of glycosaminoglycan (GAG) chains bound to a protein core. During development, CSPGs aid in axon guidance (Snow et al. 1990;

Brittis et al. 1992; Carulli et al. 2005) by serving as a molecular guardrail to prevent inappropriate targeting. In adulthood, CSPGs such as aggrecan are concentrated in synapse enwrapping structures called perineuronal nets where they have been implicated in a variety of functions including limiting axonal sprouting and neuroplasticity (Massey et al.

2006; Kwok et al. 2011; Tran et al. 2018). Following spinal cord injury, CSPGs also become highly upregulated in a gradient radiating outward from the astroglial scar epicenter as a result of various triggering elements that emanate from the leaky blood brain barrier (Schachtrup et al. 2010), inflammatory cells (Fitch et al. 1999) or fibroblastic-like cells in the lesion core (Okada et al. 2006). The gradient pattern of CSPG expression has been well characterized as a potent impediment to axon regeneration following spinal cord injury (Davies et al. 1997) and modeled in a spot assay in vitro where axons develop stalled, club-like dystrophic end balls once they encounter gradually changing concentrations of aggrecan and laminin (Tom et al. 2004).

The discovery of protein tyrosine phosphatase sigma (PTPσ) as a cognate receptor for the inhibitory actions of CSPGs was a major milestone in the search for potential therapeutic targets to promote regeneration (Shen et al. 2009). Indeed, we have recently reported the recovery of locomotor, sensory, and urinary function in an acute model of rat contusive

45 injury treated for 7 weeks with a daily systemic injection of Intracellular Sigma Peptide

(ISP) (Lang et al. 2015). ISP is a synthetic peptide designed to specifically modulate the intracellular wedge domain of PTPσ. However, the mechanism(s) underlying how ISP modification of PTPσ signaling can lead to long term CSPG disinhibition and subsequent axonal outgrowth are still unclear.

While characterizing the regeneration-promoting effects of ISP on adult dorsal root ganglion (DRG) axons using our in vitro CSPG gradient spot assay, we have observed that

ISP treatment markedly reduced the presence of the inhibitory GAG content of the substrate suggesting possible protease activity. Normal low-level protease-release by neurons to digest the extracellular matrix during their migration has been observed since the 1980s (Krystosek and Seeds 1984). For example, during development, retinal ganglion cell growth cones secrete the protease ADAMTS4 which enables their access to retino- topically appropriate locations within the CSPG-rich lateral geniculate nucleus (J.M.

Brooks et al. 2013). Furthermore, matrix metalloproteases (MMPs) have been noted for their CSPG-degrading effects (Cua et al. 2013) and MMP-2 knockout mice showed reduced serotonergic sprouting following spinal cord injury (Hsu et al. 2006). Thus, finely controlled spatio-temporal regulation of protease release by neurons is one strategy by which discrete CSPG degradation may occur to enhance axon regeneration without delivering further damage to the CNS.

Here, we present evidence that PTPσ may be regulating the release of Cathepsin B that specifically degrades CSPGs and that this is one mechanism by which peptide treated neurons may sustain axon outgrowth in a CSPG-rich environment.

2.3 Methods

Peptide Sequences

Peptides were purchased from GenScript or CS-BIO in lypholyzed form (>98% purity) and reconstituted in sterile dH20 at 2.5mM and aliquoted for storage in -20°C. Peptides were used at 2.5µM concentration unless stated.

Intracellular Sigma Peptide (ISP):

GRKKRRQRRRCDMAEHMERLKANDSLKLSQEYESI

Scrambled ISP (S-ISP): GRKKRRQRRRCIREDDSLMLYALAQEKKESNMHES

TAT: GRKKRRQRRC

Dorsal root ganglion (DRG) neuron culture

All culture experiments were performed under sterile conditions. Dorsal root ganglia

(DRGs) were cultured from adult female Sprague-Dawley rats (Harlan) according to

Case Western Reserve University IACUC guidelines. Spinal columns were extracted from the animal and a laminectomy was performed on ice under a dissection microscope

(Leica). Forceps (size 3, FST) were used to extract DRGs which were immediately placed on ice with calcium, magnesium-free media (50mL 10X Hanks Balanced Salt

Solution, 14185-052, Invitrogen, dH20, 0.174g Sodium Bicarbonate, Phenol Red).

Extracted DRGs were then axotomized with microscissors (15003-08, FST), delaminated with forceps, and cut in half before they were collected in a 1mL Eppendorf tube. Media was then replaced with pre-warmed Dispase (10165859001, Roche, 3.75

U/mL)/Collagenase II (4176, Worthington, 200U/mL) and incubated for 1 hour at 37°C,

48 then dissociated with gentle shaking for 30 minutes at room temperature. Dissociated ganglia were titurated with fire-polished Pasteur pipettes and excess myelin was filtered at 100µM (BD Falcon 352360). Following 3X5 minute 4k rotation per minute spin

(Eppendorf 5415D) washes with calcium, magnesium-free media, DRG pellet was resuspended in pre-warmed Neurobasal-A (10888-022, Invitrogen) with B-27 (17504-

044, Invitrogen), 0.02% Glutamax (35050061, Invitrogen), 0.01% Pen/Strep (15140-031,

Invitrogen) supplements at which point they were ready to be aliquoted for culture with

1µM Ara-C (C1768, Sigma) to prevent glial outgrowth.

Aggrecan spot assay

Aggrecan spots were made as previously described in Tom et al. 2004. Briefly, glass coverslips (120545082, Fisher) were acid-washed (1M hydrochloric acid) at 55°C o/n and rinsed before use in 24-well plates (35-3047, Falcon). Coverslips were incubated o/n with poly-L-lysine (100mg/mL, P1274, Sigma), washed 3X with dH20, and coated with a

40µL nitrocellulose (1.6cm2, BA83 Schleicher & Schuell)/methanol (100%, 12mL,

A412-4, Fisher) mixture on cloning rings. 700µg/mL aggrecan (A1960, Sigma) and

10µg/mL laminin (23017-015, Invitrogen) in calcium, magnesium-free media was mixed and 4 2µL spherical aliquots were carefully plated onto each coverslip and allowed to dry.

Spot assay to assess glycosamingoglycan-chondroitin sulfate proteoglycan (GAG-

CSPG) degradation

DRGs were cultured on poly-L-lysine, 1µg/mL aggrecan, and 10µg/mL laminin coated 6- well plates (35-3046, Falcon) with 1mL of media and immediately treated with vehicle or peptides for 2 or 4 days in vitro (div). Conditioned media (CM) was collected and cell strained through centrifuging or filtration (09-720-3, Fisher) before plating onto freshly made spots. CM was incubated on spots at 37°C for 2 days after which spots were stained with CS-56 (1:500, C8035, Sigma), a marker for the glycosaminoglycan (GAG) moieties of chondroitin sulfate proteoglycans (CSPGs). Each spot was imaged twice on a standard fluorescent microscope (Leica) using the same gain and exposure settings in a blinded fashion. To assess pixel intensities, a region-of-interest box with a set area was made in

ImageJ (NIH) and used for all experiments, measured the pixel intensities of the spot rim.

Five replicates with around 20 repeats were performed for each spot experiment. To validate that ISP-treated conditioned media (CM) degrades GAG-CSPGs on spots, we incubated 0.1U Chondroitinase ABC (ChABC, C3667, Sigma) at 37°C for 2 hours and quantified CS-56 immunoreactivity as described. To assess how long ISP needed to be incubated with DRGs to degrade spots, we collected media at 2div for Controls and

30min, 1, 4, 24, and 48 hours following ISP treatment before staining. To assess whether incubation with media alone would degrade CS-56, we incubated vehicle or peptides in calcium, magnesium-free media for 2div at 37°C before staining. To assess whether boiling CM would rescue CS-56 degradation, we collected CM from DRGs treated with vehicle or peptides for 2div and immediately plated one group on spots and boiled the collected CM from another group at 100C and added guanidine hydrocloride (5M) to fully denature the media before plating onto spots for 2div. For spots assaying ISP in conjunction with other drugs, DRGs were treated with vehicle or ISP and the following

50 drugs 24 hours after plating for 2div: dimethyl sulfoxide (DMSO, D8418, Sigma), Roche

Protease Inhibitor (0.1%, 05892970001, Roche), GM6001 (25µM, 364205, Calbiochem),

Anisomycin (20µg/mL, A5862, Sigma), Cycloheximide (20µg/mL, C7698, Sigma), α-

Amanitin (20µg/mL, A2263, Sigma), and Exo1 (10µg/mL, ab120292, Abcam).

Spot assays to assess dorsal root ganglia (DRG) axon outgrowth

DRGs were collected from one adult female rat and aliquoted amongst one 24-well plate with fresh spots as described above with Neurobasal-A with supplements. Vehicle or peptides were added immediately after plating for 4div at 37°C, 5% CO2. DRG coverslips were then fixed with 4% paraformaldehyde, washed with 1xPBS, and incubated with 5% normal goat serum (16210-064, Gibco) block, 0.1g bovine serum albumin (A3059,

Sigma), 0.1% Triton-X (X-100, Sigma) Block at room temperature for 1 hour. Coverslips were then stained with the following primary antibodies with block o/n at 4°C: CS-56

(1:500, Sigma), beta-tubulin III (Tuj1, 1:500, T8660, Sigma). Corresponding secondary antibodies conjugated with fluorophores were then used at o/n, 4°C incubation following

3X15min 1xPBS washes: anti-Mouse IgM Alexa Fluor-546 (1:500, A21045, Life

Technologies), anti-Mouse IgG2b Alexa Fluor-488 (1:500, A21121, Life Technologies).

Coverslips were then mounted on slides with Citifluor (17971-25, EM Sciences). The number of Tuj1+ axons crossing the CS-56 labeled spot rim was counted and normalized against the total number of Tuj1+ soma present in a blinded fashion on a standard fluorescent microscope (Leica) at 10X objective. For pretreated spots to assess axon outgrowth, the same method was used to culture DRGs, but spots were pretreated with the following for 24 hours before DRG plating: 0.1U/mL Chondroitinase ABC (ChABC

C3667, Sigma), recombinant Cathepsin B (CatB, 0.5µg/mL, 965-CY, R&D Research) activated in 0.1M Sodium Acetate, pH 5.0, recombinant Cystatin B (CSTB, 50µg, 1409-

PI, R&D Research), cell-strained CM collected from another set of 6-well DRGs treated with vehicle or peptides. Three-four replicates with up to 20 repeats were performed.

Western blot assay of glycosaminoglycan (GAG) degradation

To confirm GAG degradation, 100µL CM was collected from DRGs cultured on PLL, laminin/aggrecan as described. Cell-strained CM was incubated with 20µg/mL aggrecan at 37°C for 2 hours. 20µg protein was mixed with laemmli/beta-mercaptoethanol and denatured at 100C for 10min. Samples were loaded onto pre-cast SDS/PAGE gels (456-

1094, BD Sciences) and ran around 1.5hours at 100mV, then transferred onto PVDF membranes (Bio-Rad) o/n at 12mV on ice. Blots were blocked with Super Block (37515,

Thermoscientific) for 2 hours shaking at room temperature before incubation with the following antibodies o/n at 4°C: DIPEN (1:1000, 1042002, MD Biosciences), CS-56

(1:1000, Sigma). Following 3X15min 1xPBS-0.1%Tw-20 washes, corresponding secondary antibodies were incubated for 2 hours at RT or 4°C o/n with shaking: anti-

Mouse IgG-horse radish peroxidase (1:1000, 55563, ICN), anti-Mouse IgM-horse radish peroxidase (1:1000, AP128P, Millipore). Blots were developed using chemiluminescent kit (WBKLS0500, Milipore).

DQ Gelatin Assay

DRGs were cultured on PLL-incubated coverslips precoated with 1µg/mL aggrecan,

10µg/mL laminin, and 25μg/mL DQ gelatin (D12054, Life Technologies), which is a quenched gelatin conjugate that only fluoresces once it has been cleaved by proteases, for

2div. After fixing as previously described, DRGs were then stained with Tuj1 (1:500,

Sigma) with a Mouse IgG2b Alexa Fluor-546 secondary (1:500, Life Technologies).

Colocalization was assessed using Just Another Colocalisation Plugin (Bolte and

Cordelieres 2006) in ImageJ. Four replicates with 8 repeats each were performed.

Protease Activity (EnzChek) Assay of dorsal root ganglia (DRG) conditioned media

(CM)

To assay general protease activity in DRG CM, we utilized EnzChek (E6638,

ThermoFisher) which uses quenched casein that only fluoresces once it has been cleaved.

CM was collected from vehicle, 2.5µM ISP, or 2.5µM S-ISP treated DRGs cultured on poly-L-lysine, 1µg/mL aggrecan, and 10µg/mL laminin for 2 or 4div with 1mL media.

CM was cell strained through centrifuging and incubated with 1x EnzChek mixture in 1:4 ratio. 200µL of CM mixture was aliquoted onto a 96-well plate with 3 replications for each sample o/n at RT with gentle shaking. The plate was read at 435/535nm

(excitation/emission) and media-only was used as a blank.

Mass spectrometry of dorsal root ganglia (DRG) conditioned media (CM)

DRGs were grown on 6-well plates precoated with poly-L-lysine, 10µg/mL laminin/

1µg/mL aggrecan for 4div with 2.5µM ISP. This was repeated twice as process replicates

53 for the purpose of screening possible proteases and not for quantification. Cell-strained

CM was sent to for mass spectrometry analysis at the Case Western Reserve University

Protein Core. Proteome Discoverer version 1.3.0.0339 (ThermoFisher) was used to generate peaklist, Mascot version 2.3.00 (MatrixScience) as the search engine, Uniprot

Rat Database (December 2012 – 7,844 sequences) used as the sequence database, and an expectation value of <0.05 or lower was accepted with an estimated false discovery rate of 4.45%.

Western blot assessment of dorsal root ganglia (DRG) lysate and conditioned media

(CM)

DRGs were cultured on poly-L-lysine, laminin/aggrecan pre-coated 6-well plates with

1mL media/well and immediately dosed with vehicle or peptides for 4div. CM was collected, cell-strained and 100µg of each sample was processed for western blot as described above. To process DRG lysates for western blot analysis, DRGs were cell- scraped from 6-well plates cultured in a similar manner as described above. Lysates were incubated with Roche Protease Inhibitor and vortexed at max speed for 30min at 4°C.

Lysates were then centrifuged at max speed for 20min at 4°C. 20µg protein of each sample was processed for western blot as described above. The following primary antibodies were used for 4°C o/n incubation with gentle shaking: Cathepsin B (CatB,

1:1000, AF953, R&D), Cystatin B (CSTB, 1:500, AF1408, R&D), GAPDH (1:1000,

2118S, Cell Signaling Technologies), Histone-3 (1:1000, 4499P, Cell Signaling

Technologies). The following secondary antibodies were used following washes: anti-

Mouse IgG-horse radish peroxidase (1:1000, A9044, Sigma), anti-Rabbit IgG-horse

54 radish peroxidase (1:1000, P132P, Millipore), anti-Goat IgG-horse radish peroxidase

(1:1000, 55385, ICN).

Dorsal root ganglia (DRG) immunocytochemistry

DRGs were grown on poly-L-lysine incubated coverslips pre-coated with 1µg/mL aggrecan and 10µg/mL laminin in 24-well plate for 2div. Cell fixation and staining was performed similarly as DRG-spot staining with the following primary antibodies: Tuj1

(1:500, Sigma), Cathepsin B (CatB, 1:500, R&D), Cystatin B (CSTB, 1:500, R&D).

Corresponding previously described secondary antibodies were used.

Lenti-viral infection of Cystatin B (CSTB) over-expression on spot assays

Lenti-viral GFP control (sc-108084, Santa Cruz Biotech) and Cystatin B-overexpressing

(CSTBo/e) rat gene (LPP-CS-GS498J-Lv165-050, GeneCopoeia) constructs were purchased and aliquoted for storage at -80°C. 0.1 multiplicity of infection (MOI = cell number/virus particles in µL * stock concentration) of each lenti-viral particle group was aliquoted to inoculate DRGs on spot assays immediately upon plating for 4div. A subset of DRGs grown on aggrecan/laminin were grown on 6-well plates and inoculated with

GFP or CSTBo/e to use for western blot analyses. Fix and staining was performed as described above. Three replicates with four repeats were performed and analyzed.

Injured spinal cord tissue immunostaining

All procedures were performed according to Case Western Reserve University IACUC guidelines. Female Sprague-Dawley rats (225-250g, Harlan) were injected with ketamine

(60mg/kg)/xylazine (10mg/kg), thoracic sections T7-9 were exposed, laminectomized, and the cord was held taut to receive a 250kdyne contusive injury at T8 from an Infinite

Horizon Impact Device. Animals were injected with ISP one day after surgery

(subcutaneous, 22µg/mL with 5% dimethyl sulfoxide, DMSO) daily for 49 days. Animals were sacrificed 7 weeks after administration of last ISP injection, perfused, and fixed with 4% paraformaldehyde followed by 30% sucrose incubation overnight. Animal with the highest BBB score (20, ISP-treated) and lowest BBB score (9, DMSO control) were chosen for tissue processing. Sections (20µm/section, Hacker cryostat) were blocked with

0.1% bovine serum albumin, 0.1% Triton-X, and 5% normal goat serum in PBS (81873,

ThermoFisher). The following primary antibodies were diluted in blocking buffer and incubated overnight at 4°C: Serotonin (5-HT, 1:500, 20080, ImmunoStar), Wisteria

Floribunda Agglutinin (WFA, 1:50, L1516, Sigma), Cathepsin B (CatB, 1:500, R&D).

Following PBS-0.1%Tween-20 washes (3X15minutes), the following secondary antibodies were diluted in blocking buffer for overnight incubation: anti-Rabbit IgG

Alexa Fluor-546 (1:500, A-11035, Life Technologies) anti-Goat IgG Alexa Fluor-488

(1:500, A-11055, Life Technologies), and Streptavidin Alexa Fluor-546 (1:500, S112255,

ThermoFisher). Imaging took place on a standard fluorescent microscope (Leica).

Quantification of images below the site of injury (T8) extending to the lumbar region

(around 15 20um sections) was performed in ImageJ and normalized by area.

Injured Spinal Cord Western Blotting

Female Sprague-Dawley rats were given T8 contusions as detailed above. One day after injury, subjects were given subcutaneous dimethyl sulfoxide (DMSO) vehicle or ISP

(22µg/mL) injections for 14 days before sacrifice. Spinal cord were quickly extracted and frozen using liquid nitrogen and stored at -80°C before processing. To process spinal cords for western blotting, spinal cords from the thoracic to lumbar region were homogenized (D1000, Handheld Tissue Homogenizer, Benchmark Scientific) with

300µL lysis buffer (89900, Thermo Scientific) with a tablet of protease inhibitor

(11697498001, Sigma) per 10mL of lysis buffer. Lysates were agitated for 2 hours at 4°C then centrifuged at max speed for 20 minutes before the supernatant was collected. 20ug protein was loaded for western blotting as detailed above. The following primary antibodies were used: Cathepsin B (CatB, 1:500, R&D), GAPDH (1:1000, 2118S, Cell

Signaling Technologies) with the corresponding peroxidase-linked secondary antibodies detailed above.

Transgenic PTPσ dorsal root ganglia (DRG) extraction

PTPσ null mutants (Elchebly et al. 1999) and BALB/C wild type DRG extractions were performed similarly as rat DRG extractions with the exception that DRGs provided from one animal was used across 8 coverslips with 1µM FUdR (F0503, Sigma) to prevent glial outgrowth. Experiments for spot assay immunostaining and GAG-CSPG degradation were performed as described above. To assess Cathepsin B immunoreactivity, DRGs were plated on poly-L-lysine incubated coverslips pre-coated with 1µg/mL aggrecan and

10µg/mL laminin in 24-well plate for 2div. Cell fixation and staining was performed similarly as DRG-spot staining with the following primary antibodies: Tuj1 (1:500,

Sigma), Cathepsin B (CatB, 1:500, R&D). Corresponding previously described

57 secondary antibodies were used. Axon tips were imaged and quantified using ImageJ normalized by area. Precise sample sizes are listed in the figure legend.

Experimental Design and Statistics

Spot degradation assays were typically performed in replicates of up to 5 with around 20 repeats each (see methods subsection for details). Spot crossing assays were performed with at least 3 replicates and 4 repeats each (see methods subsection). Precise n is noted in figure legends. GraphPad Prism 5 (La Jolla, CA) was used to analyze data with the following statistical tests: Pearson’s chi test to test for normality, student’s unpaired t- test, one-way ANOVA with post-hoc Tukey’s (to test multiple groups) or Dunnett’s multiple comparisons (to test control against other groups) tests.

2.4 Results

Serotonergic sprouting correlates with glycosaminoglycan (GAG-CSPG) digestion in vivo

We had previously observed that after contusive spinal cord injury, serotonergic (5-HT) sprouting was dramatically enhanced following ISP treatment which, in turn, allowed for improved urinary and locomotor recovery (Lang et al. 2015). To explore the mechanism(s) responsible for ISP-enhanced fiber sprouting, we immunostained sections of spinal cords derived from animals that had received a thoracic level 8 contusive injury

(250kdyne) and were subsequently treated systemically with either vehicle (saline, 5% dimethyl sulfoxide, DMSO) or ISP (22µg/mL) for 7 weeks beginning one day after injury. Serotonergic sprouting was increased distal to the lesion especially in the vicinity of the ventral motor pools and near the central canal of ISP-treated animals compared to vehicle controls (Fig. 1A). Following immunostaining for both serotonin and the glycosaminoglycan moiety of CSPGs (GAG-CSPGs) using the wisteria floribunda agglutin (WFA) antibody, we found that vehicle-treated animals had profuse GAG-CSPG staining throughout the grey and white matter of the lumbo-sacral spinal cord compared to ISP-treated animals (Fig 1B). In comparison, WFA immunoreactivity of spinal cords belonging to animals that received no injury was distributed diffusely across the grey matter with noted immunoreactivity in perineuronal nets surrounding ventrally located motor neurons (Fig. 1A). In vehicle-treated, lesioned animals, serotonergic fibers were present although in minimal amounts. However, in ISP treated animals, enhanced serotonin immunostaining occurred in regions where WFA was lacking (Fig. 1B). Indeed,

59 the staining patterns seemed reciprocal with an almost lock-and-key configuration (Fig.

1A). This striking pattern led us to speculate whether ISP treatment enhanced the ability of fibers to sprout through the secretion of proteases which would diminish the presence of inhibitory GAG-CSPGs. To better explore this hypothesis, we harvested adult DRGs to study in vitro since it was not feasible to extract adult serotonin neurons in significant numbers. Indeed, DRG cultures plated on our aggrecan spot assay and treated with ISP developed “shadows” of CS-56 immunostaining (an antibody specific for the GAG domain of CSPGs) beneath their axons as they extended across a previously inhibitory

CSPG gradient (Fig. 1C). This led us to hypothesize that protease activity in DRGs may be induced by ISP treatment.

ISP induces DRGs to degrade CSPGs through upregulation of protease secretion

To begin investigating whether protease activity was linked downstream to PTPσ activity, we treated DRGs with vehicle control, varying doses of ISP (0.75µM-5µM), or scrambled-ISP (S-ISP, 2.5µM) and collected the conditioned media (CM) which was then incubated with freshly made aggrecan spots for two days in vitro (div) (Fig. 2A-C). These spots were then stained with CS-56 and the pixel intensities of the CSPG gradient were measured. To validate this aggrecan spot assay as a measure for CS-56 immunoreactivity, we additionally incubated spots with 0.1U/mL Chondroitinase ABC (ChABC) and found that ChABC treatment reduced CS-56 immunoreactivity as expected (Fig. 2B-C).

Interestingly, S-ISP (2.5uM) also slightly decreased CS-56 immunoreactivity.

Importantly however, ISP (2.5uM) far more dramatically decreased GAG-CSPG over S-

ISP. The slight amount of apparent protease activity stimulated via the scrambled peptide

60 may be due to the known biological activity that TAT itself (an amino acid sequence included in ISP as well as S-ISP that allows cell penetrance of our peptide) is able to induce such as enhancing neuroprotection or neuroinflammation through Erk1/2 and Akt signaling (Williams et al. 2010; P. Liu et al. 2014; Youn et al. 2015). We have found no evidence that S-ISP promotes any functional benefits after SCI (Lang et al., 2015). We did find that CS-GAG digestion was ISP dose-dependent and became significantly enhanced over vehicle control when peptide concentrations reached 2.5µM (Fig. 2C). As a control, we incubated media alone with vehicle, or the same concentrations of ISP, S-

ISP, or TAT (Fig. 2E). We found that these peptides alone (no cells present) were unable to significantly digest CSPGs (Fig. 2E). Instead, ISP-degradation of GAG-laden CSPGs increased over time and was dependent on the presence of DRGs (Fig. 2E). GAG-CSPG degradation may reflect an enhanced secretion of stored proteases or an increased production of proteases or both.

To confirm that aggrecan degradation was occurring independent of our spot assay conditions, we performed western blot analyses of CM collected from treated DRGs that were then incubated with aggrecan (20µg/mL) for two hours (Fig. 2F-H). CS-56 immunoblotting of incubated CM and aggrecan mixtures confirmed a dose-dependent

ISP-induced decrease of CS-56 immunoreactivity (Fig. 2F). Interestingly the immunoreactivity with DIPEN, an antibody specific for a neo-epitope of the aggrecan core protein revealed only once cleaved (Hughes et al. 1995), also increased in an ISP- dose dependent manner (Fig. 2G). As a further control for these assays, we incubated media alone with the same concentration of aggrecan for up to 4div and found that CS-56

61 was not decreased overtime, nor was DIPEN immunoreactivity detected (Fig. 2I). This suggested that a decrease in CS-56 and reciprocal increase in DIPEN expression patterns are ISP-dependent but only in the presence of cells.

To further characterize protease activity, we used a general protease assay (EnzChek) that measures cleaved casein, which also fluoresces once it becomes cleaved by a variety of proteases present in the CM collected from DRG cultures incubated with vehicle control,

ISP or S-ISP for 1, 2, or 4div (Fig. 3C). Just as in our spot assays, we noticed that protease activity increased over time and that ISP (2.5µM) treatment significantly increased protease activity over vehicle control by 4div. In addition, collected CM was thoroughly denatured by boiling at 100°C and incubated with guanidine hydrochloride

(5M). We found that denaturing the CM was able to fully rescue GAG-CSPGs from degradation induced by ISP incubation (Fig. 3D).

While we confirmed that GAG-CSPG degradation was enhanced by ISP through our spot assay and western blots, we sought to corroborate even further that protease(s), in addition to being secreted into the media, were also being focally induced along the regenerating axon. To visualize local GAG degradation by ISP treated DRG axons, we plated DRGs on modified aggrecan spots that were formulated with 10µg/mL laminin,

1µg/mL aggrecan, and 25µg/mL DQ Gelatin, which is a specially quenched gelatin that fluoresces once cleaved by a protease (Fig. 3A-B). Using this modified aggrecan spot, we found axons that colocalized with a trail of cleaved, fluorescing gelatin as they attempted to extend past the aggrecan gradient (an example is shown in Fig. 3A). Quantification of

62 these modified spots revealed that DRG neurons normally show a baseline of gelatin cleavage; however, treatment with ISP (2.5µM) significantly increased the amount of cleaved gelatin colocalized with axons labeled with beta-tubulin III (Tuj1) (Fig. 3B). We further incubated CM with protease inhibitors including a low concentration of the globally effective Roche protease inhibitor (0.1%) and a broad matrix metalloprotease

(MMP) inhibitor, GM6001 (25µM). The non-specific Roche protease inhibitor was able to rescue ISP-induced GAG-degradation, but ISP treatment in the presence of GM6001 still allowed for GAG-degradation (Fig. 3E). This suggests that ISP may be upregulating proteases other than the various MMPs that are inhibited by GM6001. The addition of the

Roche protease inhibitor also resulted in a decrease in axons crossing the aggrecan gradient in the presence of ISP, suggesting that proteases may, indeed, be playing a role in enhancing axon outgrowth by increasing CSPG digestion (Fig. 2,3).

ISP upregulates Cathepsin B secretion in DRG cultures

We next asked how ISP might be inducing protease activity. To begin exploring whether enhanced transcription or translation of certain proteases was responsible for ISP- induction of GAG-degradation, we treated DRGs with vehicle control or ISP (2.5µM) with translation inhibitors, Anisomycin (20µg/mL) and Cycloheximide (20µg/mL), or a transcription inhibitor, α-Amanitin (20µg/mL), and incubated spots with the collected

CM. GAG-CSPG degradation still occurred in all treatment groups once ISP was added suggesting that neither transcription nor translation were requisite for ISP-dependent

GAG-CSPG degradation (Fig. 4A). However, when Exo1 (10µg/mL), an exocytosis inhibitor (Feng et al. 2003), was added in conjunction with ISP, CS-56 immunoreactivity

63 was restored (Fig. 4B). Thus, ISP seems to be inducing exaggerated protease activity in the CM through enhancing protease secretion.

To discover which protease(s) may be hyper-secreted following PTPσ modulation, we subjected ISP-treated DRG CM to mass spectrophotometry analysis and found that

Cathepsin B (CatB) was the most abundant CSPG-degrading protease (Extended Fig. 4-

1) (Fosang et al. 1992; Mort et al. 1998). Previous work has identified Cathepsin B to be enriched primarily in neurons in the gray matter of the spinal cord and brain (Ellis et al.

2005). Western blot analysis of vehicle control, ISP or S-ISP (2.5µM) treated CM confirmed that Cathepsin B activity increased overtime (Fig. 4C) and was, indeed, found in DRG cellular lysates (Fig. 4D). Immunostaining of DRG neurons in vitro (plated on laminin/ low aggrecan) additionally revealed that Cathepsin B could be found in neuronal somata as well as throughout the axon and axonal tips (Fig. 4F). Cathepsin B is a lysosomally-derived peptidase that is important in numerous cellular homeostatic processes including protein degradation and turnover in lysosomes, autophagy, cell signaling, and innumerable other physiological functions (Turk et al. 2012). Interestingly,

Cathepsin B has been previously reported to also reveal a DIPEN-specific epitope once it cleaves aggrecan much like that which we have observed (Fig. 2G) (Mort et al. 1998).

To begin exploring how Cathepsin B may be further controlled by PTPσ modulation, we immunostained for an endogenous inhibitor of Cathepsin B, Cystatin B (CSTB) (Turk and Bode 2001). While we found trace amounts of Cystatin B in ISP-treated DRG CM

(Fig. 4C), Cystatin B seemed to be decreased by ISP intra-cellularly (Fig. 4D). Indeed,

Cathepsin B (Fig. 4F) as well as Cystatin B (Fig 4E) can be detected throughout DRG neurons including within their axons.

Cathepsin B degradation of aggrecan enhances axon outgrowth

We next sought to test whether specific Cathepsin B-degradation of CSPGs would enhance axon outgrowth past a CSPG gradient. We first tested recombinant Cathepsin B and Cystatin B using a general protease assay (EnzChek) and found that higher concentrations of recombinant Cathepsin B were, indeed, able to cleave casein, but that this activity is dampened by the addition of Cystatin B (Fig. 5A). Recombinant Cathepsin

B (0.5µg/mL) was then used to pretreat aggrecan spots much like ChABC (Fig. 5B). In addition to recombinant Cathepsin B, we also pretreated spots with CM from DRGs treated with vehicle, ISP or S-ISP (2.5µM) with or without the addition of Cystatin B

(50µg/mL) that would putatively inhibit any ISP-secreted Cathepsin B. As expected, ISP- treated CM yielded GAG-CSPG degradation significantly over media-only positive control, vehicle, or S-ISP controls (Fig. 5B). Additionally, ChABC (0.1U/mL) was able to decrease CS-56 immunoreactivity as expected. The addition of recombinant Cathepsin

B was also able to degrade GAG-CSPGs in a similar manner. However, incubating recombinant Cystatin B with ISP-treated DRG CM was able to rescue Cathepsin B- induced GAG-CSPG degradation (Fig. 5B). The degradation of GAG-CSPGs by pretreatment of the spot assay also correlated with an increase in axon outgrowth across the outer rim (Fig. 5C-D). As expected, ChABC treatment yielded significant axon crossings past the aggrecan gradient compared to media-only, vehicle CM, and S-ISP CM pretreatments. ISP-treated CM and recombinant Cathepsin B pretreatments also similarly

65 enhanced axon crossings. However, the addition of recombinant Cystatin B to recombinant Cathepsin B pretreatment attenuated this effect (Fig. 5C-D).

To further test whether Cystatin B- inhibition of Cathepsin B had any functional effects on axons crossing the CSPG gradient, we overexpressed Cystatin B (CSTBo/e) using lenti-viral particles expressing the Cystatin B gene under the EFa1 promoter in our DRG culture. Overexpression of Cystatin B was confirmed in comparison to control GFP lentiviral infected DRG cultures using western blot analysis (Fig. 6A). Indeed, overexpression of Cystatin B was able to rescue ISP-induced GAG-CSPG degradation in our spot assays

(Fig. 6B). Furthermore, overexpression of Cystatin B attenuated ISP-induced axonal outgrowth past a gradient of aggrecan in our spot assay (Fig. 6C-D). Taken together, this suggests that ISP-induced degradation of GAG-CSPG is, indeed, occurring through

Cathepsin B expression and that Cathepsin B activity is important for axon outgrowth through extracellular matrix laden with CSPGs.

As Cathepsin B is associated with lysosomes (Sloane et al. 1981), we co-labeled DRG neurons in vitro (on laminin and low aggrecan) with Lamp1, a lysosomal marker, and

Cathepsin B to better visualize Cathepsin B/lysosomal interactions (Fig. 6E). As expected, the majority of Cathepsin B immunoreactivity was co-localized with lysosomes along the axon and soma of DRG neurons. This suggests that lysosomes and Cathepsin B are, indeed, in an appropriate region of the neuron to effect GAG-CSPG degradation at the leading edge of the axon as it grows.

To explore whether Cathepsin B-induced digestion of the perineuronal net may be occurring in vivo, we co-labeled Cathepsin B with serotonergic axons in sections from

T8-contused SCI animals that received either daily vehicle or ISP (22µg/mL) subcutaneous injections for 49 days after injury, with a 24hr delay before treatment was initiated. Interestingly, serotonergic axons seemed to highly express Cathepsin B following immunostaining (Fig. 7). It is possible that the pattern of low WFA immunoreactivity and high serotonergic expression (Fig 1) may be occurring through

ISP-enhanced Cathepsin B degradation of the proteoglycan laden perineuronal net and extracellular matrix. To further correlate low WFA immunoreactivity with Cathepsin B activity as a result of ISP treatment, we processed the spinal cords of similarly contused rats that were given 14 days of ISP (22µg/mL) or vehicle subcutaneous injections.

Following western blotting, we found that ISP treatment induced greater active Cathepsin

B immunoreactivity over control (Fig. 7B). Through upregulated Cathepsin B activity, serotonin fibers may be able to better clear their inhibitory GAG-CSPG laden environment to allow for enhanced sprouting.

Linking Cathepsin B activity with PTPσ

Finally,we inquired whether Cathepsin B activity was enhanced in PTPσ null animals independent of peptide modulation. As the cognate receptor of GAG-CSPGs, the DRGs of PTPσ null animals have been extensively reported to show enhanced axonal outgrowth despite the presence of CSPGs (Shen et al. 2009) and we hypothesize that without PTPσ, these DRGs may show some elevation of protease activity. Similar to ISP-treated DRGs

(Fig. 1C), we found that PTPσ mutant DRGs plated on aggrecan spots showed digested

67 trails of CS-56 (Fig. 8A). To better quantify this effect, we returned to our CM spot assay and found that CM collected from transgenic PTPσ animals was able to decrease CS-56 immunoreactivity compared to CM made by their wild-type littermates (Fig. 8B).

Interestingly, the reduction of CS-56 immunoreactivity did not synergize with ISP treatment as ISP treated null animals yielded a similar level of CS-56 reduction as PTPσ null DRGs alone. Finally, to correlate the reduction of CS-56 to Cathepsin B, we immunostained and quantified Cathepsin B immunoreactivity in the axons of PTPσ null

DRGs (Fig. 8C-D). We found significantly increased Cathepsin B immunoreacivity compared to wild-type neurons that was found along the axons as well as their growing tips (Fig. 8 C-D). Together, our results link the enhanced activity of Cathepsin B to PTPσ as one of the many underlying mechanisms that may be contributing to enhanced axon outgrowth on CSPG-rich environments.

2.5 Discussion

We have described how Intracellular Sigma Peptide (ISP) modulation of PTPσ enhances

GAG-CSPG degradation, ultimately leading to relief of proteoglycan-mediated axon growth inhibition. Furthermore, we have identified Cathepsin B as a major enzyme that is secreted by receptor modulated DRGs and likely serotonergic axons as well. While initial

CGRP staining in our thoracic level 8 contusion model did not show obviously enhanced

DRG sprouting following ISP treatment, perhaps more sophisticated labeling techniques that can reveal the detailed anatomy of DRGs will be required to determine if ISP treatment fosters DRG sprouting in vivo. We propose that exaggerated secretion of

Cathepsin B may be occurring at the leading edge of ISP-treated axon growth cones as they negotiate their way past an inhibitory CSPG barrier. Peptide modulation of PTPσ appears to be relatively protease specific and, thus, in stark contrast to the “protease storm” that typically occurs after CNS injury (Noble et al. 2002; Haoqian Zhang et al.

2010).

The ability of growing neurons to locally secrete extracellular matrix-remodeling proteases is a well characterized phenomenon (Krystosek and Seeds 1984; Bai and Pfaff

2011). For example, peripheral axon regeneration following a sciatic crush lesion is known to be involved with protease remodeling of Schwann cell derived extracellular matrix (Siconolfi and Seeds 2001). Also, a pre-conditioning lesion of the peripheral fibers of DRGs that further improves regeneration after a delayed second lesion is, at least partially, effected by the augmented expression of proteases (Minor et al. 2009).

Regenerating olfactory axons have been noted for their ability to release proteases (Cao et al. 2012; Ould-yahoui et al. 2013), which can continue even after they are transplanted into the spinal cord (Pastrana et al. 2006). Duchossoy et al. also observed protease- remodeled scar lesions following spinal cord injury that formed pathways for ingrowing neurites (Duchossoy 2001). Aside from neurons (Krystosek and Seeds 1984; Zuo et al.

1998), many different migratory cell types utilize fine temporal-spatial regulation of proteases to remodel the surrounding extracellular matrix along their prospective pathways. Metastasizing tumor cells, in particular, enhance a program of focalized protease and Cathepsin B secretion during migration (Olson and Joyce 2015).

While we have identified Cathepsin B in DRGs as the chief protease secreted following

ISP treatment, it is possible that other proteases may be secreted or activated as well. Zuo et al., for example, have identified DRG neuron activation of MMP-2 that serves to remodel inhibitory extracellular matrix and enhance axon regeneration by degrading

CSPGs in the Bands of Bungner while sparing axon-promoting laminin (Zuo et al. 1998).

Given the relatively low densities of our DRG cultures, we were only able to clearly identify robustly secreted Cathepsin B, but cannot exclude other proteases secreted at lower levels that may act in concert to degrade inhibitory CSPGs surrounding the neuron.

Cathepsin B has been described as a “hub” enzyme that can initiate a network of protease activity including that of MMP-2 and 9 (Gondi and Rao 2013). For example, Padamsey et al. (2017) have recently described finely regulated exocytosis of Cathepsin B at the dendritic spine that, in turn, activates MMP-9 to maintain spine motility. It is possible

70 that enhanced Cathepsin B activity seen in our model may also be initiating a network of protease activity (Padamsey et al. 2017). Aside from degrading CSPGs, Cathepsin B in neurons has also been described to further digest amyloid-beta in mouse models of

Alzheimer’s Disease to functionally improve memory (D.-S. Yang, et al. 2011a) .

We have also described that peptide modulation of PTPσ results in an intracellular decrease in Cystatin B, an endogenous Cathepsin B inhibitor, which may participate to further strengthen Cathepsin B activity. Studies by the Nixon group have shown that genetic deletion of Cystatin B and subsequent enhanced Cathepsin B activity works to reverse autophagic dysfunction (D.-S. Yang et al. 2011b). As inhibition of PTPσ has been previously linked to increased autophagic flux (Martin et al. 2011), we surmise that a strong link between lysosomal regulation and autophagy may also exist, as autophagosomes must eventually fuse with lysosomes to complete protein degradation and turnover. Furthermore, we have identified lysosomal immunostaining along the DRG axon growth cone. It is possible that the bubbling vesicular structures that we first noted in abundance within early developing dystrophic growth cones in our proteoglycan gradient spot assay (Tom et al. 2004) are cycling autophagosomes. This has substantial implications for the possible roles PTPσ and CSPGs may play in a variety of traumatic but also perhaps neurodegenerative diseases. For example, CSPGs are associated with senile plaques of Alzheimer’s disease (DeWitt et al. 1993), and mouse models of

Alzheimer’s disease develop well characterized dystrophic axons around senile plaques filled with Cathepsin-laden lysosomes (S. Lee et al. 2011). Whether these lysosomes are

71 the result of similar CSPG-PTPσ interactions is an interesting question for future research.

The regulation of protease release by CSPG-PTPσ interactions may additionally have implications for synaptogenesis during embryonic development or plasticity in adulthood.

The family of LAR receptors including PTPσ have been noted for the roles they play in synapse formation during development where they may act as adhesion molecules to structurally stabilize emergent synapses (Mironova and Giger 2013; Um and J. Ko 2013).

Given what is known about PTPσ in inhibiting axon growth and encouraging synaptogenesis (Filous et al. 2014; Lang et al. 2015), it is possible that PTPσ acts as a switch to regulate these two processes. CSPGs bound to PTPσ in the context of the growth cone, for example, inhibits axon elongation (receptor on); however, blockade of the receptor (receptor off) enables growth cones to secrete extracellular matrix- remodeling proteases to enhance outgrowth. When a growth cone reaches its target and perineuronal net proteoglycans upregulate, it is conceivable that PTPσ switches “on” to immediately limit protease secretion in order to cease rapidly elongating growth and help foster engagement at the synapse. Indeed, there is a rising interest in PTPσ as a presynaptic receptor acting to inhibit axon branching and initiate synaptogenesis (Horn et al. 2012; J.S. Ko et al. 2015). Following injury in adulthood, ISP may be effectively inactivating PTPσ to allow dystrophic axons to re-enter a renewed protease-secreting growth state. Indeed, we found that PTPσ null DRGs themselves show enhanced

Cathepsin B immunoreactivity along their axons. Certainly, there is an important

72 biological need to strategically link and tightly regulate protease secretion at the growth cone depending on whether it is still elongating or engaging with its postsynaptic target.

We further report that following contusive cord injury and systemic ISP treatment, sprouting serotonergic axons show high expression of Cathepsin B within regions denuded of CSPG staining. While the isolation of ample amounts of mature serotonergic neurons for the purpose of in vitro characterization of CSPG degradation was not possible (hence, we turned our attention to mature DRGs), the remarkable reciprocal pattern of dense serotonergic axons and decreased matrix may be the result of enhanced

Cathepsin B activity as it is in sensory neurons. Recently, Bijata et al. have linked serotonergic receptor 5-HT7 to MMP-9 release that resulted in dendritic spine elongation in the presence of the perineuronal net (Bijata et al. 2017). Following many models of

CNS injury, serotonergic axons have additionally been widely reported to be especially plastic and robustly able to sprout (Hawthorne et al. 2011; Jin et al. 2016; Yuanyuan Liu et al. 2017). Whether this robust regenerative phenotype is due, at least in part, to their ability to degrade CSPGs through Cathepsin B activity will need to be further explored.

It is becoming more evident that proteases play nuanced roles in synaptogenesis, extracellular matrix remodeling, and neuronal plasticity. Our work links PTPσ modulation with protease activity. While the precise molecular connections between these processes will need to be investigated, we propose that continued study of PTPσ modulation would help to elucidate how axon regeneration may be encouraged in many models of CNS trauma or disease.

2.6 Figures

Figure 1

ISP treatment enhances glycosaminoglycan-chondroitin sulfate proteoglycans

(GAG-CSPG) degradation by neurons. A) Systemic ISP enhances GAG-CSPG amelioration following spinal cord injury. Rats received 250kdyne thoracic level 8 contusion and treated for 7 weeks with subcutaneous injections of DMSO vehicle or

22ug/mL ISP beginning one day after injury. Tissue was collected 7 weeks after last ISP treatment and stained to visualize serotonergic (5-HT) axons and GAG-CSPGs (WFA).

Non-injured spinal cord of an adult rat was immunostained with WFA to visualize normal GAG-CSPG pattern. Scale bar=500µm. B) Quantification of 5-HT (n=25, 41 images; **p<0.0015, df=64, un-paired t test) and WFA (n=39, 47 images; ***p<0.0001, df=84 un-paired t test) immunoreactivity in sections caudal to the injury site. C) Dorsal root ganglion (DRG) axons (Tuj1) leave digested shadows as they cross the high CSPG gradient (CS-56) in our aggrecan spot assays when treated with low concentrations of ISP

(1.25µM). Scale bar=50µm. Red arrows indicate regions of absent GAG-CSPGs colocalized with neuronal expression. Graph depicts: median (line), quartiles (boxes), range

(whiskers)

Figure 2

ISP promotes GAG chain degradation in coverslip-bound aggrecan/laminin spot assays. A) Conditioned media (CM) from 4 days in vitro (div) or 0.1U/mL

Chondroitinase ABC (ChABC) treated DRGs were incubated with aggrecan spots for

2div, then stained with CS-56 and analyzed. B) Representative aggrecan spot images.

Scale bar=200µm. C) ISP degrades aggrecan GAG-CSPGs in a dose-dependent manner. n=120, 73, 69, 92, 77, 59, 50; *p<0.05, **p<0.005, ***p<0.001; df=6, One-way ANOVA

Tukey’s posthoc. D) ISP-degradation of GAG chains is time-dependent and significant at

48 hours of incubation with DRGs. n=47, 31, 32, 41, 30, 39, 78; ***p<0.001; df=6, One- way ANOVA, Dunnett’s post-hoc. E) As a control, equal molar concentrations of peptide alone in media do not degrade GAG chains. Not significant (ns). n=34, 27, 28, 22; p=0.1748; df=3, One-way ANOVA Tukey’s post-hoc. F) Western blots of CM from

DRGs treated with varying concentrations of ISP incubated with 20ug/mL aggrecan confirm CS-56 spot degradation, n=4 blots G) DIPEN, a neo-epitope present once aggrecan is cleaved, increases with ISP dose, n=3 H) Western blots of CM from vehicle control, 2.5uM ISP, and 2.5uM S-ISP incubated with 20ug/mL aggrecan blotted with CS-

56 and DIPEN, n=2 I) Control western blots of media incubated for varying times show intact CS-56 and no DIPEN signal, n=3. Graphs depict: median (line), quartiles (boxes), range (whiskers).

Figure 3

ISP promotes GAG chain degradation through increasing protease activity. A)

Representative image of an ISP-treated DRG axon digesting DQ Gelatin as it crosses the

CSPG gradient in our spot assay and vehicle control DRG axon. Scale bar=50µm B)

Quantification of ratio of cleaved DQ Gelatin overlapping Tuj1-labeled axons. n=34, 32;

P=0.0011; df=15, Unpaired student t-test. C) EnzChek measures fluorescence of cleaved casein of CM from DRGs treated with CON, 2.5µM ISP or SISP for 1, 2, or 4div. n=11,

16, 16, 16, 16, 16, 19, 18, 20; ***p<0.001; df=8, One-way ANOVA Dunnett’s posthoc.

D) Aggrecan spot GAG chain-degradation is rescued following boiling CM, then incubating with guanadine hydrochloride (5M) before plating. n= 71, 49, 50, 51, 68, 50;

**p<0.005; df=5, One-way ANOVA Dunnet’s posthoc. E) Aggrecan spot degradation was rescued by 0.1% Roche general protease inhibitor cocktail, but not broad MMP inhibitor, GM6001 (25µM). n= 114, 161, 197, 97, 100, 63; **p<0.005, df=10. Not significant (ns), One-way ANOVA Tukey’s posthoc. F) Axon crossings, normalized by the total number of neurons in the spot, were decreased with protease inhibitor treatment. n=28, 10, 10, 29, 17, 20; **p<0.005; df=5, One-way ANOVA Tukey’s posthoc. Graph depicts: median (line), quartiles (boxes), range (whiskers).

Figure 4

ISP promotes secretion of Cathepsin B (CatB). A) ISP-dependent aggrecan spot degradation is not dependent on transcription or translation as shown by Anisomycin

(20ug/mL), Cycloheximide (20ug/mL), or alpha-Amanitin (20ug/mL) treatment of

DRGs. n=103, 91, 54, 50, 73, 57, 72, 60; not significant (ns), One-way ANOVA Tukey’s posthoc. B) Inhibiting exocytosis with a high concentration of Exo1 (10ug/mL) rescues

GAG chain degradation by ISP. n=115, 187, 160, 76; *p<0.05, ***p<0.001; df=3, One- way ANOVA Tukey’s posthoc. C) Western blot of Cathepsin B (CatB) and Cystatin B

(CSTB) from DRG CM collected after 2 or 4div treated with vehicle control, 2.5µM ISP or S-ISP, n=5. The same blot was stained with total protein dye, Coomassie Blue. D)

Western blot of CatB, CSTB, or GAPDH from DRG lysate treated with vehicle control,

2.5µM ISP or S-ISP for 4div, n=3. E) DRGs stained with Tuj1 and CSTB or F) CatB.

Scale bar=50µm. Graph depicts: median (line), quartiles (boxes), range (whiskers).

Extended Figure 4-1 Mass spectrometry analysis of CM taken from DRG cultures plated on uniform laminin and low aggrecan substrate treated with 2.5µM ISP treatment for

4div.

Figure 5

Pretreatment of aggrecan spots with ISP-treated conditioned media (CM) or recombinant Cathepsin B (rCatB) enhances axon crossings through the aggrecan gradient. A) EnzChek protease assay of recombinant CatB +/- recombinant Cystatin B

(rCSTB). n=6; *p<0.05; df=4, One-way ANOVA Dunnet’s post-hoc. B) Aggrecan spot degradation with CM from vehicle, 2.5µM ISP or S-ISP treated DRGs, or media only,

0.1U/mL Chondroitinase ABC (ChABC), 50ng/mL recombinant CatB, 10ng/mL recombinant CSTB alone or in combination with vehicle or 2.5µM ISP treated DRG CM. n=106, 157, 141, 136, 21, 45, 11, 63, 47; **p<0.005; df=12, One-way ANOVA Tukey’s posthoc. ISP CM vs. ChABC not significant (n.s.) C-D) Axon crossings normalized by total neurons present on aggrecan spots pretreated with CM collected from vehicle,

2.5µM ISP or SISP, or 0.1U/mL ChABC, recombinant CatB (0.5ug/mL) with or without recombinant CSTB (50ug/mL) with representative images. Scale bar=50um. Dotted lines represent CSPG border. Inset depicts axon crossing CSPG gradient. n=11, 20, 12, 12, 12,

17, 30, 29; *p<0.05, ***p<0.001; df=7, One-way ANOVA Dunnett’s posthoc. Graph depicts: median (line), quartiles (boxes), range (whiskers).

Figure 6

Overexpression of Cystatin B (CSTBo/e) decreases ISP-treated DRG axon crossings through CSPGs. A) Western blot of DRG lysates inoculated with lenti-viral particles expressing GFP or CSTBo/e constructs. B) Aggrecan degradation from CM collected from GFP control or CSTBo/e DRGs treated with vehicle or 2.5µM ISP. n=152, 63, 98,

80; ***p<0.0001, df=5, One-way ANOVA Tukey’s posthoc. C) Spot crossings of DRGs treated with lenti-viral particles expressing GFP or CSTB overexpressing constructs. n=12; *p<0.05, df=3, One-way ANOVA Tukey’s posthoc. D) Representative images of

DRG axons crossing aggrecan spots. Dotted lines represent CSPG border. Inset depicts axon crossing CSPG gradient. Scale bar=50µm. E) CatB is associated with lysosomes

(Lamp1) in DRGs. Scale bars=50µm.Graph depicts: median (line), quartiles (boxes), range (whiskers).

Figure 7

Serotonergic axons express Cathepsin B (CatB). A) Spinal cord tissue processed from

T8-contused rats that were treated with DMSO vehicle or 22ug/mL ISP for 7 weeks following injury. Following 7 weeks after the last ISP treatment, spinal cord tissue was stained to visualize serotonin (5-HT) and CatB. Scale bar=500µm. B) Western blot of injured spinal cord lysates collected from female rats given a thoracic level 8 (T8) contusion and treated with vehicle or 22ug/mL ISP for 14 days. n=2.

Figure 8

Genetic loss of PTPσ correlates with Cathepsin B (CatB) activity and CSPG degradation. A) DRGs extracted from PTPσ +/- animals were cultured on aggrecan spot stained with CS-56 and Tuj1 display digest trails. Arrows point to digested CS-56 left by axon tip. Scale bar=50um. B) Conditioned media harvested from BALB/C wild type or

PTPσ +/- or -/- animals treated with vehicle or 2.5uM ISP were incubated onto new aggrecan spots and stained with CS-56 before quantification. n=100, 56, 51, 32, 63, 64;

***p<0.0001, df=7. One-way ANOVA, Tukey’s posthoc. C-D) Representative images of

CatB in BALB/C wildtype and PTPσ -/- DRGs (Tuj1). Scale bar=20um. Quantification of CatB immunoreactivity. n=45, 30; ***p<0.0006, df=30. Unpaired t-test. Graph depicts: median (line), quartiles (boxes), range (whiskers).

III. MODULATION OF PROTEIN TYROSINE PHOSPHATASE

RECEPTOR SIGMA ENHANCES MATRIX METALLOPROTEASE-2

RELEASE IN OLIGODENDROCYTE PROGENITOR CELLS TO

RELIEVE CHONDROITIN SULFATE PROTEOGLYCAN

INHIBITION AND REMYELINATION

This chapter may also be found in full in the following publication:

Fucheng Luo*, Amanda Phuong Tran*, Li Xin, Chandrika Sanapala, Bradley T Lang,

Jerry Silver, Yan Yang. Modulation of receptor protein tyrosine phosphatase σ enhances

MMP-2 activity to promote remyelination and functional recovery in animal models of multiple sclerosis. Under review, Nature Communications. April 2018.

*Both authors contributed equally to this work

3.1 Abstract

Multiple Sclerosis (MS) is characterized by focal CNS inflammation leading to the death of oligodendrocytes (OLs) with subsequent demyelination, neuronal degeneration, and severe functional deficits. Inhibitory chondroitin sulfate proteoglycans (CSPGs) are increased in the extracellular matrix in the vicinity of MS lesions and are thought to play a critical role in myelin regeneration failure. We now show that CSPGs curtail remyelination through binding with their cognate receptor, protein tyrosine phosphatase σ

(PTPσ) on oligodendrocyte progenitor cells (OPCs). We report that inhibition of

CSPG/PTPσ signaling by systemically deliverable Intracellular Sigma Peptide (ISP), promotes OPC migration, maturation, remyelination, and functional recovery in animal models of MS. Furthermore, we report a novel downstream molecular target of PTPσ modulation in OPCs involving up-regulation of the protease MMP-2 that allows OPCs to enzymatically digest their way through CSPGs. In total, we demonstrate a critical role of

PTPσ/CSPG interactions in OPC remyelination in MS.

3.2 Introduction

Multiple Sclerosis (MS) is a chronic autoimmune-mediated demyelinating disease characterized by a dramatic loss of clusters of oligodendrocytes (OLs), demyelination, and irreversible neurologic disability(Patrikios et al. 2006). Although remyelination can occur spontaneously, it ultimately fails in regions that develop scar-like, proteoglycan-laden plaques(Franklin and ffrench-Constant 2008). The underlying mechanisms of failed oligodendrocyte progenitor cell (OPC) differentiation, maturation, and remyelination are still not well-understood. Recent studies have identified the regulatory effects of chondroitin sulfate proteoglycans (CSPGs) on OPC maturation and function(Pendleton et al. 2013; Keough et al. 2016).

CSPGs are structural extracellular matrix (ECM) molecules consisting of chains of sulfated glycosaminoglycans (GAGs) attached to a core protein. Up-regulation of CSPGs is a hallmark of the scarring process in the CNS and has been well characterized following spinal cord injury (McKeon et al. 1991; McKeon et al. 1995; Jones et al. 2003; Tran et al.

2018), stroke (Carmichael et al. 2005; Deguchi et al. 2005), and MS (Sobel and Ahmed

2001; H. Mohan et al. 2010; Lau et al. 2012). In MS, upregulated CSPGs such as aggrecan and versican have been detected within active demyelinating lesions (Sobel and Ahmed 2001; Kippert et al. 2009; A. Chang et al. 2012). While permissive laminins promote the spreading, survival, and maturation of OLs (Buttery and ffrench-Constant

1999; Colognato et al. 2005), recent studies have shown that increased concentrations of CSPGs can outcompete growth promoting ECM and curtail mouse or human

OPC/OL migration, morphological process extension, and maturation (Siebert and

Osterhout 2011; Pendleton et al. 2013). CSPG inhibition could be relieved by enzymatic degradation through Chondroitinase ABC to enhance OL maturation in vitro (Siebert et al. 2011). In vivo, although CSPGs increase temporally in Lysolecithin (LPC)-induced lesions prior to the onset of remyelination (Fuller et al. 2007; Lau et al. 2012), improved remyelination can occur following CSPG-targeting treatments such as proteoglycan synthesis inhibitors β-d-xyloside (Lau et al. 2012) or flurosamine (Keough et al. 2016).

The transmembrane protein tyrosine phosphatase-sigma (PTPσ), and related phosphatase leukocyte common antigen-related (LAR), have been identified as receptors for the inhibitory actions of CSPGs (Shen et al. 2009; D. Fisher et al. 2011). However, the role of

PTPσ in OPC/CSPG interactions and MS disease progression is not well understood. A recent study has suggested that ablating or blocking PTPσ may actually exacerbate the course of disease (Ohtake et al. 2017) while others have found that knock out of the PTPσ gene stimulates OPCs to increase outgrowth and myelination in vitro despite the presence of aggrecan (Pendleton et al. 2013). Recently, the Silver laboratory has developed a systemically delivered synthetic peptide, Intracellular Sigma Peptide (ISP), that modulates

PTPσ and relieves CSPG-mediated inhibition leading to functional recovery following SCI

(Gardner and Habecker 2013; Lang et al. 2015; H. Li et al. 2015).

Here, we asked whether ISP treatment would also promote the regeneration of myelin in the setting of demyelinating disease using two different demyelinating mouse models. Indeed, ISP allows OPCs to overcome the inhibitory effects of CSPGs to

92 promote remyelination as well as robust functional recovery. Further, we demonstrate a novel mechanism of action underlying CSPG/PTPσ signaling whereby ISP treated

OPCs are stimulated to increase protease activity, especially of MMP-2. We also document that the peptide helps to create a diminished pro-inflammatory environment.

In turn, enhanced enzyme production in the context of an altered immune response specifically degrades inhibitory CSPGs which increases OPC migration into and differentiation within demyelinated, CSPG-laden territories. These findings may have strong clinical significance to foster the development of improved CSPG-targeted therapeutic approaches to promote OL regeneration and remyelination within demyelinated lesions in diseases such as MS.

3.3 Methods

Animals

All animal care and animal procedures were approved by the Institutional Animal Care and

Use Committee of Case Western Reserve University School of Medicine. Wild-type

C57BL/6 mice were purchased from the Jackson Laboratory (Stock No. 000664) and housed at Animal Research Center of Case Western Reserve University, Mice were maintained with a 12-h light/dark cycle. Both male and female mice were included in this study.

Experimental autoimmune encephalomyelitis (EAE) model

For induction of EAE, C57BL6/J female mice at 10-week-old of age were immunized with

MOG35-55 together with complete Freund’s adjuvant emulsion (Hooke Laboratories,

MOG35-55 EAE Induction kit, EK-2110) according to the manufacturer’s instruction. Using the EAE Induction Kit resulted in 98% successful disease induction. All EAE animals were monitored daily and scored using a clinical scale from 0 to 5 (0: no abnormality; 1: limp tail; 2: limp tail and hind legs weakness; 3: limp tail and complete paralysis of hind legs; 4, hind leg and partial front leg paralysis; 5: moribund. Once EAE mice scored for 1 or 3, they were randomly recruited into treatment group and vehicle group. For the treatment group,

5-7 mice were given daily intraperitoneal injections of ISP (20µg/day), or 5% DMSO in saline, 100µl) for Vehicle group. Experiments were blinded and animals were scored daily.

For ISP Onset: ISP treatment was given at onset of sickness scored by clinical scale. For

ISP Peak: ISP treatment was given at peak of sickness scored by clinical scale.

Lysolecithin (LPC)-induced focal demyelination in mice

12-week-old C57BL/6 male mice were anaesthetized using isoflurane and a laminectomy was performed. 1.5µl of 1% LPC was infused into the dorsal column between T11 and T12 spinal cord at a rate of 0.25µl/min. The needle was removed after a delay of 5 min to minimize back flow and the lesion closed. Starting at 24h post-surgery, mice were treated daily with ISP (20 µg/day) or vehicle (5% DMSO in saline, 100µl) by subcutaneous injection near the injury site. The mice were euthanized at days 7, 14 and 21. Spinal cords were dissected for further western blot, histology, and ultrastructural analysis. Control animals received an equivalent injection of saline, and tissue was collected. For the second injection of the MMP-2 inhibitor (10 µg/1.5 µl, 444244, Calbiochem), lentiviral particles expressing shRNA targeting mouse MMP-2 (1 µl, LPP-MSH027657-LVRU6GP,

GeneCopoeia) or 0.9% saline, animals were anesthetized at 1 d (for lentiviral particles) or 4 d (for MMP-2 inhibitor) after LPC lesion and MMP-2 inhibitor or lentiviral particles was delivered to the same area using the above paradigm. Animals were allowed to recover and sacrificed at 14 days post lesion (dpl) or 18 dpl. Lesion sizes were determined by staining of serial sections with eriochrome cyanine staining.

LPC-induced demyelination in mouse cerebellar slice cultures

Cerebellar slice cultures were performed as previously described (Hui Zhang et al. 2011).

Briefly, 300um-thick cerebellar slices were cut from P8-12 mouse cerebellum using a Leica vibrating microtome (Leica, VT1000S) and cultured in medium containing 50% basal medium eagle medium, 25% Heat-inactivated horse serum, 25% Hank’s solution, 2.5%

95 glucose, 1% glutamine and penicillin-streptomycin. After 4 days in vitro (div), 0.5 mg/ml lysolecithin (LPC) was added for 17-18h to induce demyelination. Slices were then incubated with 2.5 µM ISP for 8 days. For GM6001 experiments, 25uM GM6001 (2903,

Tocris), 100nM MMP-2 inhibitor (444244, Calbiochem), and/or 2.5uM ISP or SISP was incubated for 9 days. Remyelination was examined by semi-quantitative Western blot of

MBP and immunofluorescence stain of MBP and NF200.

Immunostaining of cultured slices

Slices were fixed with 4% PFA, delipidated, and washed three times in PBS, blocked in

PBS containing 0.1% Triton X-100 and 5% normal goat serum and incubated with anti-

MBP (SMI-99P, Covance, 1:300), and anti-NF200 (N4142, Sigma, 1:250) antibodies overnight at 4°C. Slices were then washed in PBS and incubated in Alexa Fluor-conjugated secondary antibodies (1:500, invitrogen) for 2h. Slices were mounted in Vectashield mounting medium with DAPI (Vector Laboratories) and analyzed using Leica DFC500 fluorescence microscope.

Purified mouse Oligodendrocyte Progenitor Cell (OPC) cultures

OPCs were prepared from newborn C57BL/6 mice as described previously(Luo et al.

2016). Cell culture plates were pre-coated with IgM (10 µg/ml, Millipore) in 50mM Tris-

HCl and followed primary mouse antibody A2B5. Dissociated cells were incubated in the pre-coated culture dishes for 30 min at 37°C and then non-adherent cells were gently removed. A2B5+ OPCs were released by 0.05% trypsin in DMEM at a purity of ~96%.

Purified OPC cells were expanded in DMEM/F12 medium supplemented with N2, 20ng/ml

PDGF, 20ng/mlFGF, 5ng/mlNT-3, 10ng/ml CNTF, Glutamine (200 mM).

Conditioned media (CM) protease activity assay

Around 1x106 OPCs were plated per well on PLL, 1ug/mL laminin, and 2ug/mL aggrecan coated 6-well plates and treated with vehicle, 2.5uM ISP, or SISP for 2div at 37°C. CM was collected, cell-strained, concentrated (Milipore Ultracel YM-3), and placed on ice until assayed. ThermoFisher Protease Assay Kit (E66383, EnzChek, ThermoFisher) was used to assay protease activity. 1x of the EnzChek mixture was mixed 1:1 with the CM from each group and incubated at room temperature overnight with gentle shaking. 3 replicates were performed for each sample. Samples were analyzed using a spectrophotometer at

502/528nm excitation and emission to assess cleaved and fluorescing casein. Fluorescence units reported have been blanked with EnzChek and cell culture media mixture.

CSPG gradient crossing assay and gradient quantification

CSPG gradients were prepared as described previously (Tom et al.). 24-well glass coverslip were coated with poly-L-lysine and nitrocellulose, and a mixture of 700ug/mL aggrecan (A1960, Sigma) and 10ug/mL laminin (11243217001, Sigma) spotted on the coated coverslip. After drying, coated coverslips were then incubated with laminin at 37°C for 3h. Purified OPCs were plated at a density of 10,000/coverslip and cultured in

DMEM/F12 medium containing with N2, PDGFRα (20ng/ml), FGF (20ng/ml), NT-3

(5ng/ml), CNTF (10ng/ml), Glutamine (200 mM). Coverslips were stained with CS56

(C8035, Sigma, 1:500) and O4 antibodies (Hybridoma Core Cleveland Clinic, 1:10). O4-

97 positive cells crossing the aggrecan border were counted for each spot. For CSPG gradient quantification, 1x106 OPCs were plated per well on PLL, 1ug/mL laminin, and 2ug/mL aggrecan coated 6-well plates. OPCs treated with vehicle, 25uM GM6001 (2983, Tocris),

100nM MMP-2 inhibitor (444244, Calbiochem), 2.5uM ISP, or 2.5uM SISP for 2div at

37°C. CM was harvested and incubated with freshly made spots. Spots were incubated with

CM for 2div at 37°C then stained with CS56 and laminin (L9393, Sigma, 1:1000) antibodies and consistently imaged. Using ImageJ software (NIH), pixel intensities of the

CS56 or laminin spot rims were quantified using the same region of interest saved in

ImageJ.

For cultured spinal cord explants cultured on spots, the spinal cords of cervical and thoracic of P1 mouse pups were chopped into 1-2 mm tissue pieces and transferred to the coverslip.

Explants were cultured in DMEM/F12 medium with 15% FBS (Hyclone), 10ng/ml PDGF

(Sigma) and N2 supplement (Invitrogen). For ISP treatment, 2.5uM ISP was added to the media at the time of plating. 25uM GM6001 (2983, Tocris), 100nM MMP-2 inhibitor

(444244, Calbiochem), and/or 2.5uM ISP or SISP was used.

Peptide Sequences

Peptides were purchased from GenScript or CS-Bio in 1mg lypholyzed quantities (>98% purity) that were diluted to 2.5mM in dH20 and aliquoted at -20°C until ready for use as previously described(Lang et al. 2015).

Intracellular Sigma Peptide (ISP):

GRKKRRQRRRCDMAEHMERLKANDSLKLSQEYESI

Scrambled ISP (SISP):

GRKKRRQRRRCIREDDSLMLYALAQEKKESNMHES

Western blot analysis of aggrecan and laminin

CM was harvested from 1x106 OPCs incubated on PLL, 1ug/mL laminin, and 2ug/mL aggrecan coated 6-well plates. OPCs were treated with vehicle, 2.5uM ISP, or 2.5uM SISP in conjunction with 10ug/mL Exo1 (ab120292, Abcam), 25uM GM6001 (2983, Tocris), or

100nM MMP-2 inhibitor (444244, Calbiochem) for 2 days at 37°C. 100uL CM of each cell-strained group was incubated with 20ug/mL aggrecan and/or 10ug/mL laminin for 2 hours in 1mL Eppendorf tubes at 37°C. As a positive control, OPC media was incubated with aggrecan and incubated in the same fashion. Western blots were then performed as described below with incubation against CS56 and/or laminin antibodies.

Western blot analysis of OPC lysate or CM

To assess MMP-2 or 10 in OPC CM or lysates, OPCs were incubated on precoated PLL, aggrecan, and laminin as described above. OPCs were treated with vehicle, 2.5uM ISP or

SISP for 2 days at 37C. CM was then harvested and concentrated using Milipore Ultracel

YM-C centrifugal filter units at max speed (Eppendorf Centrifuge 5415D) for 30 minutes at 4°C. 50ug of concentrated CM was loaded into each lane. OPC lysates were assessed using western blot techniques described below with 20ug protein loaded from each group.

Western blot analysis

Tissue samples or cerebellar slices or purified OPC cells were homogenized with RIPA lysis buffer and protein concentration was determined by Pierce BCA protein assay kit according to the manufacture instruction (Thermo Fisher). Then, equal amounts of protein were loaded to 15% SDS-PAGE gels, and electrophoretically transferred to PVDF membranes (Millipore). The membranes were blocked in 0.1% TPBS buffer with 5% non- fat milk for 1h at room temperature and probed with indicated primary antibodies overnight at 4°C and followed by secondary antibody conjugated to horseradish peroxidase (HRP).

The following primary antibodies were used: MBP (SMI-99P, Covance, 1:1000), CS56

(C8035, Sigma, 1:1000), Laminin (L9393, Sigma, 1:1000), GAPDH (AF5718, R and D

Systems, 1:1000), MMP-2 (AF1488, R and D Systems, 1:1000), MMP-10 (MAB910, R and D Systems, 1:1000), and β-actin (sc-47778, Santa Cruz, 1:1000). Enhanced chemiluminescence was performed with a West Pico Kit (Thermo Fisher) and detected by

FluorChem E system (ProteinSimple, USA). The density of bands was quantified using

ImageJ software (NIH).

Gelatin Zymography

CM was collected from 1x106 OPCs incubated on PLL, 1ug/mL laminin, and 2ug/mL aggrecan coated 6-well plates. OPCs were treated with vehicle, 2.5uM ISP or SISP for 2 days at 37°C. CM was concentrated using Milipore Ultracel YM-3 centrifugal filter units as previously described. 40ug undenatured protein from concentrated CM or 25ng of activated recombinant MMP-2 (PF023, Milipore) and MMP-9 (PF140, Milipore) was loaded with 1x laemmli buffer (1610747, Bio-Rad) onto 10% gelatin zymograms (1611167, Bio-Rad) and ran at 100mV for 1.5 hours on ice. Zymograms were then gently shaken with 1x renaturing

100 buffer (1610765, Bio-Rad) for 30 minutes at RT then incubated with 1x developing buffer

(1610766, Bio-Rad) O/N at 37C. Developed Zymograms were then gently washed with dH20 and incubated O/N with 0.1% Coomassie Blue dye (27816, Sigma). Following washes with destaining buffer (40% MeOH, 10% Acetic Acid), Zymograms were imaged and the amount of gelatin degradation was assessed using ImageJ in a method described in the western blot section.

Protease Array Screen

OPCs were cultured on 6-well plates for 4div with vehicle control or 2.5uM ISP treatment.

Conditioned media was collected from each group and cell-strained before incubation with blot array provided in R & D Systems Protease Array Kit (ARY025). Instructions from kit were followed with overnight incubation of collected media at 4°C. Control and ISP blots were developed together with the same exposure time and pixel intensities of array was assess using ImageJ (NIH).

Luxol Fast Blue (LFB) myelin staining and quantification

LFB staining was performed according to the manufacturer’s instruction (#26681, Electron

Microscopy Sciences). For spinal cord sections, 20 µm coronal sections were incubated in

LFB solution in 56°C overnight and then rinsed sequentially with 95% alcohol and distilled water. The sections were then in 0.1% lithium carbonate solution and dehydrated with a series of graduated ethanol, cleared with Histoclear, and mounted. A set of serial matched sections were imaged and analyzed. Images (5 to 6 sections/animal) were captured under

101 light microscope. The demyelinated areas (lack of LFB staining) were quantified using

ImageJ software. For EAE sections, demyelinated areas were measured and represented as a percentage of total area of spinal cord. For sections of LPC model, lesion volumes were calculated by the lesion area from serial sections throughout the entire lesion based on the equation for volume of cylinder (V=lesion area x length of lesion).

Immunocytochemistry

For MMP-2/O4 staining, OPCs were plated onto 24-well coverslips that were precoated with PLL, 1ug/mL laminin, and 2ug/mL aggrecan and incubated with vehicle or 2.5uM ISP for 2 days at 37°C. Cultured OPC or OL cells were fixed in 4% PFA and followed blocking in PBST solution (10% normal goat serum and 0.2% Triton–X100 in PBS). Diluted primary antibodies were incubated with samples overnight at 4°C and followed by appropriate secondary antibody goat anti-mouse or anti-rabbit IgM or IgG conjugated with

Alexa Fluor 488 or 594 (1:500, Invitrogen). The following primary antibodies were used:

PDGFRα (ab65258, Abcam, 1:250), O4 (Hybridoma Core, Cleveland Clinic), MBP (SMI-

99P, Covance, 1:300), MMP-2 (Ab19167, Millipore, 1:200), and CS56 (C8035, Sigma,

1:250). Cells were mounted with Vecta Shield mounting medium with DAPI (Vector

Laboratories).

TUNEL, Ki67 immunocytochemistry and quantification

To assess proliferation, OPCs were plated on 24-well coverslips that were precoated with

PLL, 1ug/mL laminin, and 2ug/mL aggrecan. OPCs were treated with vehicle, 100nM

102

MMP-2 inhibitor (444244, Calbiochem), and/or 2.5uM ISP or SISP immediately upon plating for 2 days at 37°C. Coverslips were fixed and stained using the same method described in the immunocytochemistry section with Ki67 (550609, BD Pharmingen, 1:500) and O4 (1:10). Coverslips were imaged and counted. To assess apoptosis, OPCs were cultured in the same fashion and incubated with vehicle or 1ug/mL LPC for 2 hours at 2 days following vehicle, 2.5uM ISP or SISP incubation. Coverslips were then fixed with

PFA and methanol, then stained using APO-BrdU TUNEL assay kit (A23210,

ThermoFisher). We followed the manufacturer’s guidelines with an overnight incubation of the primary antibody at room temperature. Coverslips were costained with DAPI (D9542,

Sigma, 1:10,000) then imaged and counted.

Immunohistochemistry

Mice were anesthetized with avertin and perfused with PBS and 4% paraformaldehyde

(PFA). Spinal cords were dissected and post-fixed in 4% PFA overnight at 4°C and equilibrated in 20% sucrose. 20um-thick sections were pretreated with Reveal Decloaker

Solution (RV1000M, Biocare Medical) for antigen retrieval according to the manufacturer’s instructions. After blocking, sections were incubated with primary antibodies overnight at 4°C and followed by appropriate secondary antibodies conjugated with Alexa fluorescence 488 or 594. The following primary antibodies were used: MBP

(SMI-99P, Covance, 1:300), NeuF200 (N4142, Sigma, 1:250), CS56 (C8035, Sigma,

1:250), MMP-2 (AB191677, Sigma, 1:500), laminin (L9393, Sigma, 1:500), CAT301

(MAB5284, Millipore, 1:250), versican (AB1032, Millipore, 1:250), Iba1 (019-19741,

103

WAKO, 1:250), GFAP (MAB360, Millipore, 1:250), Olig2 (AB9610, Millipore, 1:250),

CC1 (OP80, Millipore, 1:250), iNOS (#610329, BD Biosciences, 1:250), Arginase-1 (sc-

18355, Santa Cruz, 1:50). For each staining, at least 3 individual animals/group were examined and images were captured with a Leica DFC500 fluorescence microscope.

Staining was quantified using Image J software (US National Institutes of Health, USA).

Fluorescence intensity was calculated as percentages of the mean value of the naïve controls.

Tissue preparation for electron microscopy (EM) analysis

For ultrastructural analyses of myelination, anesthetized animals were perfused with 2% glutaraldehyde/ 4% paraformaldehyde in 0.1 M sodium carcodylate buffer, pH 7.4

(Electron Microscopy Sciences). Lesioned areas of the LPC or EAE-induced spinal cords from ISP-treated or control animals were dissected and post-fixed in 1% OsO4 for 2hs.

Coronal sections (500 µm) of spinal cord were prepared (Leica, Vibratome), dehydrated, stained with saturated uranyl acetate and embedded in a Poly/Bed812 resin (Polysciences

Inc.). 1 µm-thick sections were cut and stained with toluidine blue, and matched areas were selected for EM analysis. For ultrastructure analysis, ultrathin sections (0.1 µm) were cut and visualized using an electron microscope (JEOL100CX) at 80kV. G-ratios were calculated from 50-100 randomly selected myelinated axons by measuring the myelin thickness of the inner and outer diameter of the myelin sheath.

shRNA knock down of MMP-2

104

MMP-2 knock down was mediated through lentiviral particles expressing shRNA targeting mouse MMP2 driven by the U6 promoter (LPP-MSH027657-LVRU6GP, GeneCopoeia).

Lentviral particles for shRNA scrambled constructs (LPP-CSHCTR001-LVRU6GP-100-C,

GeneCopoeia) were used as the corresponding controls. OPC cultures were infected for at least 48 hours before experiments at the multiplicity of infection (MOI) of 1. Constructs were validated using western blots analysis of infected OPC cultures for MMP-2

(AB191677, Sigma, 1:500).

Statistical analysis

All data analyses were performed using GraphPad Prism 6.00. Data are shown as mean±SEM. p<0.05 is deemed statistically significant. Statistical analysis was performed by two-tailed unpaired Student’s t tests, one-way or one-way ANOVA with post-hoc analysis by Tukey’s multiple comparison test, Dunnett’s multiple comparison test, or

Sidak’s multiple comparison test. Quantifications were performed in a blinded fashion. No statistical tests were used to predetermine sample sizes, but our sample sizes are similar to those generally employed in the field. Data distribution was assumed to be normal, but this was not formally tested. All experiments were performed at least three times independently.

105

3.4 Results

Increased CSPGs and receptor PTPσ expression in lesions of EAE and LPC demyelinating MS mouse models

We characterized CSPG expression in demyelinating lesions of MOG35-55-induced chronic progressive EAE and LPC-induced acute focal demyelination. Demyelinated

EAE and LPC lesions in the white matter of the spinal cord were visualized with

Luxol Fast Blue (LFB) myelin staining (Sup. Fig.1). As expected, LFB staining decreased in the lesions of both models. Immunostaining of sections of spinal cord tissue revealed upregulated CSPG expression in demyelinating lesions of EAE- and

LPC-afflicted animals compared to vehicle controls (Sup. Fig. 1). Furthermore, CSPG upregulation progressively increased in the EAE-lesioned spinal cord from 28 to 41 days after immunization (Sup. Fig. 1A). Tissue sections collected from animals at 7 and 14 days post-LPC injection in the dorsal spinal cord (Sup. Fig. 1C,D) similarly showed increased production of CSPGs in demyelinating lesions. Increased CSPGs in focally demyelinated areas led us to hypothesize that CSPGs negatively influence OPCs in lesion sites through PTPσ signaling, which may ultimately affect their ability to remyelinate the cord.

CSPGs are known to signal through the receptors PTPσ, LAR(Dyck et al. 2015), and the

Nogo receptors 1 and 3(Kaplan et al. 2015). We used a previously published RNA- sequencing transcriptome database of mouse cerebral cortex to search for the gene expressions of PTPRS (PTPσ), PTPRF (LAR), PTPRD (PTPδ), and RTN4R (Nogo

106 receptors) during OPC development(Ye Zhang et al. 2014). PTPRS gene transcripts

(FPKM) were the most abundant type of CSPG receptors in developing OPCs (Sup. Fig.

2B). Immunostained OPCs/OLs cells derived from wild type mouse pup brains

(postnatal day 1-2) revealed that PTPσ was expressed in the somata and processes of immature Olig2+ and mature CC1+ or myelin basic protein (MBP)+ cells (Sup. Fig.

2A,C). Western blot analyses also indicated an upregulation of PTPσ in the lesioned spinal cord of EAE- or LPC-induced demyelinating models at day 28 (EAE) and day 7

(LPC) post injections (Sup. Fig. 2D). In EAE-induced animals, double- immunostaining was also performed with antibodies against PTPσ and the OPC marker, Olig2, to reveal increased PTPσ co-labeled with Olig2+ OPCs in demyelinating lesions (Sup. Fig. 2E). These findings suggest that PTPσ is expressed and upregulated in cells of oligodendrocyte lineages following EAE or LPC induced disease, and that this receptor presents a tractable target to study the effects of CSPGs and/or receptor manipulations in MS models.

Modulation of PTPσ with ISP promotes functional recovery and remyelination in an EAE animal model

We next tested Intracellular Sigma Peptide (ISP) in an EAE mouse model, which recapitulates chronic progressive demyelination disease processes. Following MOG35-55 immunization, animals received intraperitoneal ISP injections (20µg/mouse, daily) for 41 days at the beginning (EAE ISP Onset) or the peak of sickness (EAE ISP Peak) determined by clinical scoring (Fig. 1A). The control group was injected with 5% DMSO vehicle in parallel. Functional recovery was initially observed in the Onset group after

107

~10-12 days of ISP administration (ie. day 23 post immunization). ISP improved clinical scores from 3.5–4 (severe paralysis) to 2-1.5 (limp tail and hind limb weakness). After 20-22 days of ISP treatment (~33 days post immunization), several animals in the Onset group recovered with clinical scores improving to 0.5-1 (limp tail) (Fig. 1B,C, movie). In contrast, control animals remained severely paralyzed with scores remaining around 3.5-4. EAE ISP Peak animals also improved significantly with ISP treatment; however, ISP given at the onset of disease allowed for better hind limb recovery (Fig. 1B,C). These improvements were also closely correlated with histological improvements. Lesion sizes were especially reduced in Onset treated animals as indicated by LFB myelin staining (Fig. 1 D,E). Conversely, MBP immunostaining was denser in animals treated with ISP for 41 days compared with control EAE animals (Fig. 1F). Western blotting of MBP protein isoforms also showed restoration of MBP expression in ISP-treated mice (Fig. 1G). Ultrastructural analyses revealed increased myelinated/remyelinated axons in the EAE-lesioned spinal cord in the

ISP-treated mice compared to controls (Fig. 1H). Quantitative analysis confirmed an increase of myelinated/remyelinated axons in the ISP-treated group (Fig. 1I) and the G- ratio, which indicates myelination thickness by normalizing the diameter of myelination by axon diameter, was lower in the ISP-treated group compared to the vehicle-treated group (Fig. 1J). Importantly, our results suggest that ISP acted to enhance myelin regeneration rather than prevent demyelination especially since demyelination baselines

(LFB) at 18 days following EAE induction were not significantly different between the two groups at this early stage of disease progression (Sup. Fig. 4E,F).

108

ISP treatment is associated with decreased CSPG expression in demyelinated lesions over time as well as an altered inflammatory response in EAE

In addition to observations that CSPGs markedly decreased following ISP treatment of EAE-induced animals (Fig. 4A, Cat301) we also found altered macrophage dynamics in the same animals. To begin, macrophages (Iba1) appeared to colocalize with aggrecan (Cat301) especially in the white matter (Fig 4A), which may be due to activated macrophages depositing or, more likely, phagocytosing aggrecan, which they are not known to produce (Martinez et al. 2015). We additionally observed decreased Iba1 immunostaining as well as decreased amounts of aggrecan within

Iba1+ macrophages in ISP-treated EAE animals compared to controls (Fig. 4A). We performed further quantification of Iba1 and GFAP at 41 days following EAE induction and saw significant decreases in both microglia/macrophages and reactive astrocytes respectively following ISP treatment (Sup. Fig. 4A, B). A recent study by

Dyck et al. (Dyck et al. 2018) also reported that modulation of LAR family receptor phosphatases with synthetic peptides, including ISP, skewed microglia/macrophages towards an M2 polarization following spinal cord injury. Indeed, immunostaining for markers identifying M1 (iNOS) and M2 (Arginase-1) macrophage polarization revealed supporting evidence that ISP treatment modulates the inflammatory environment in EAE animals (Sup. Fig. 4C,D). Of note, while microglia/macrophages seem to produce PTPσ after injury, reactive astrocytes do not

(Lang et al. 2015; Dyck et al. 2018).

ISP enhances the rate of myelin repair in LPC-induced demyelination

109

We next asked whether ISP modulation of PTPσ-CSPG interactions has similar effects in acute focal demyelination induced by LPC injected into the dorsal column white matter of young adult C57BL6/J mice treated either with ISP (20µg/day, subcutaneous) or control vehicle starting at 1-day post LPC injection. After ISP treatment, LPC-induced lesion volumes were significantly reduced at 14 and 21-days post lesion (dpl) compared with vehicle-treated animals as shown by LFB myelin staining (Fig. 2A,B). As indicated in Figure 2, vehicle-treated animals had an average lesion volume of 1.508±0.069 mm3,

1.035±0.06mm3 and 0.738±0.027mm3 after 7, 14 or 21dpl, respectively. In contrast, ISP- treated animals showed reduced lesion volume from an average of 1.535±0.058mm3 at

7dpl to 0.613±0.043mm3 at 14dpl (Fig. 2B). By 21dpl, we found extensive lesion repair and reduced lesion volumes in ISP-treated mice (1.535±0.058mm3 to 0.2±0.041mm3)

(Fig. 2B). Immunostaining consistently indicated increased MBP expression in LPC- lesions of ISP-treated mice compared with vehicle-treated mice after 14 and 21dpl (Fig.

2C). Quantitative western blot analysis confirmed increased expression of MBP in LPC- lesioned animals after ISP treatment at 14dpl (Fig. 2D). Finally, ultrastructural analysis confirmed the number of remyelinated axons in ISP-treated mice at 14dpl compared to controls (Fig. 2E,F). Consistent with these results, quantitative analyses of the G-ratio between ISP- and vehicle-treated mice revealed increased thickness of myelin sheaths in

ISP-treated mice at 14dpl (Fig. 2G). These experiments showed that ISP treatment accelerated the rate of myelin repair in vivo.

To better visualize the ability of ISP to impact OPCs and subsequent remyelination, we utilized a well-established ex vivo model of myelinating mouse cerebellar slice

110 culture derived from postnatal day 8-10 pups treated with 0.1% LPC for 17-18 hours to induce demyelination(Hui Zhang et al. 2011; Luo et al. 2014). We found that naïve slice cultures developed abundant myelinated axons as shown by MBP and neurofilament (NF200) colocalization (Fig. 3A, Con). LPC treatment, however, caused profuse demyelination with the production of punctate and disorganized myelin

(Fig. 3A, 1 day in vitro (div)). After 8div, the demyelinated phenotype was still prominent and remyelination was delayed in treated LPC slices compared with vehicle

(Fig. 3A, LPC+Veh, 8div). In contrast, LPC-demyelinated slices treated with ISP for 8div showed increased remyelination (Fig. 3A, LPC+ISP, 8div) compared to vehicle- treatment. Although increased MBP expression was seen in vehicle-treated slices by

14div after LPC treatment, the expression of MBP was still disorganized and failed to colocalize well with axons (Fig. 3A, LPC+Veh, 14div). However, ISP treatment for

14div resulted in abundant remyelinated axons (Fig. 3A,B LPC+ISP, 14div), which was confirmed with western blot analysis and quantification (Fig. 3C). These slice culture experiments confirm that following LPC treatment, ISP enhances the rate of remyelination perhaps by influencing OPCs directly or possibly microglia/macrophages as well.

Interestingly, this ISP enhanced rate of remyelination was correlated with a decrease in aggrecan presence (Cat301) by 8div in our cerebellar slice cultures (Supp. Fig. 3), although aggrecan expression was similar between ISP and vehicle groups at 4div (Supp.

Fig. 3). By 14div, we observed simultaneous remyelination through MBP staining and

111 decreased CSPG expression in both groups although ISP-treated slices had enhanced

MBP expression and greater CSPG decrease.

To further investigate whether CSPGs are affected by ISP treatment, we double immunostained for MBP and CSPGs in LPC-injected animals. Control LPC-lesioned animals showed upregulated levels of GAG-CSPGs (CS56) and aggrecan (Cat301) that was inversely correlated with decreased expression of MBP at 14dpl (Fig. 4B). In contrast, ISP-treated animals showed quicker reduction of CSPGs (CS56 and Cat301) and enhanced myelin expression compared to controls (Fig. 4B,D). It is important to note that while myelin repair does normally occur in LPC-lesioned animals, CSPG degradation is much slower than that which occurs after peptide treatment (Supp.

Fig. 3A,B). Thus, we found that ISP treatment not only enhanced the rate of myelin repair, but was associated with more rapidly decreased CSPG expression over time.

In LPC-injected animals, versican immunostaining was strongest in the penumbra of the lesion where reactive astrocytes (GFAP+) were found (Fig 4C). This pattern of

CSPG deposition confirms recently reported findings by Keough et al. (Keough et al.

2016) who also reported increased versican secretion by reactive astrocytes. We also examined inflammatory cells and astrocytes in locally demyelinated LPC lesions and altered the timing of ISP treatment to begin our investigation of the mechanism(s) underlying decreased CSPG expression following ISP treatment. Immediately (rather than a 1 day delay) upon LPC injection, mice received ISP (20ug day/mouse, subcutaneous) for 7 days. Staining of the lesion epicenter revealed no change in the

112 amount of activated microglia (Iba1), reactive astrocytes (GFAP), or MBP myelin protein expression between ISP-treated animals and control groups at this early stage (Fig.

4C,D). This suggests that the extent of LPC-induced injury was initially similar between

ISP and control groups. Again, aggrecan staining was colocalized with Iba1+ macrophages (Fig. 4C). However, CSPG expression (CS56, Cat301) was significantly reduced after ISP treatment, suggesting that ISP may be involved in the enhanced degradation of CSPGs in demyelinating lesions.

CSPG reduction by ISP enhances OPC survival, differentiation, and migration

ISP-induced CSPG disinhibition could result in enhanced remyelination by regulating

OPC proliferation, survival, differentiation, or migration. To distinguish between these possibilities, we began with quantification of proliferating OPCs by immunostaining with

Olig2 and Ki67 antibodies in the LPC lesion at 7dpl. We found that the percentage of

Olig2+ OPCs was significantly increased in the lesions of ISP-treated mice compared to vehicle-treated mice (Sup. Fig. 5A,B), but found no differences in Olig2+/Ki67+ cells between the groups (Sup. Fig. 5B). To further examine the effects on OPC proliferation by ISP, we peptide treated cultured OPCs for 2div on a low concentration of laminin

(1µg/mL) and aggrecan (2µg/mL) and found no significant differences in proliferation rates between treated and control groups (Sup. Fig. 5C,D). To investigate whether apoptosis of OPCs was affected by CSPGs, we performed TUNEL staining of OPCs also cultured on low concentrations of laminin and aggrecan and found that ISP treatment decreased the percentage of apoptotic cells (Sup. Fig. 5E,F). ISP also decreased OPC

113 death when similarly cultured OPCs were challenged with LPC (1µg/mL, 2 hours) (Sup.

Fig. 5E.F).

To examine the effects of ISP treatment on OPC differentiation, we quantified the number of differentiated CC1+ oligodendrocytes in both EAE animals treated with ISP

(41dpl) and in LPC-injected animals (14dpl). The percentage of CC1+ oligodendrocytes was significantly enhanced in lesions of ISP-treated mice compared to controls (Sup Fig.

6A-D). ISP-enhanced OPC maturation was also confirmed in vitro with immunopurified

OPCs (P1-2 WT mice) cultured on aggrecan and laminin pre-coated coverslips.

Immunostaining of early OPCs (O4+) and mature OLs (MBP+) showed that CSPGs reduced the progressive maturation of OPCs as seen through reduced process lengths of

O4- and MBP-expressing cells grown on CSPGs compared to OPCs grown on non-

CSPG control substrates (Sup. Fig. 6E,F). Process outgrowth and maturation of

OPCs grown on CSPGs were largely rescued with ISP treatment (quantified by analyzing MBP+ footprints of cells grown on CSPG with or without ISP treatment)

(Sup. Fig. 6E,G). These findings indicate that ISP may be enhancing survival and differentiation, instead of proliferation, of OPCs in demyelinated lesions.

ISP may also be promoting the migration of OPCs into the lesion site where they can survive and subsequently differentiate into their myelinating forms. To explore

CSPG/receptor effects on OPC migration, we utilized spinal cord explants derived from P2 WT pups grown on our CSPG gradient spot assay that has been previously used as a potent in vitro model of the inhibitory gradient distribution of CSPGs found

114 in glial scars(Lang et al. 2015; Tom et al.). ISP-treated early (PDGFRα+) and pre-mature

(O4+) OPCs derived from the explant were able to cross the CSPG-enriched outer-rim of the gradient spot. In control explants, few cells were able to migrate across this inhibitory territory (Fig. 5A-C). Thus, in addition to relieving CSPG-related apoptosis and maturational defects, ISP may also be promoting the migration of OPCs into the lesion site where they can survive and subsequently differentiate into their mature myelinating forms. These observations taken together with the reduction of CSPGs in both slice culture (Sup. Fig.3) and in vivo models of MS (Fig. 4) after ISP treatment lead us to hypothesize that targeting PTPσ through ISP induces increased secretion or activation of endogenous proteases.

ISP treatment enhances protease-dependent enzymatic digestion of CSPGs

In addition to observations of reduced CSPG expression in ISP-treated ex vivo and in vivo demyelination models, we noticed that ISP-treated PDGFRα+ OPCs left “shadows” of possibly digested GAG-CSPG areas where they infiltrated the aggrecan rim of our spot assays (Fig. 5D, arrows). The entire outer proteoglycan rim was also reduced in diameter in the presence of ISP (Fig. 5D, compare rim widths). To begin investigating whether protease activity was occurring, we returned to our spot assay to better characterize putative aggrecan degradation. Conditioned media (CM) was collected from immunopurified OPCs treated with vehicle, ISP, or SISP (scrambled ISP) and plated onto freshly made spots. ISP-treated OPC CM significantly reduced CS56 expression compared to vehicle or scrambled peptide controls as well as a no cell control (Sup. Fig.

8A-C). Interestingly, the laminin portion of the spot was completely spared as visualized

115

by immunostaining (Sup. Fig. 8D,E). We also confirmed these results with western blot

analyses of OPC CM incubated with aggrecan (20µg/mL) and laminin (10µg/mL)

collected from OPCs treated with vehicle, ISP, or SISP present (Fig. 5E-G).

To independently characterize ISP induction of OPC protease activity, we performed a general enzyme activity assay (EnzChek Kit) based on quenched casein fluorescence and found that ISP treatment of OPCs, indeed, increased protease activity (fluorescence A.U.) compared to vehicle and SISP controls (Fig. 5H,I). Furthermore, ISP increased aggrecan digestion in a dose-dependent manner as visualized through western blot analysis of ISP treatment of OPC CM incubated with aggrecan (Sup. Fig. 8G,H).

ISP increases protease to enhance OPC migration and remyelination

To begin identifying which critical proteases ISP may be regulating, we incubated

vehicle or ISP-treated OPC CM with a protease array blot. We found an increased signal

for various enzymes belonging to several classes of proteases (eg. ADAMTS, Kallikreins,

Cathepsins, MMPs) in the ISP-treatment group that are potentially (if produced in

sufficient amounts) capable of digesting CSPGs (Sup. Fig. 7). Interestingly, three

laminin degrading proteases such as Cathepsins L and V and MMP-10 were reduced,

suggesting some level of specificity in the regulation of the enzyme cascade that is linked

with PTPσ modulation. This result may help explain why we have observed unchanged

laminin expression in our ISP-treated in vitro assays (Fig. 5E,G, Sup. Fig. 8D,E). To

confirm the results from our protease array, we performed western blot analyses of

multiple upregulated proteases including MMP-2, 9, and Cathepsin B in ISP or control

116 treated OPC CM and found that MMP-2 was readily detectable and clearly enhanced after ISP treatment (Fig. 6C,D). Staining of cultured OPCs showed that MMP-2 is expressed in O4+ OPCs within their processes and appears to intensify after peptide treatment (Fig. 6H).

To confirm that OPC CM-derived MMP-2 activity is enhanced by ISP treatment, we performed gelatin zymography and found that MMP-2 gelatin-degrading activity was significantly increased upon ISP treatment over controls (Fig. 6A,B). MMP-9 activity was barely visible by gelatin zymography (Fig. 6A), and undetected by western blot analysis (data not shown). We also blotted for MMP-10, a protease that degrades fibronectin, laminin, and elastin, and found that it is secreted in far lower amounts than

MMP-2 (Fig. 6C). Vehicle, ISP, and SISP-treated OPC cultures seem to secrete MMP-10 in equally low amounts (Fig 6C,E), suggesting that enhanced MMP-2 secretion by ISP may be specific to PTPσ modulation. In addition to ISP-treated OPC CM, OPC lysates were also analyzed by western blot and showed a decrease in MMP-2 expression normalized over GAPDH loading control suggesting that MMP-2 secretion may be enhanced by ISP (Fig 6F). To test this, we incubated aggrecan with OPC CM treated with an exocytosis inhibitor, Exo1 (10µg/mL), in conjunction with ISP. At sufficient concentrations, Exo1 has been observed to reversibly inhibit exocytosis through its inhibition of the Arf GTPase(Feng et al. 2003). We found that Exo1 partially rescued aggrecan GAG digestion (Fig. 6I,J). We also performed the same experiment with a broad MMP inhibitor, GM6001 (25uM), and a specific MMP-2 inhibitor (OA-Hy,

Calbiochem, 100nM) with ISP and found that GAG digestion was partially rescued in

117 both cases indicating that ISP-induced CSPG degradation may very well be perpetrated by the metalloprotease family and MMP-2 predominantly (Fig. 6I,J). GM6001 and an

MMP-2 inhibitor additionally rescued CS56 spot degradation (Sup. Fig. 9A-D).

We returned to the spot assay to test whether MMP inhibition decreases OPC migration across the CSPG rim. Treatment of OPCs with GM6001 (25uM) and the specific MMP-2 inhibitor (OA-Hy, 100nM) effectively halted OPC entry into the CSPG-rich area even in the presence of ISP (Fig. 7A,B). This suggests that ISP-induction of enhanced OPC migration may be dependent on MMPs. Finally, to test the functional necessity of MMPs on remyelination, we treated LPC-demyelinated cerebellar slices with GM6001 or the specific MMP-2 inhibitor in conjunction with ISP. Colocalization of MBP and neurofilament+ axons was, indeed, decreased with GM6001 and the MMP-2 inhibitor treatment despite the presence of ISP (Fig. 7C,D). Furthermore, 2div inhibition of MMP-

2 in LPC-challenged (1µg/mL, 2 hours) OPCs cultured on low concentrations of aggrecan and laminin ablated the survival-promoting effects of ISP as assessed through

TUNEL staining (Sup. Fig. 9E). The MMP-2 inhibitor, however, did not seem to increase apoptosis compared to vehicle control on aggrecan/laminin alone (Sup. Fig. 9E). Specific

Inhibition of MMP-2 additionally negated gains in MBP footprints of mature oligodendrocytes grown on low concentrations of aggrecan/laminin (Sup. Fig. 9F,G).

To further elucidate the necessity of MMP-2 activity following ISP treatment in enhancing remyelination, we utilized a lentiviral particle-delivered shRNA construct. We first validated this shRNA approach using lentiviral delivery as well as western analysis

118 to knock down MMP-2 in OPC cultures infected for 48 hours (Sup. Fig.10A). shRNA knock down of MMP-2 was able to reduce the area of extended MBP+ processes of OLs cultured on aggrecan in vitro (Sup. Fig. 10B,C) and the ability of OPCs to migrate past a high aggrecan barrier (Sup. Fig. 10D,E) despite ISP treatment. As expected, shRNA knock down of MMP-2 also attenuated ISP-induced remyelination in cerebellar slice cultures (Fig. 7E,F).

We next characterized MMP-2 expression in vivo using immunohistochemistry and found that while the dorsal column of the naïve spinal cord expressed a baseline of some MMP-

2 protein, LPC injection into the same area somewhat increased MMP-2 expression at

14dpl (Fig. 8A,B). However, ISP treatment markedly enhanced MMP-2 expression in the

LPC-injected site, which was also confirmed with western blot analysis of the affected spinal cord areas (Fig. 8C). Immunostaining of ISP-treated LPC demyelinated cords showed MMP-2 colocalizing with Olig2-identified OPCs (Fig. 8D), but also with Iba1- labeled immune cells (Fig. 8E). To explore further the necessity of MMP-2 activity in

ISP-enhanced remyelination, we returned to the LPC model and analyzed lesion volume following myelin staining at 18dpl (Fig. 8F,G) with an MMP-2 inhibitor (OA-Hy) or shRNA construct targeting MMP-2 delivered with lentiviral particles. Interestingly, pharmacological inhibition of MMP-2 increased the lesion volume over vehicle control suggesting that baseline levels of MMP-2 may be facilitating slow remyelination in this model. The addition of MMP-2 pharmacological inhibition or MMP-2 shRNA-mediated knock down attenuated the enhanced remyelinating effects of ISP which correlated with a concomitant increase in CS56 immunoreactivity (Sup. Fig. 11A,B), and decreased

119 accumulation of Olig2+ OPCs in the lesion epicenter (Sup. Fig. 11C,D). Together, these results indicate the importance of PTPσ-regulated MMP-2 secretion by OPCs, but also possibly microglia, to assist them in their migration and ability to remyelinate despite high CSPG deposition after focal demyelinating injury.

3.5 Discussion

We have elucidated a critical role for the CSPG receptor PTPσ following its modulation in the restoration of OPC homeostasis in a variety of MS models where proteoglycan deposition during lesion associated scar formation potently inhibits OPC migration, differentiation, and remyelination. Here we report that targeting PTPσ with a systemically delivered peptide enhances the rate of myelin repair in LPC-induced lesions and stimulates robust myelin regeneration and functional recovery following chronic demyelinating EAE.

We also present a novel finding linking PTPσ modulation with altered immune polarization and enhanced protease activity. This underscores the important role that CSPGs play following CNS demyelinating diseases and identifies a strategy that can target MS lesions broadly throughout the neuraxis to relieve CSPG mediated inhibition.

The downstream mechanisms following peptide modulation of the receptor PTPσ that allow cells to overcome CSPG inhibition are largely unknown. Do treated cells simply fail to recognize the inhibitory chondroitin GAG chains or do they selectively modify the substrate to remove only inhibitory components of the ECM, sparing those that promote their growth? Protease activity is heavily regulated at several levels including transcription,

120 translation, secretion, localization, activation, and inhibition to prevent unfettered, potentially damaging activity (Haoqian Zhang et al. 2010). Left uncontrolled, proteases are able to degrade a broad range of proteins with potentially disastrous outcomes as seen in

“protease storms” following a variety of CNS injuries. In the relapsing phase of MS, nonspecific protease up-regulation associated with rampant inflammation and myelin degeneration have been well characterized (Brkic et al. 2015). However, the beneficial effects of finely regulated protease secretion following injury to promote tissue repair are becoming more appreciated. Here, we present a novel finding linking PTPσ modulation with enhanced MMP-2 protease activity by OPCs, which aids in their digestion through

CSPG-laden demyelinated plaque that envelops the MS-like lesion. Through in vitro, ex vivo and in vivo assays, we have identified the necessity of MMP-2 activity through PTPσ modulation not only for OPC migration but also for improved OPC survival, maturation, and remyelination. MMP-2 upregulation has been previously identified to allow stem cells to invade CSPG-containing regions of the glial scar (Veeravalli et al. 2009) and improve remyelination of peripheral axons by Schwann cells in culture (Lehmann et al. 2009).

Interestingly, previous work on damaged peripheral nerves where regeneration occurs spontaneously has also noted the fine-tuned regulation of neuronal MMP-2-induced degradation of CSPGs but with laminin sparing (Zuo et al. 1998). We have also observed laminin sparing and concomitant CSPG degradation with ISP treatment, which may be one mechanism by which OPCs are able to infiltrate into and survive otherwise CSPG-dense demyelinated lesions. This highlights the precise regulation of proteases by OPCs, and possibly immune cells, through PTPσ to promote controlled CSPG degradation. Whether

121

MMP-2 plays other roles in OPC homeostasis independent of CSPG disinhibition will need to be further explored.

Importantly, while we were able to detect MMP-2 in our various assays, we cannot exclude the possibility that other proteases may also be activated either in parallel or along the same

MMP-2 signaling axis. Other proteases that could be involved might include a variety of other MMPs such as MMP-12, which has been shown to regulate OPC maturation and morphological differentiation (Larsen and Yong 2004). Although found at very low concentrations in our assays, MMP-9, which has been shown to facilitate remyelination

(Larsen et al. 2003), may be upregulated as OPCs mature. Additionally, aggrecan- degrading ADAMTS4 has recently been found to be restricted to mature OLs and implicated in efficient myelination (Pruvost et al. 2017). In fact, knockout of ADAMTS4 seems to impair myelination enough to impair motor function in mice.

Recently, Dyck et al. (Dyck et al. 2018) found that daily treatment of spinal cord injured rats with our LAR family receptor blocking peptides dampened the pro-inflammatory environment induced by traumatic injury. Indeed, peptide treated macrophages from injured cords were found to express more reparative phenotype M2 markers and anti- inflammatory factors such as IL-10. Work from the Bradbury group involving large-scale

CSPG-degrading Chondroitinase ABC therapies following spinal cord injury also skewed macrophages toward the M2 spectrum (Bartus, James, Didangelos, Bosch, Verhaagen,

Yanez-Munoz, et al. 2014) again with enhanced IL-10 secretion (Didangelos et al. 2014).

Our study begins to confirm this observation in a peptide treated EAE model of MS where

122 we observed decreases in CSPG load as well as the destructive M1 macrophage phenotype marker iNOS and increases in M2 associated Arginase-1.

What might be the further effect of modulating the inflammatory environment by ISP on the behavior of microglia/macrophages or the evolution of CSPGs in the MS lesion? Why do we see decreased numbers of microglia/macrophages and/or less aggrecan within macrophages over time after peptide treatment? It has been shown that incubation of microglia/macrophages with CSPG-secreting reactive astrocytes (DeWitt et al. 1998) or

CSPGs in solution or bound to a substrate (Shaffer et al. 1995; Dyck et al. 2018) curtails the rate at which they phagocytose debris and degrade extracellular proteins, such as β- amyloid, in vitro. Our study suggests, in conjunction with Dyck et al. (Dyck et al. 2018), that ISP may be markedly enhancing microglial phagocytic capacity. By modulating the cognate receptor of CSPGs, OPCs and microglia may be working together to more rapidly clear inhibitory CSPGs and other cellular remnants. Thus, in altering the pro-inflammatory environment toward an M2 state (Kigerl et al. 2009) and inducing focal protease activity by selective cells, ISP may provide additional CSPG disinhibition, culminating in myelin regeneration. Whether modulation of PTPσ by peptide treatment additionally enhances phagocytic or protease activity in Iba1+ cells in the setting of MS to further aid in CSPG clearance, as well as cellular debris removal in general, will require further investigation.

123

3.6 Figures

124

Figure 1 ISP promotes functional and histological recovery in EAE mouse model. A) Diagram of ISP administration to EAE mice at the beginning or the peak of sickness determined by clinical score. B) Clinical score of disease severity in MOG35-55-induced EAE mice treated with ISP or vehicle daily beginning at the onset or the peak of disease. C) Mean improvement in disease score per animal of EAE cohort in B. (n=9 (EAE vehicle group), 15 (EAE ISP onset group) and 12 (EAE ISP peak group), ANOVA F(2,33)=20.96;

Tukey’s multiple comparison test, PEAE+Veh versus EAE ISP onset <0.0001 , PEAE+Veh versus EAE ISP peak =0.0004 ). D, E) Luxol fast blue (LFB) staining of myelin, demonstrating normal myelin integrity in ISP-treated EAE mice at 48 days post induction, in contrast to the marked loss of myelin present in the spinal cord of vehicle-treated control. Dashed lines demarcate lesion areas. Scale bar=100 µm. (n=5 mice/group, P=0.0002, t=6.647, df=8; Two-tailed unpaired Student’s t test). F) Double immunostaining for MBP and neurofilament-200 (NF200) in the thoracic spinal cord of vehicle- and ISP-treated EAE mice at 48 days post-induction. Dashed lines demarcate lesion areas. Scale bar=100 µm. G) Western blot analysis of MBP expression in spinal cord tissue of vehicle or ISP-treated control mice and EAE mice at 48 days post-induction. Data are normalized to β-actin protein expression. (n=4 mice/group, ANOVA F(3,12)=26.68; Tukey’s multiple comparison test, Pcon versus EAE+Veh = 0.0001, PEAE+Veh versus EAE+ISP = 0.0037). H) Electron micrographs from ventral lumbar spinal cords of vehicle and ISP-treated EAE mice 48 days following induction. Scale bar=2 µm. I) Number of myelinated axons in the spinal cord lesions of vehicle and ISP-treated EAE mice. (n=3 mice/group, P=0.0014, t=7.815, df=4; two-tailed unpaired Student’s t test). J) Quantification of the g-ratios (axon diameter/fiber diameter) of myelinated fibers in the ventral lumbar spinal cords of vehicle and ISP-treated EAE mice. (EAE+Veh group, g-ratio= 0.8874 ±0.001901; EAE+ISP group, g-ratio=0.8423ISP group; n=134 remyelinated axons from 3 mice/groupP<0.0001, t=14.31, df=266; two-tailed unpaired Student’s t test). The data are presented as mean±s.e.m. *P<0.05, **P<0.01, *** p < 0.001.

125

Figure 2 ISP promotes remyelination in the spinal cord of lysolecithin (LPC)-demyelinated mice. A) Representative LFB-stained sections of LPC lesions from the spinal cords of vehicle or ISP treated mice. Dashed lines demarcate lesion areas. Scale bar=100 µm. B) Quantitative analysis of the volume of lesioned spinal cord in vehicle or ISP-treated mice at

126

3, 7, 14 and 21dpl. (n=4 mice/group, ANOVA F(3,12)=16.41; Sidak’s multiple comparison test, 14dpl: PLPC+Veh versus LPC+ISP = 0.0003; 21dpl: PLPC+Veh versus LPC+ISP <0.0001). C) Double immunostaining for MBP and DAPI in the spinal cord of vehicle- and ISP-treated mice at 14 and 21dpl. Dashed lines demarcate lesion areas. Scale bars=100 µm. D) Western blot analysis of MBP expression in spinal cord tissue of vehicle or ISP-treated mice at 14dpl. Data are normalized to β-actin protein expression. (n=4 mice/group, ANOVA

F(3,12)=24.21; Tukey’s multiple comparison test, Pcon versus LPC+Veh < 0.0001, PLPC+Veh versus

LPC+ISP = 0.0276). E) Representative electron microscopy images of LPC lesions from the spinal cord of vehicle or ISP-treated mice at 14dpl. Scale bar=5 µm. F) The number of myelinated axons in LPC-induced lesions from vehicle or ISP-treated mice at 14dpl. (n=3 mice/group, P=0.0001, t=14.26, df=4; two-tailed unpaired Student’s t test). G) The myelin g-ratio in the LPC-lesions of vehicle or ISP-treated mice at 14dpl. (LPC+veh group, g- ratio= 0.9103 ±0.003583; LPC+ISP group, g-ratio=0.8741±74103599; n=139 remyelinated axons from 3 mice/groupP<0.0001, t=7.142, df=276; two-tailed unpaired Student’s t test). The data are presented as mean±s.e.m. *P<0.05, **P<0.01, *** p < 0.001.

127

Figure 3 ISP accelerates remyelination in LPC treated organotypic cerebellar cultures. A) Representative immunohistochemistry images of MBP and neurofilament-200 (NF200) show normal myelination in naïve (Con) sections, LPC-induced demyelination at 1 day in vitro (div), and increased remyelination after ISP treatment in LPC-demyelinated cerebellar slices at 8div and 14div. Scale bar=100 µm. B) Bar graph illustrates relative MBP immunoreactivity (i.e. co-localization of MBP and NF200) in cerebellar slices compared to no treatment (100% as control). (n=9 slices from 3 independent replicates per group,

ANOVA F(5,48)=230.4, Tukey’s multiple comparison test, 8dpl: PLPC+Veh versus LPC+ISP <

128

0.0001, 14dpl: PLPC+Veh versus LPC+ISP < 0.0001). C) Western blot analysis of MBP expression in vehicle or ISP-treated cerebellar slices at 8div and 14div. Data are normalized to β-actin protein expression. (n=3 independent replicates per group. 8div: ANOVA F(3,8)=58.89,

Tukey’s multiple comparison test, PVeh versus LPC+Veh < 0.0001, PLPC+Veh versus LPC+ISP = 0.0187;

14div: ANOVA F(3,8)=6.281, Tukey’s multiple comparison test, PVeh versus LPC+Veh < 0.048,

PLPC+Veh versus LPC+ISP = 0.025.) The data are presented as mean±s.e.m. *P<0.05, **P<0.01, *** p < 0.001.

129

130

Figure 4 ISP decreases chondroitin sulfate proteoglycan (CSPG) load in both EAE and LPC models. A) Representative immunohistochemistry images of Iba1 and Cat301 (aggrecan- specific antibody) show decreased accumulation of CSPG and microglia/macrophages in the thoracic spinal cord of ISP-treated compared to vehicle treated mice EAE mice at 41 days post-induction. Scale bar=100 µm. B) Representative immunohistochemistry images of MBP, Cat301 and CS56 (glycosaminoglycan specific antibody) show decreased accumulation of CSPG after ISP treatment at 14dpl in LPC demyelination mice. Scale bar=100 µm. C) Representative immunohistochemistry images of Iba1, GFAP, MBP, Cat301 and CS56 show decreased accumulation of CSPG after ISP treatment at 7dpl in LPC demyelination mice. Scale bar=100 µm. D) Relative quantification of immunofluorescence intensity of Iba1, GFAP, MBP, Cat301 and CS56 in the spinal cord of vehicle or ISP-treated LPC mice at 7dpl. (n=3 mice/group, Iba1: P=0.1848, t=1.6, df=4; Cat301: P=0.0092, t=4.719, df=4; GFAP: P=0.1111, t=2.039, df=4; CS56: P=0.0028, t=6.55, df=4; MBP: P>0.9999, t=0, df=4; Versican: P=0.0095, t=4.669, df=4; two-tailed unpaired Student’s t test). The data are presented as mean±s.e.m. **P<0.01.

131

Figure 5 ISP increases CSPG-degrading protease activity. A,B) CSPG gradient crossing assay shows that ISP treatment promotes the crossing of PDGFRα+ or O4+ OPCs through the gradient of CSPG. Scale bar=100 µm. C) Quantification of immunostaining for the amount of PDGFRα+ or O4+ OPCs crossing the CSPG barrier after vehicle or ISP treatment. (n=9 spots from 3 independent replicates. PDGFRα: P<0.0001, t=7.99, df=16; O4: P<0.0001, t=9.419, df=16, two-tailed unpaired Student’s t test). The data are presented as mean±s.e.m.

132

D) Representative immunostained images of CS56 and PDGFRα+ OPCs on CSPG barrier depicting CSPG degradation after ISP treatment as they cross the barrier to leave CS56 “shadows” (inset and arrows). E) To investigate protease activity, OPC conditioned media (CM) was treated with vehicle control or 2.5uM ISP or scrambled-ISP (SISP) and incubated with aggrecan (20ug/mL) or laminin (10ug/mL), then analyzed through western blots. F) Quantification of glycosaminoglycan moiety through CS56 immunoblotting reveals significant (One-Way ANOVA, Dunnett’s posthoc test, P=0.0432, F(2,12)=4.131, N=5 western blots) CS56 degradation following ISP treatment. G) Quantification of laminin immuoblotting shows no significant changes (One-Way ANOVA, Tukey’s posthoc test, P=0.9024, F(2,15)=0.1034), N=6 western blots). H) Quenched casein in EnzChek protease activity assay fluoresces once it becomes cleaved. I) Quantification of EnzChek protease activity assay reveals significant protease activity (One-Way ANOVA Dunnett’s posthoc test, P=0.0015, F(2,18)=9.534, N=26 from 7 replicates) in OPC CM treated with 2.5uM ISP over control. Graphs indicate scatterplot representations of western blots or mean of each replicate with standard error means. *P<0.05, **P<0.01, *** p < 0.001. n.s., not significant.

133

Figure 6 ISP increases MMP-2 secretion and activity. To further characterize protease activity, A) cultured OPCs were treated with vehicle control or 2.5uM ISP or SISP, concentrated, then loaded onto gelatin SDS/PAGE gels for zymography analysis. 25ng of recombinant MMP-

134

9 or MMP-2 served as positive controls. B) Quantification of active MMP-2 lanes of gelatin zymography reveals significant MMP-2 activity following ISP treatment over control (One-Way ANOVA, Dunnett’s posthoc test, P=0.0161, F(2,12)=5.937, N=5 zymograms). C) Enhanced MMP-2 expression following ISP treatment was confirmed in western blots of concentrated OPC conditioned media (CM). D) Quantification of MMP-2 immoblotting was significantly enhanced following ISP treatment over control (One-Way ANOVA, Dunnett’s posthoc test, P=0.0150, F(2,15)=5.63, N=6 western blots). E) In contrast, MMP-10 immunoblotting was not significant among treatments (One-Way ANOVA, Tukey’s posthoc test, P=0.9619, F(2,15)=0.03899, N=6 western blots). I) To explore whether ISP induces secretion of proteases to degrade CSPGs, cultured OPCs were treated with the following drugs with or without 2.5uM ISP: exocytosis inhibitor Exo1 (10ug/mL), general metalloprotease inhibitor GM6001 (25uM), or specific MMP-2 inhibitor (OA-Hy, Calbiochem, 100nM). Collected CM was incubated with aggrecan (20ug/mL) and immunoblotted with CS56. J) Quantification of CS56 reveals significant ISP-induced degradation of CSPGs over control, Exo1+ISP, GM6001+ISP, and OA-Hy + ISP (One-Way ANOVA, Tukey’s posthoc test, P=0.0010, F(7,32)=4.749 N=5 western blots). H) Immunostaining of O4+ (red) OPCs shows MMP-2 concentrated in OPC soma and processes. Graphs indicate scatterplot representations of western blots or mean of each replicate with standard error means. *P<0.05. n.s., not significant.

135

136

Figure 7 ISP-induced MMP-2 activity increases OPC migration and remyelination through CSPG disinhibition. To ask whether ISP-induced protease activity is involved in OPC migration and remyelination, A) cultured OPCs were plated onto coverslips with CSPG spot gradients and treated with vehicle control, 2.5uM ISP or SISP, 25uM GM6001 +/- ISP or SISP, or 100nM MMP-2 inhibitor (OA-Hy, Calbiochem) +/- ISP or SISP. The amount of O4+ OPCs crossing the CSPG barrier was counted and B) quantified. ISP treatment (N=37 spots, 5 replicates) significantly induced greater O4+ OPC migration past the CSPG barrier compared to control (One-Way ANOVA, Tukey’s posthoc test, P=0.0002, F(11,48)=5.013), GM6001+ISP (N=20 spots), or OA-Hy + ISP (N=20 spots) treatments. N(Spots)=31 Control, 26 SISP, 18 GM6001, 20 GM6001+SISP, 18 OA-Hy, 19 OA- Hy+SISP) C) To test whether remyelination was affected by protease inhibition, P7-9 cerebellar slices were all treated with LPC for 18 hours, then treated with control vehicle, 2.5uM ISP or SISP, 25uM GM6001+/- ISP, or 100nM OA-Hy +/- ISP for 9 days before staining for neurofilament (NF200) or MBP. D) Remyelination was quantified through MBP and neurofilament colocalization. ISP treatment (N=35 images from 4 replicates with up to 13 sections total) significantly increased MBP-neurofilament colocalization over control (One-Way ANOVA, Tukey’s posthoc test, P=0.0001, F (8, 88) = 13.61), GM6001 + ISP (N=28), and OA-Hy + ISP (N=23) groups. N(images)=17 Control, 22 SISP, 48 GM6001, 20 GM6001+SISP, 22 OA-Hy, 22 OA-Hy +SISP Graphs indicate scatterplot representations of mean of each replicate with standard error means. *P<0.05, **P<0.01, *** p < 0.001. E-F) Cerebellar slice cultures were treated with lentiviral constructs for 48 hours before LPC treatment. Vehicle or ISP (2.5uM) treatment followed for 6 days in vitro. MBP immunofluorescence (green) was then quantified with NF200 (red). (One-Way ANOVA, Tukey’s posthoc test. P=0.0005. F(3, 116)=6.395. N=around 30 images from 10 slices). Graphs indicate scatterplot representations of mean of each replicate with standard error means.

137

138

Figure 8 ISP promotes myelin repair through increasing MMP-2 expression in LPC-induced demyelination model of mice. A) Representative images from immunohistochemistry of MMP-2 and DAPI show increased levels of MMP-2 in the spinal cord of ISP-treated mice at 7 days post LPC injection (dpl) compared to naïve or LPC vehicle control cords. Dashed lines demarcate lesion areas. Scare bar=100 µm. (B) Relative quantification of immunofluorescence intensity of MMP-2 in the spinal cord of ISP-treated mice at 14dpl

(n=3 mice/group, ANOVA F(2,6)=48.12, Tukey’s multiple comparison test, Pcon versus LPC =

0.0075, PLPC versus LPC+ISP = 0.0056). (C) Western blot analysis of MMP-2 expression in spinal cord tissue of naïve, vehicle, or ISP-treated mice at 14dpl. Data normalized to β- actin protein expression. (n=3 mice/group, ANOVA F(2,6)=33.14; Tukey’s multiple comparison test, Pcon versus LPC = 0.0168, PLPC versus LPC+ISP = 0.0143). (D) Representative immunohistochemistry images show Olig2+ OPCs express MMP-2 in the spinal cord of ISP-treated mice at 14dpl. White arrows indicate the colocalization of MMP-2 and Olig2. (E) Representative immunohistochemistry images show Iba1+ cells (microglia/macrophage) express MMP-2 in the spinal cord of ISP-treated mice at 14dpl. White arrows indicate the colocalization of MMP-2 and Iba1. Scare bar= 100 µm.(F) Representative eriochrome cyanine (myelin) staining of LPC lesions from the spinal cords of naïve, vehicle, ISP, MMP-2 inhibitor (OA-Hy), or MMP-2 shRNA treated mice. Dashed lines demarcate lesion areas. Scale bar=100 µm. (G) Quantitative analysis of the volume of lesioned spinal cord in vehicle, ISP, MMP-2 inhibitor (OA-Hy), or MMP-2 shRNA treated mice at 18dpl. (n=4 mice/group, ANOVA F(5,18)=169.7; Tukey’s multiple comparison test, PLPC versus LPC+ISP < 0.0001; PLPC+ISP versus LPC+ISP+MMP-2(-) <0.0001; PLPC+ISP versus

LPC+ISP+MMP-2 shRNA <0.0001; PLPC versus LPC+MMP-2(-) =0.0279 ). The data are presented as mean±s.e.m. *P<0.05, *** p < 0.001.

139

Supplementary Figure 1 Increased CSPG load in mouse models of MS. A-B) Representative LFB-stained sections and immunohistochemistry images of Cat301 and CS56 show CSPG accumulation in the thoracic spinal cord of EAE mice at 28 and 48 days post-induction. Scale bar=100 µm. Quantification of pixel intensities of Cat301 (aggrecan CSPG) and CS56 (glycosaminoglycan moieties of CSPGs) depicted. (Cat301: n=3 mice/group, ANOVA

F(2,6)=163.9, Tukey’s multiple comparison test, Pcon versus EAE D28 =0.0007, PEAE D28 versus EAE

D41 = 0.0001; CS56: n=3 mice/group, ANOVA F(2,6)=168.7, Tukey’s multiple comparison test, Pcon versus EAE D28 =0.0052, PEAE D28 versus EAE D41 < 0.0001).

140

(C-D) Representative LFB-stained sections and immunohistochemistry images of Cat301 and CS56 show CSPG accumulation in the lesion site after LPC demyelination at 7dpl and 14dpl. Scale bar=100 µm. Quantification of pixel intensities of Cat301 and CS56 depicted

(Cat301: n=3 mice/group, ANOVA F(2,6)=269.7, Tukey’s multiple comparison test, Pcon versus 7 dpl < 0.0001, P7dpl versus 14 dpl < 0.0001; CS56: n=3 mice/group, ANOVA F(2,6)=105,

Tukey’s multiple comparison test, Pcon versus 7 dpl < 0.0001, P7dpl versus 14 dpl = 0.0011).

141

Supplementary Figure 2 PTPσ expression is enhanced following EAE and LPC. A) Representative immunocytochemistry images of PTPσ and Olig2 show PTPσ expression on oligodendrocyte progenitor cells grown in vitro. B) Graphical representation of cell- specific PTPRD (PTPδ), PTPRS (PTPσ), PTPRF (LAR) and RTN4R (Nogo) mRNA levels obtained from a publicly available RNA-sequencing transcriptome database (https://web.stanford.edu/group/barres_lab/). FPKM represents fragments per kilobase of

142 transcript sequence per million mapped fragments. C) Representative immunohistochemistry images of PTPσ, CC1, and MBP show PTPσ expression in CC1+ or MBP+ cells in OPCs culture in vitro. D) Western blot analysis of PTPσ expression in the thoracic spinal cord of EAE or LPC-induced demyelination mice. Data are normalized to

β-actin protein expression (n=4 mice per group. Pcon versus EAE =0.012, t=3.558, df=6; Pcon versus LPC=0.0012, t=5.775, df=6; two-tailed unpaired Student’s t test). E) Representative immunohistochemistry images of PTPσ and Olig2 show increased expression of PTPσ in Olig2+ cells in the spinal cord of EAE mice at 28 days post-induction. Scale bar=50 µm. The data are presented as mean±s.e.m. *P<0.05, **P<0.01.

143

144

Supplementary Figure 3 ISP increases CSPG clearing as remyelination occurs. A) Representative immunohistochemistry images of MBP and Cat301 show reduced abundance of CSPG after ISP treatment in LPC-demyelinated cerebellar slices at 8dpl and 14dpl. Scale bar=100 µm. (B) Relative quantification of immunofluorescence intensity of Cat301 after Vehicle or ISP treatment in LPC-demyelinated cerebellar slices at 4dpl, 8dpl and 14dpl (n=9 slices from 3 independent replicates per group, two-way ANOVA F(2,6)=30.78, Sidak’s multiple comparison test, 8dpl: PLPC+Veh versus LPC+ISP < 0.0001, 14dpl: PLPC+Veh versus LPC+ISP = 0.0325). ). *P<0.05, *** p < 0.001, n.s: no significance.

145

Supplementary Figure 4 ISP modulates inflammation in EAE models of mice. A) Representative immunohistochemistry images of Iba1 and GFAP show decreased activation of microglia and astrocytes respectively in the spinal cord of ISP-treated mice at day 41 following EAE induction. Scale bar= 100 µm. (B) Relative quantification of immunofluorescence intensity of Iba1 and GFAP in the spinal cord of ISP-treated mice at day 41 (n= 3 mice/group Iba1:

146

P=0.0016, t=7.608, df=4; GFAP: P=0.0113, t=4.448, df=4. two-tailed unpaired Student’s t test). (C) Representative immunohistochemistry images of iNOS (M1 microglia marker) and Arginase-1 (M2 microglia marker) show increased M2 microglia and decreased M1 microglia in the spinal cord of ISP-treated mice at day 28. Scale bar= 100 µm. (D) Quantification of relative immunofluorescence of iNOS and Arginase-1 in the spinal cord of ISP-treated mice at day 41 (n= 3 mice/group iNOS: P=0.0008, t=9.114, df=4; Arginase-1: P=0.0014, t=7.824, df=4. two-tailed unpaired Student’s t test). (E) Representative Luxol fast blue staining images show no differences in demyelination area of Vehicle or ISP-treated mice at day 18. (F) Quantification of demyelination area in the spinal cord of Vehicle or ISP-treated mice at day 18 indicates similar levels of demyelination in EAE model (n= 3 mice/groupP=0.8559, t=0.1936, df=4, two-tailed unpaired Student’s t test). *P<0.05, *** p < 0.001, n.s: no significance.

147

Supplementary Figure 5 ISP promotes OPC recruitment and survival on CSPGs. A) Representative images of Ki67 immunostaining showing proliferating OPCs (Olig2+) in the spinal cord of vehicle- and ISP-treated mice at 7dpl. White arrows show Olig2+/Ki67+ cells. Scale bar=100 µm. B) Quantification of immunostaining showing Olig2+ cells/mm2, Ki67+ cells/mm2 and Olig2+/Ki67+ cells/mm2 in the spinal cord of vehicle- and ISP-treated mice at 7dpl. (n=3 mice/group, Olig2: P=0.0012, t=1.6, df=4; Ki67: P=0.0449, t=2.883, df=4; Olig2+/Ki67+:

148

P=0.056, t=2.666, df=4. two-tailed unpaired Student’s t test). The data are presented as mean±s.e.m. C) Animals treated with vehicle or ISP for 7div following spinal cord LPC injections. Sections were stained with Olig2, Ki67, and DAPI B) Quantification of Olig2, Ki67, and colocalized Olig2 and Ki67 in LPC sections. C) OPCs were cultured on aggrecan/laminin-precoated coverslips, treated with vehicle control, 2.5uM ISP or SISP, and stained with O4 and Ki67. D) Ki67 was not significantly changed among the groups (One-Way ANOVA, Tukey’s posthoc test, P=0.8998, F(2,9)=0.1068, N=12 images each with 4 replicates and 3 repeats each). E) OPCs were cultured on aggrecan/laminin and treated with control or 2.5uMISP for 2div before incubation with vehicle or LPC (1ug/mL) for 2 hours before DAPI and TUNEL staining. F) Quantification of TUNEL+ over DAPI+ cells reveal significant survival of ISP-treated (N=40 images, 4 replicates) OPCs over control (One-Way ANOVA, Tukey’s posthoc test, P=0.0021, F(3,36)=5.96). ISP-treatment enhanced survival of LPC-treated cells. (N(images)=36 control, 28 each LPC-treated group). Graphs indicate scatterplot representations of mean of each replicate with standard error means. *P<0.05, **P<0.01, *** p < 0.001.

149

150

Supplementary Figure 6 ISP enhances OPC process outgrowth and maturation. A) Representative immunohistochemistry images of Olig2 and CC1 in the thoracic spinal cord of vehicle or ISP-treated EAE mice at 41 days post-induction. Scale bar=50 µm. B) Quantification of immunostaining for Olig2+ cells/mm2 and CC1+ cells/mm2 in the thoracic spinal cord of vehicle or ISP-treated EAE mice at 41 days post-induction. (n=3 mice/group, Olig2: P=0.0189, t=3.811, df=4; CC1: P=0.0019, t=7.259, df=4. two-tailed unpaired Student’s t test). C) Representative immunohistochemical images of DAPI and CC1 in the spinal cord of vehicle or ISP-treated LPC mice at 14dpl. Dashed lines demarcate lesion areas. Scale bar=100 µm. D) Quantification of immunostaining for normalized CC1+ oligodendrocytes density at 14dpl. (n=3 mice/group, P=0.0044, t=5.789, df=4. two-tailed unpaired Student’s t test). E) Representative immunohistochemical images of the maturation of OPCs after plating onto poly-l-lysine or CSPGs in the presence of ISP or vehicle. Scale bar=100 µm. F) Quantification of the relative proportion of maturing OLs after OPCs plating onto poly- l-lysine (control) or CSPGs in the presence of ISP or vehicle. (n= 3 independent replicates , O4: ANOVA F(2,6)=1.321, P=0.3347; MBP: ANOVA F(2,6)=30.99, Tukey’s multiple comparison test, PCon versus CSPGs+Veh = 0.0005, PCSPGs+Veh versus CSPG+ISP =0.0175). G) Comparison of the size of MBP+ footprints after OPCs plating onto poly-l-lysine (control) or CSPGs in the presence of ISP or vehicle at 6 days. (n= 3 independent replicates,

ANOVA F(2,6)=40.96, Tukey’s multiple comparison test, PCon versus CSPGs+Veh = 0.0003,

PCSPGs+Veh versus CSPGs+ISP =0.0052). The data are presented as mean±s.e.m. *P<0.05, **P<0.01, *** p < 0.001. n.s., not significant.

151

Supplementary Figure 7 ISP-treated Conditioned Media Protease Array. To begin screening which proteases may be upregulated by ISP treatment, cultured OPCs were treated with vehicle control or 2.5uM ISP for 4 days in vitro. Conditioned media was then incubated with qualitative protease array (R & D Systems) and developed. % Change in pixel intensities of ISP- treated vs. control OPC CM was then calculated.

152

Supplementary Figure 8 ISP enhances CS56 degradation in a dose-dependent manner. ISP-treated OPC CM is capable of degrading CS56 spots. A) OPCs treated with vehicle control, 2.5uM ISP or SISP was incubated for 2 days in vitro before CM was collected and incubated with

153 aggrecan/laminin spots. An additional subset of spots was incubated for the same duration with media only. Spots were immunostained with CS56 or laminin and the pixel intensities of the spot rim were recorded. B) Representative images of CS56-stained spots with yellow region of interest indicating measured portion of the spot. C) Quantification of CS56- immunostained spot indicates ISP-treated OPC CM significantly degrades CSPGs over control (One-Way ANOVA, Dunnett’s posthoc test, P=0.0001, F(5,72)=45.19). N(images, 5 replicates)=69 Control, 104 ISP, 95 ISP, 31 Media only) D) Representative images of laminin-stained spots. E) Quantification of laminin-stained spots indicate no significant changes between groups (One-Way ANOVA, Tukey’s posthoc test, P=0.0818, F(3,85)=2.312), N(images, 5 replicates)=133 Control, 134 ISP, 114 SISP, 34 Media Only) F) To confirm CS56 degradation, OPCs were treated with varying doses of ISP or vehicle control and incubated with a fixed concentration of aggrecan (20ug/mL) for 2 hours before western blot analysis. G) Western blot analysis of CS56 and subsequent H) quantification of CS56 band indicate significant ISP-treated CS56 degradation over control at 2.5 and 5uM doses (One-Way ANOVA, Tukey’s posthoc test, P=0.0049, F(5,12)=6.112, N=3 western blots). Graphs indicate scatterplot representations of western blots or mean of each replicate with standard error means.

154

155

Supplementary Figure 9 Protease inhibitors attenuate ISP-induced CSPG degradation and subsequent CSPG- OPC disinhibition. To assess functional effects of protease inhibitors on OPCs, A-B) CS56-immunostained spots were incubated with vehicle control, 2.5uM ISP, or 25uM GM6001 +/- ISP treated OPC CM. Quantified CS56 immunoreactivity reveals significant ISP-induced CS56 degradation over control and GM6001+ISP (One-Way ANOVA, Tukey’s posthoc test, P=0.0001, F (3, 37) = 21.43), N(images, 3 replicates)=24 Control, 17 ISP, 19 GM6001, 20 GM6001+ISP). Similarly, C-D) aggrecan spots treated with vehicle control, 2.5uM ISP, or 100nM MMP-2 inhibitor (OA-Hy) +/- ISP revealed ISP-induced degradation over vehicle control and MMP-2 inhibitor (OA-Hy) + ISP (One-Way ANOVA, Tukey’s posthoc test, P=0.001, F (3, 44)=31.50). N=24 images each) E) To assess whether the MMP-2 inhibitor affected apoptosis, OPCs cultured on aggrecan/laminin precoated coverslips were treated with ISP, 100nM MMP-2 inhibitor (OA-Hy) +/- ISP for 2 days in vitro. A subset of treated OPCs was additionally challenged with LPC (1ug/mL) incubation for 2 hours before TUNEL/DAPI staining. MMP-2 inhibitor (OA-Hy) did not significantly increase TUNEL+ cells/total cells. However, MMP- 2 inhibitor (OA-Hy) even with ISP treatment increased apoptosis following LPC treatment over LPC-ISP treatment (One-Way ANOVA, Tukey’s posthoc test, P=0.0001, F(7,21)=29.66), N=2 replicates, 2 wells each). F) 100nM MMP-2 inhibitor (OA-Hy) treatment of OPCs cultured on aggrecan/laminin additionally decreased MBP footprint significantly over ISP treatment alone (One-Way ANOVA, P=0.0001, F(3,112)=228.3), N(cells, 5 replicates total)=128 Control, 126 ISP, 191 OA-Hy, 222 OA-Hy+ISP). Graphs indicate scatterplot representations of mean of each replicate with standard error means.

156

Supplementary Figure 10 shRNA knock down of MMP-2 decreases OPC maturation and migration on CSPGs to limit remyelination in cerebellar slices. A) OPC cultures infected with control lentiviral scrambled shRNA or lentiviral particles expressing shRNA construct targeting MMP-2 for 48 hours before western blotting to assess MMP-2 knock down compared to GAPDH. B-C) Scrambled or shRNA targeting MMP-2 lentiviral-infected OPCs cultured on laminin and low concentration of aggrecan (1ug/mL) were immunostained with MBP following vehicle or ISP (2.5uM) treatment for 48 hours. MBP area was quantified. (One- Way ANOVA, Tukey’s posthoc test, P=0.013, F(3, 158)=6.677, N=100 cells from 2 repeats) D-E) OPCs (O4, red) were infected with lentiviral constructs 48 hours before plating onto spot assays to assess migration across aggrecan (green) with or without ISP

157

(2.5uM). Number of OPCs crossing aggrecan spot were counted (One-Way ANOVA, Tukey’s posthoc test, P=0.0015, F(3, 44)=6.056. N=40 spots from 2 repeats).

158

Supplementary Figure 11 MMP-2 mediates ISP-induced remyelination in LPC-demyelinated mouse model. (A) Representative images from immunohistochemistry of CS56 and DAPI from the spinal cords of vehicle, ISP or MMP-2 inhibitor (OA-Hy)-treated mice show MMP-2-mediated CS56 degradation at 14 days post LPC injection. Dashed lines demarcate dorsal white matter of spinal cord. Scare bar= 100 µm. (B) Relative quantification of immunofluorescence intensity of CS56 in the spinal cord of vehicle, ISP or MMP-2 inhibitor (OA-Hy)-treated mice at 14dpl (n=3 mice/group, ANOVA F(3,8)=76.08, Tukey’s multiple comparison test, PLPC versus LPC+ISP < 0.0001, PLPC+ISP versus LPC+ISP+MMP-2(-) = 0.0079,

PLPC versus LPC+MMP-2(-) = 0.0026). (C) Representative images from immunohistochemistry of Olig2 and DAPI from the spinal cords of vehicle, ISP or MMP-2 inhibitor (OA-Hy)-treated mice show MMP-2-mediated OPCs migration at 14 days post LPC injection. Dashed lines demarcate lesion areas. Scare bar= 100 µm. (D) Quantification of the number of Olig2+ cells in the spinal cord of vehicle, ISP or MMP-2 inhibitor (OA-Hy)-treated mice at 14dpl

(n=3 mice/group, ANOVA F(3,8)=21.58, Tukey’s multiple comparison test, PLPC versus

LPC+ISP = 0.0074, PLPC+ISP versus LPC+ISP+MMP-2(-) = 0.0088, PLPC versus LPC+MMP-2(-) = 0.0385).

159

IV. GENERAL DISCUSSION

Chondroitin sulfate proteoglycans (CSPGs) have conventionally been considered repulsive components of the extracellular matrix, especially by Developmental Neuroscientists (D.J.

Silver and J. Silver 2014; Wiese and Faissner 2015). Ongoing work, including studies by

Lang et al., have modified this observation to highlight the ability of a gradient of CSPGs to permanently immobilize approaching axons and trap them in a dystrophic state. The

NG2 proteoglycan, for example, has been previously shown to stabilize the approaching growth cone as an irregular synapse (Filous et al. 2014; Son 2015), and dystrophic growth cones have been observed to persist even 40 years after injury to the spinal cord (Ruschel et al. 2015). How dystrophic growth cones may be transformed and intrinsically galvanized into a regenerating state has been a long-standing source of interest in the Silver lab.

Indeed, the idea that CSPGs are inhibitory ECM proteins with far reaching consequences for a variety of other cell types found within the CNS is becoming more appreciated.

Work delineated by this thesis hopes to add to this field of work by proposing one mechanism by which not only axons, but oligodendrocyte progenitor cells use to extricate themselves from CSPG inhibition. This is accomplished through modulation of the receptor protein tyrosine phosphatase sigma (PTPσ), found in both peripheral neurons and oligodendrocyte progenitor cells, which enhances protease secretion and activity in the

ECM that degrades CSPG to relieve the immediate environment of the growth cone or oligodendrocyte progenitor cell of CSPG inhibition.

160

4.1.1 Summary and Discussion of Modulation of Protein Tyrosine Phosphatase

Receptor Sigma Leading to Enhanced Cathepsin B Activity in Peripheral

Axons

Our studies link, for the first time, selective protease release following modulation of PTPσ with a synthetic peptide. This was first observed in vitro through diminished GAG-CSPG

(CS-56) staining in the Silver spot assay. Upon closer inspection with lower doses of ISP, certain axons were visualized to leave a trail of digested GAG-CSPG similarly to observations by the Seeds group whose neuroblastoma clonal cell lines (Krystosek and

Seeds 1981) and dissociated mouse dorsal root ganglion (DRG) neurons (McGuire and

Seeds 1990; Hayden and Seeds 1996) cultured on fibrin slides also showed selective release of proteases at the “trailblazing” tip of the neurite to leave regions of digested substrate.

The Seeds group has additionally discovered this phenomenon in the developing cerebellum, specifically in granule neurons as they migrated from the external granule layer

(Friedman and Seeds 1994). Since this initial discovery, the use of selective protease secretion has been observed in many other instances of migrating axons during development. For example, Brooks et al. found that ADAMTS-4, or aggrecanase, secretion by retinal axons allowed for their initial entry into the aggrecan-laden lateral geniculate nucleus before corticogeniculate neurons (J.M. Brooks et al. 2013). In this instance, finely controlled protease secretion not only enabled retinal migration into their intended target but coordinated the temporal innervation of corticogeniculate neurons as well.

161

While developing neurons clearly utilize protease secretion to negotiate through a complicated ECM landscape, it is becoming clear that this strategy is also recapitulated by regenerating axons in adulthood. For example, regenerating olfactory receptor cells, which periodically turnover and therefore must project back to the olfactory bulb to reestablish connections with the CNS, utilize a variety of proteases including β-secretase 1 (BACE1) to maneuver their way to the olfactory bulb (Rajapaksha et al. 2011; Cao et al. 2012).

Additionally, pre-conditioning injuries, which have been well characterized to enhance transcription of regenerating-associated genes to enhance axon regeneration following spinal cord injury (Neumann and Woolf 1999) require protease activity to promote axonal outgrowth (Minor et al. 2009). In fact, extracted peripheral axons from tPA knockout animals failed to regenerate past a myelin substrate even after a preconditioning injury

(Minor et al. 2009). Thus, the enhancement of selective protease activity may be one mechanism of many that shifts adult axons to a regenerating “mode.”

In our own cultures where axonimized DRG neurons normally extend their processes across a growth-conducive environment, some protease activity was observable even without ISP treatment as seen in a DQ gelatin protease activation assay. In this assay, DQ gelatin fluoresces when cleaved. Fluorescing substrate was visible in the vehicle-treated control DRG culture around the soma; however, ISP-treatment greatly enhanced this basal level of protease activity beyond the soma along the axons. Moreover, protease GAG-

CSPG degrading activity increased in an ISP dose-dependent manner suggesting that PTPσ modulation and activity of the receptor itself may be graded.

162

Although protease release by neurons has been previously observed, exactly how and when such secretion occurs has been unclear. Using ISP as an exogenous “switch” to modulate

PTPσ dynamics allowed us to more easily delve into some mechanism underlying PTPσ- related protease activity. Even in conjunction with transcription and translation inhibitors including anisomycin/α-amanatin and cycloheximide respectively, ISP treatment of DRGs enhanced GAG-CSPG degradation suggesting that protease activity was independent of these processes. The addition of an exocytosis inhibitor, Exo1, however greatly reduced

GAG-CSPG digestion even in the presence of ISP. Whether PTPσ specifically regulates exocytosis of proteases alone, exocytosis as it occurs in growth cones in general, or the cellular signaling pathway leading to enhanced exocytosis will need to be further studied.

In addition, our work has identified cathepsin B through mass spectrometry and western blot of conditioned media as the protease whose secretion was enhanced following ISP treatment. Although cathepsin B was positively identified, this study does not exclude the possibility of other proteases working in conjunction to degrade GAG-CSPGs surrounding

DRGs. That said, cathepsin B is an interesting protease in that not only was it robustly enhanced and therefore easily identified in our experiments, but cathepsin B has been shown to play a unique role as an activating hub in tumorigenic cells as they metastasize

(Olson and Joyce 2015). Indeed, selective protease release to remodel the ECM has been more thoroughly studied in the context of cancer metastasis. Cathepsin B, for instance, has been reported as an integral player in the tumor microenvironment as its secretion and activity was localized around tumor cells to degrade components of the basement membrane to aid in angiogenesis (S.-H. Chang et al. 2009). Indeed, remodeling of the

163

ECM has been reported to be orchestrated by a variety of other proteases aside from cathepsin B. Unique here is cathepsin B’s ability to propel or initiate this network of protease activation in the area where activated cathepsin B is secreted. For example, cathepsin B has been known to activate MMPs in the tumor microenvironment without changes in the levels of MMP itself through cathepsin B-led degradation of TIMPs, the endogenous inhibitors of MMPs (Kostoulas et al. 1999). Cathepsin B may be playing a similar role in axons to robustly enhance GAG-CSPG degradation by coordinating the activation of a variety of other proteases such as MMP-2 (Zuo et al. 1998) or tPA

(Siconolfi and Seeds 2001) whose enhanced secretion was not identified in our assays, but have been reported by others to exert similar ECM remodeling roles by peripheral axons.

Another interesting facet of cathepsin B is its connection to lysosomes. Cathepsin B, along with other family members of this family including cathepsins L, S, and Z, have been well- known as lysosomal proteases that aid in proteolytic degradation within these specialized structures (Olson and Joyce 2015). Indeed, staining for lysosomes using Lamp1 revealed colocalizion of the majority of cathepsin B within these structures which are found not only in the somata of peripheral neurons, but throughout their axons and growth cones as well.

Work by the Nixon group has highlighted the importance of lysosomal function in axon growth and homeostasis. For instance, inhibition of lysosomal proteolysis through de- acidification or cathepsin inhibition resulted in an accumulation of autophagic vacuoles in bulbous swells, and ultimately, dystrophic axons (S. Lee et al. 2011). Additionally, cathepsin B activity as a part of lysosomal function and macroautophagy (autophagy, hereafter) are inextricably linked. Recent studies have come to better appreciate this

164 connection as blockage of autophagosome-lysosomal fusion led to diminished lysosomal activation (Zhou et al. 2013) and inhibition of lysosomal activation in the face of sustained autophagy resulted in neuritic dystrophy (Bordi et al. 2016). The link between autophagosome-lysosomal fusion and axonal dystrophy may be an intriguing clue to the identity of the heretofore unidentified cycling vesicles first discovered by Tom et al. in dystrophic growth cones trapped by a gradient of CSPGs (Tom et al. 2004).

How cathepsin B itself is regulated downstream by PTPσ modulation is broached by western blot analyses in our study. While it seems as though ISP treatment enhances the secretion and abundance of cathepsin B extracellularly as identified through western blot immunostaining of conditioned media, the protein levels of intracellular cathepsin B seem to be stable. Intriguingly, we found a decrease in intracellular protein levels of cystatin B, an endogenous inhibitor of cathepsin B, instead. Indeed, immunostaining of cystatin B revealed broad expression throughout the axons and somata of DRGs. Normally, cytosolic cystatin B protects cell constituents from digestion from cathepsins leaking from lysosomes

(Y. Ma et al. 2017), especially cathepsins B, H, and L (Chung et al. 2016).

Although genetic knockout of cystatin b has been implicated in Unverricht-Lundborg type progressive myoclonus epilepsy (Eldridge et al. 1983; Joensuu et al. 2014), little is known about the roles cystatin b may play in CNS homeostasis or how it is regulated. Some burgeoning clues, however, suggests intriguing connections between its inhibition of cathepsin B and the protease-digested ECM phenotype our study reports. For example, in the developing rat cerebellum, cystatin B expression is undetectable in radial glial cells as

165 visualized through brain lipid-binding protein (BLBP) immunostaining at seven days; however, by 90 days BLBP and cystatin B staining colocalizes along the somata and axonal projections (Riccio et al. 2005). Perhaps this correlates with an increase of cathepsin B and other protease activity during development to remodel the ECM. In another example, knocking out cystatin B in mice manifests as progressive myoclonus epilepsy that may be caused by a diminished number of GABAergic terminals that is not due to a decline in the number of inhibitory interneurons itself (Joensuu et al. 2014). This is especially intriguing because aggrecan and other CSPGs are predominantly found in perineuronal nets encasing GABAergic somata and their proximal dendrites throughout the

CNS and have been known to play a role in stabilizing synapses during the critical period of development (Pizzorusso et al. 2002). Whether knocking out cystatin B leads to enhanced cathepsin B and other protease activation that attenuates the development of perineuronal nets around these GABAergic terminals will need to be further studied.

But how is regulation of this inhibitor accomplished downstream PTPσ modulation? We have performed initial protease characterization of PTPσ null mice to find that protease activity was enhanced in the conditioned media of PTPσ knockout DRG neurons. This may be due to enhanced cathepsin B activity in the axon as pixel intensity of cathepsin B from immunostaining was found to be increased in the axons of PTPσ knockout DRGs compared to control wild type. However, it is still unknown whether cathepsin B activity is linked to a decrease in cystatin B in the PTPσ knockout. Further characterization of cystatin

B in the mutant background will need to be performed to begin elucidating how PTPσ modulates cathepsin B in neurons.

166

Cystatin B has additionally been found to co-immunoprecipitate with a cluster of proteins from rat cerebellar lysate including RACK1, β-spectrin, and neurofilament light polypeptide (NF-L), which are three proteins well-known for their roles in cytoskeletal dynamics and cell migration (Di Giaimo et al. 2002). To begin, RACK1 is a multifunctional scaffolding protein that has been characterized to regulate epithelial to mesenchymal transition in many cancer cells leading to cell motility and migration through modulation of the focal adhesion site (Duff and Long 2017). Specifically, siRNA knockdown of RACK1 has been found to suppress β1-integrin and MMP-2 and 9 expressions in human glioma cell lines (Lv et al. 2016). How RACK1 binding with cystatin B alters cathepsin B activity will need to be further studied. At focal adhesion sites,

RACK1 additionally integrates signaling from transmembrane receptors to the cytoskeleton. Work on PTPµ has found that this receptor complexes with PKCdelta through RACK1 to enhance neurite outgrowth of retinal ganglion cells (Rosdahl et al.

2002). Further work will be needed to determine whether PTPsigma may also be associated with or modulating RACK1 in a similar capacity with cystatin B.

β-spectrin and NF-L are cytoskeletal proteins that are highly integrated with cell signaling processes including cell adhesion and mobility through Rac and Rho GTPase activation

(Rui Zhang et al. 2013). Rac and Rho GTPases have been well described to coordinate multiple signaling pathways to orchestrate dynamic reorganization of the actin and microtubule cytoskeleton to affect axon outgrowth at the growth cone. How cystatin B binding to β-spectrin and NF-L may be affecting cytoskeletal dynamics and whether this process is regulated by PTPσ will require further study.

167

The PTPσ sister receptor, LAR, has additionally been found to complex with Trio at focal adhesion sites (Debant et al. 1996) which is especially intriguing because Trio regulates axon guidance through remodeling and assembly of the actin cytoskeleton (Bateman and

Van Vactor 2001) and itself contains spectrin-like repeat domains that confer binding to components of the cytoskeleton (Djinovic-Carugo et al. 2002). As LAR and PTPσ share structural similarities and overlapping functions, it may prove fruitful to ask whether cystatin B modulation, enhanced exocytosis, and other dynamic axonal changes through

PTPσ may be orchestrated through Trio at the focal adhesion site.

While cathepsin B activity has been robustly found in our in vitro DRG experiments, a larger question looms on whether the same mechanism underlies enhanced serotonergic sprouting we have previously identified in vivo. Lang et al. found that ISP treatment of spinal cord contused rats enhanced serotonergic sprouting caudal to the lesion site, and that the extent of serotonergic sprouting correlated with a return in a variety of functions.

Moreover, gains in coordinated locomotor function seem to be dependent on serotonin as treatment with a serotonin antagonist, methysergide, ablated differences in function between ISP-treated and vehicle control groups. This is not the only instance where serotonergic sprouting coincides with functional improvements following spinal cord injury

(Alilain et al. 2011; D. Kim, Zai, Liang, Schaffling, Ahlborn, and Benowitz 2013a; Freria et al. 2017). Indeed, serotonin neurons have been identified as the neuronal subtype with projecting axons that most robustly sprout following a variety of CNS injuries (Hawthorne et al. 2011; Jin et al. 2016; Kajstura et al. 2017) in response to a wide range of treatments including pharmacological (Pearse et al. 2004; D. Kim, Zai, Liang, Schaffling, Ahlborn,

168 and Benowitz 2013b), stem cell (X. He et al. 2012), rehabilitative (Gonzalez-Rothi et al.

2015), and chondroitinase injections (Alilain et al. 2011; Bartus et al. 2014).

The mechanisms underlying serotoninergic sprouting and how this may be different than other neuronal subtypes is an ongoing source of interest (Hawthorne et al. 2011). In our study, we observed an intriguing observation that may suggest one method by which serotonergic sprouting may occur. Immunostaining for GAG-CSPGs (WFA) and serotonin axons (5-HT) following spinal cord injury showed a reciprocal patterning of decreased

GAG-CSPG expression where serotonin was highly expressed in ISP-treated spinal cord sections. In contrast, immunostaining in the vehicle control spinal cords showed a decrease in serotonergic sprouting and an increase in GAG-CSPG expression. The ISP-treated serotonin/GAG-CSPG pattern fit in an almost “lock-and-key” manner suggesting that serotoninergic-led protease digestion of the surround GAG-CSPG may be taking place.

Although we were unable to extract enough adult serotonergic neurons to meaningfully assess their protease activity in vitro, cathepsin B and serotonin immunostaining were performed in the same spinal cord injured animals to reveal that cathepsin B highly colocalized with serotonergic neurons. The idea that axons may utilize proteases to barrel through an inhibitory ECM is not new; however, it is compelling to consider whether serotonergic neurons are especially effective in this process.

While studying protease release by growth cones have been difficult due to the large amount of growth cones necessary for biochemical analyses, research into dendritic spine plasticity have yielded relevant results. Recently, Bijata et al. have identified that

169 simulation of the 5-HT7 receptor leads to selective MMP-9 activity resulting in local ECM remodeling, cell adhesion receptor proteolysis, and dendritic spine elongation (Bijata et al.

2017). The 5-HT7 receptor is especially interesting in the context of axon regeneration as its transient upregulation in developing hippocampal and peripheral sensory neurons correlates with a time of axonal outgrowth (Gaspar et al. 2003). Additionally, the 5-HT7 receptor has been previously linked with neurite outgrowth as embryonic day 18 hippocampal cultures treated with a specific inhibitor against 5-HT7 showed decreased neurite length in vitro (Rojas et al. 2014). 5HT-7 regulation of neurite outgrowth has additionally been previously linked to its activation of the Gα12 subunit, which in turn regulates cytoskeletal dynamics through RhoA and Cdc42 to enhance neurite outgrowth in embryonic day 18 hippocampal neurons in vitro (Kvachnina et al. 2005). Work from Bijata et al. links local protease activity for the first time to a serotonergic receptor. In the same vein, Padamsey et al. also found that MMP-9 activity was increased after stimulation of hippocampal spines to result in ECM remodeling and long lasting enlargement of the dendritic spine head. Importantly, MMP-9 activity was initiated and sustained by cathepsin

B activity which cleaved TIMP-1, the endogenous inhibitor of MMP-9 (Padamsey et al.

2017). Padamsey et al. furthermore found that this cascade of protease activity was accomplished only after back-propagating action potentials induced lysosomal calcium release that triggered lysosomal fusion with the plasma membrane and subsequent exocytosis of cathepsin B. In this way, cathepsin B release and amplified protease activity through MMP-9 allowed for spinal plasticity induced by activity. Whether a similar process occurs in the growth cones of sprouting serotonergic neurons following PTPσ modulation will need to be further explored.

170

4.1.2 Summary and Discussion of Modulation of Protein Tyrosine Phosphatase

Receptor Sigma Leading to Enhanced MMP-2 Activity in Oligodendrocyte

Progenitor Cells

In Multiple Sclerosis-like EAE, another model of CNS injury where upregulated CSPGs inhibit regenerative processes, ISP also yielded functional locomotor recovery. Underlying this recovery is ISP-induced enhanced remyelination as assayed through myelin staining of sectioned spinal cord tissue, western blot immunostaining of MBP, and electron microsopy of myelin sheaths surrounding axons of the spinal cord. Indeed, ISP-enhanced remyelination was repeated in an LPC as well as an ex vivo cerebellar slice model of demyelination.

Closer examination of LPC-induced demyelinated lesions of the dorsal aspect of the spinal cord revealed increased immunostaining of GAG-CSPGs (CS-56) and the protein core of aggrecan, Cat-301 (Fryer et al. 1992). Interestingly, immunostaining revealed that CSPGs predominantly occupied the lesion epicenter that has become demyelinated from the LPC injection into the spinal cord. As ISP is given, however, the CSPG content of the lesion epicenter decreased as MBP immunostaining increased overtime. Specifically, we surmise that this MBP infiltration is due to remyelination instead of MBP sparing as LPC lesion sizes of vehicle control and that ISP-treated cords were not significantly different at seven days post injection. Cerebellar slices treated with ISP additionally showed enhanced rates of remyelination concomitant with a decrease in CSPG staining compared to vehicle control.

171

Furthermore, immunostaining of different glia revealed increased reactive astrocytes in the lesion epicenter (GFAP) and a correlating increase of reactive astrocyte-secreted versican as previously identified by Keough et al and others. ISP treatment, however, reduced the amount of versican immunostaining presumably through protease-led digestion. The CSPG source in LPC-injected models of demyelination may very well be predominantly secreted by reactive astrocytes as Iba1 immunostaining was not significantly different between vehicle control and ISP-treated LPC spinal cords.

Intriguingly, this may be different in EAE models where aggrecan core protein seems to be localized within Iba1-enriched demyelinated plaques. Indeed, reactive macrophages have been found to secrete versican and other CSPGs in vitro (Makatsori et al. 2003; Lindholm et al. 2005). PTPσ has additionally been linked to the onset of EAE progression putatively by T-cells (Ohtake et al. 2017). How ISP treatment affects PTPσ signaling in invading macrophages will need to be further examined with further investigation although recent studies have begun to paint a compelling role for PTPσ in immune cell signaling.

Our in vivo analyses of CSPGs in demyelinated models in mice nonetheless confirm findings that CSPGs arising from reactive astrocytes and a broadly inflamed CNS environment play a potently inhibitory role in remyelination. One way this inhibition could be relieved is through protease-led digestion of the CSPG protein itself, which was suggested through progressively decreased immunostaining of CSPGs in a variety of demyelinated models. Indeed similarly to peripheral axons, modulation of PTPσ in pure cultures of oligodendrocyte progenitor cells induced enhanced protease activity. This was

172 made evident by oligodendrocyte progenitor cells cultured on the Silver spot assay with

ISP. Surprisingly, oligodendrocyte progenitor cells migrating past the CSPG gradient also left “shadows” of digested GAG-CSPG.

Certainly, other peripheral myelinating glia have additionally utilized this secretory protease strategy to aid in myelination. For example, Schwann cells require BACE1 to remyelinate peripheral nerves (X. Hu et al. 2015) and cathepsin E has been noted to promote Schwann cell migration and myelination (Lutz et al. 2013). Moreover, MMP-2 and 9 digestion of dystroglycan in a finely compartmentalized manner are also required for normal Schwann cell functions including differentiation and myelination (Court et al.

2011; Y. Kim et al. 2012). Among olfactory ensheathing glia, which myelinate olfactory axons in the olfactory epithelium and nerve, MMP-2 has additionally been identified as important in their migration along inhibitory ECM substrates in vitro (Gueye et al. 2011), in degrading neurocan (Yui et al. 2014), and in promoting axon regeneration when transplanted into a contused rat spinal cord (Pastrana et al. 2006).

In our own biochemical and western blot analyses, we have identified MMP-2 as the chief protease responsible for GAG-CSPG degradation. Prior to our studies, MMP-2 and MMP-9 activity has been weakly linked to PTPσ in a trabecular meshwork cell line (Zaiden and

Beit-Yannai 2015). Zaiden et al. found that overexpressing PTPσ in a human-derived normal trabecular meshwork cell line was itself able to increase the levels of the pro-form of both gelatinases without altering the active forms of these proteases. This study, unfortunately, lacked vital details concerning the type of ECM substrate used to culture

173 these cells and specific tools to modulate the overexpressed receptor, which obfuscates the interpretation of exactly how PTPσ modulates gelatinase activity in this cell system.

In our own studies, MMP-2 seems to be expressed throughout the somata of oligodendrocyte progenitor cells as well as the processes. It is interesting to contrast this with the identification of cathepsin B in peripheral neurons. While DRGs have also been noted to express MMP-2 to degrade CSPGs (Zuo et al. 1998), cathepsin B expression was not detectable through immunoblotting of ISP-treated oligodendrocyte progenitor cell conditioned media although this does not exclude its secretion, or the secretion of other proteases unidentified by our immunoblotting, by oligodendrocyte progenitor cells. One possible explanation of this may be that high levels of cathepsin B expression seems to be preferentially expressed in neurons although other glia have also been shown to express cathepsin B mRNA (Petanceska et al. 1994).

MMP-2 expression by ISP-treated oligodendrocyte progenitor cells also preferentially digested aggrecan over laminin as seen through spot assay quantification and western blot analyses. This is an important distinction because while aggrecan is inhibitory to oligodendrocyte progenitor cell mobility, homeostasis, and maturation, laminin promotes cellular mobility. For instance, oligodendrocyte progenitor cells have been found to migrate, through “crawls” and “jumps” along laminin-rich blood vessels from their progenitor domains to myelinate axons throughout the CNS during development (Tsai et al. 2016). This preferential mode of aggrecan digestion by MMP-2 has also been

174 implicated in DRGs which use this strategy to regenerate along the adult periphery (Zuo et al. 1998).

Similar to our characterization of ISP-treated peripheral neurons in vitro, cultured oligodendrocyte progenitor cells seem to secrete MMP-2 into the conditioned media as ISP treatment of oligodendrocyte progenitor cells increased MMP-2 expression in the media while depleting intracellular MMP-2 in oligodendrocyte progenitor cell lysate. The addition of an exocytosis inhibitor, Exo1, additionally decreased ISP-induced CSPG digestion.

MMP-2 activity is necessary for oligodendrocyte progenitor cell migration past the high aggrecan barrier in the spot assay as well as its ability to remyelinate following demyelinating LPC insult in a cerebellar slice model as simultaneous ISP and MMP-2 inhibitor treatment and knock down of MMP-2 by shRNA delievered by lentiviral particles severely reduced these events. MMP-2, of course, may be acting in concert with other proteases to execute these effects. Like cathepsin B, MMP-2 activity may be acting as a hub to orchestrate the activation of a network of proteases perhaps including MMP-9 (Uhm et al. 1998; Larsen et al. 2003) and ADAMTS-4 (Pruvost et al. 2017), which among others, have been implicated in remyelination.

In addition to enhanced migration, we also observed that specifically inhibiting MMP-2 produced deleterious effects in normal oligodendrocyte progenitor cell homeostasis. It has been previously established, and confirmed in our studies, that CSPGs inhibit oligodendrocyte progenitor cell proliferation, differentiation, process outgrowth, and survival (Harlow and Macklin 2014; Keough et al. 2016), and in this way, ultimately

175 inhibits remyelination. The addition of an MMP-2 inhibitor alone was able to increase apoptosis in a population of oligodendrocyte progenitor cells cultured on laminin and low aggrecan in the presence of demyelinating LPC. To be clear, the addition of the MMP-2 inhibitor alone, without LPC treatment, was unable to increase apoptosis over vehicle control. Even the addition of ISP was unable to significantly rescue MMP-2 inhibitor- induced oligodendrocyte progenitor cell apoptosis suggesting the important role of MMP-2 in oligodendrocyte progenitor cell survival following insult. Additionally, MMP-2 seems to be important in oligodendrocyte progenitor cell differentiation. As oligodendrocyte progenitor cells mature, they begin to express MBP and expand the territory of their processes in sheath-like projections (Baumann and Pham-Dinh 2001; Bergles and

Richardson 2015). Treating oligodendrocyte progenitor cells on laminin and low aggrecan substrates with the addition of an MMP-2 inhibitor reduced the volume of these projections even though MBP expression was still evident. The addition of the MMP-2 inhibitor alone was able to inhibit process outgrowth potently compared to vehicle control.

MMP-2 may be exerting other biological processes independent of CSPG degradation to enhance oligodendrocyte progenitor cell maturation and survival. While CSPGs themselves potently inhibit oligodendrocyte progenitor cell homeostasis to ultimately prevent the ability of oligodendrocyte progenitor cells to remyelinate in their mature forms, proteases have also been implicated in affecting oligodendrocyte progenitor cell homeostasis as well, as their substrates include other proteins aside from those in the ECM. Regarding migration, for example, tPA has been implicated in enhancing oligodendrocyte progenitor cell migration along blood vessels during development in a chemotaxic manner by binding,

176 in a catalytically independent process, to the epidermal growth factor (EGF) receptor on oligodendrocyte progenitor cells in order to more efficiently sense the chemokine fibroblast growth factor (FGF)-2 in vitro (Leonetti et al. 2017). Proteolytic processing also affects oligodendrocyte progenitor cell proliferation. MMP-9 has been found to decrease oligodendrocyte progenitor cell proliferation following injury through shedding of membrane-bound NG2 and the NR1 subunit of the NMDA receptor, which itself has been implicated in oligodendrocyte progenitor cell maturation (H. Liu and Shubayev 2011).

MMP-12 has also been observed to enhance oligodendrocyte process extension and lineage progression (Larsen and Yong 2004) possibly through a chemokine-processing or ADAM-

10 and 17 Notch receptor cleaving manner (Pan and Rubin 1997). Thus, while CSPGs potently inhibit oligodendrocyte progenitor cell lineage progression and ultimately processes associated with promoting remyelination through PTPσ, proteases regulate all these processes including digestion of the CSPG itself.

4.2 Study Implications

Clearly, CSPGs and subsequent PTPσ signaling play an important role in CNS injuries.

Moreover, our findings highlight PTPσ as an attractive pharmacological target for neurodegenerative diseases where CSPGs play a role in inhibiting regenerative processes.

In our own work, we have found that systemic ISP injections following demyelinating models of EAE and LPC injections into the spinal cord have accelerated remyelination through relief of CSPG inhibition. Coupled with previous findings which found that ISP enhances axon regeneration or sprouting (Lang et al. 2015), systemic ISP treatment could

177 potentially promote functional recovery following chronic stages of Multiple Sclerosis where oligodendrocyte progenitor cell apoptosis is rampant and CSPG load in demyelinated plaques are abundant, resistant to remyelination, and impede voluntary motor functions (Franklin and ffrench-Constant 2008; Lau et al. 2012). In chronic cases of spinal cord injury, too, remyelination may occur spontaneously (Hermelinda et al. 1998), but not nearly enough to restore fast saltatory conduction to newly regenerated axons due to a dearth in necessary growth factors to promote the process and death and inhibition of oligodendrocyte progenitor cells (H.-F. Wang et al. 2017). Recent work from the Zhigang

He lab has further highlighted the limitations of axon regeneration-promoting therapies alone as newly regenerated retinal ganglion axons fail to restore sufficient electrophysiological function without proper myelination (Bei et al. 2016). Targeting PTPσ with ISP therefore presents a multi-pronged approach to promote not only axon regeneration, but also to restore function in chronic stages of disease progression.

As CSPGs are implicated in a variety of CNS injuries, it stands to reason that systemic ISP treatment may have clinical relevance to these other injuries as well. For example, CSPGs are upregulated in the lesion epicenter resulting from traumatic brain injury (Yi et al. 2012) and ischemic reperfusion injuries after strokes (Huang et al. 2014) due to reactive astrocytes and formation of a glial scar. Thus, modulation of PTPσ would presumably promote axonal sprouting into the infarcted or lesioned sites and aid with oligodendrocyte progenitor cell survival while depleting local CSPG load.

178

Aside from CNS injuries, CSPGs are beginning to gain greater recognition in their more nuanced roles in neurodegenerative diseases. CSPGs, specifically their GAG chains surrounding select neurons in structures called perineuronal nets, are increasingly studied for their effects on the control of plasticity and memory in adulthood outside of developmental pruning during the critical period. Chondroitinase ABC injections into different mouse models of Alzheimer’s disease, for instance, has yielded improvements in neural plasticity and memory function. Specifically, P301S tau mutants, which gradually express neuronal pathology similar to late stages of human Alzheimer’s disease, show complete loss of novel object memory that is reversible following bilateral injections of chondroitinase ABC (S. Yang et al. 2015). Moreover, APP mutants, which develop deficient memory function as they develop amyloid plaques, preserve the ability to differentiate contextual memory and LTP following an injection of chondroitinase ABC into the hippocampus (Végh et al. 2014). While further studies will be needed to assess the role of CSPGs binding to PTPσ between hippocampal neurons and perineuronal nets, studies thus far suggest that modulation of PTPσ in Alzheimer’s disease may prove fruitful especially as PTPσ is involved in synaptogenesis (Um and J. Ko 2013). CSPG-laden perineuronal nets, furthermore, may have significant impact in the spinal cord in the context of neurodegenerative diseases. PTPσ, for example, has been shown to be expressed in motor neurons (Sajnani-Perez et al. 2003) where they may negatively impact these neurons in the context of amyotrophic lateral sclerosis (ALS). In SOD1 mutant mice, which models ALS, CSPGs have been found to be upregulated in the ventral motor neuron pools of the spinal cord compared to age-matched controls (Mizuno et al. 2008). While perineuronal nets typically encapsulate a subset of the soma of these ventral motor neurons,

179

SOD1 mutants seem to display up-regulation of versican and phosphacan, which have been characterized to peak at the early symptomatic stage. Notably, progressive loss of motor neurons correlates with astrogliosis, which may partially explain the upregulation of select

CSPGs (Shijo et al. 2017). The connection between ALS and CSPGs-PTPσ is indeed compelling and will require further studies to see if modulating PTPσ with ISP would further improve symptoms through the attenuation of CSPG-induced inhibition.

Presented here is the novel finding linking PTPσ and protease release in both peripheral axons and oligodendrocyte progenitor cells, but what are the implications of this work moving forward? Broadly, these studies may shed light on one mechanism by which CNS cells may selectively relieve CSPG inhibition in their immediate environment. This punctate protease release is in stark contrast to the “protease storms” released following insult to the CNS that positively feeds back into inflammatory processes to promote cell death and further injury. Work from this thesis further highlights the nuanced role of selective protease release in important processes such as promoting or restoring cell mobility. While this phenomenon has been extensively studied in the context of cancer metastasis (Olson and Joyce 2015), selective protease release to remodel the cell’s immediate environment is becoming more appreciated in axon regeneration and especially dendritic spine plasticity, which can be abundantly assayed in vitro, and may have additional implications for how this receptor is involved in plasticity beyond a post-injury context, specifically in synaptogenesis.

4.2.1 PTPσ as a Switch in Plasticity

180

Through studies involving chondroitinase ABC injections, which digest GAG chains present in perineuronal nets, into the naïve adult brain we have gleaned the importance of

CSPGs in maintaining both structural as well as activity-dependent plasticity. In a classic example of CSPGs’ impact on activity-dependent plasticity, Pizzorusso et al. showed that injections of chondroitinase ABC in the adult visual cortex initiated a period of experience- dependent plasticity in adulthood, which resulted in an ocular dominance shift. Usually this period of experience-dependent plasticity occurs during neonatal development in a critical period that typically closes with the development of perineuronal nets from P16-P21 in mice (Gogolla et al. 2009). Further exploration of this phenomenon has showed that chondroitinase ABC digestion of GAG chains of the perineuronal net increases the number of synaptic puncta with resulting functional changes on these neurons’ electrophysiological characteristics (Pyka et al. 2011; Favuzzi et al. 2017). Functionally, chondroitinase ABC injections have been linked to alterations in LTP, and ultimately, functional memory tasks

(Gogolla et al. 2009; Banerjee et al. 2017). Together, this suggests that CSPGs enriched in perineuronal nets play an important role in limiting neural plasticity including affecting synaptogenesis.

The effects of CSPGs on synaptogenesis may be, in part, due to protein binding and signaling events instigated through PTPσ. The RPTP family, which includes LAR and

PTPδ in addition to PTPσ, has been described as a presynaptic hub-like receptor that organizes synapse complexes by binding to multiple postsynaptic adhesion partners to propel synapse self-assembly (Takahashi and Craig 2013). PTPσ itself has been found to

181 cooperatively bind to cell adhesion and synaptogenesis-promoting proteins such as

Tropomyosin receptor kinase (Trk)-C (Takahashi et al. 2011), neurotrophin-3 (NT-3) (Han et al. 2016), and netrin-G ligand-3 (NGL-3) (Woo et al. 2009; Kwon et al. 2010).

Importantly, PTPσ initiates synaptic organization in excitatory synapses first by mediating cell-cell adhesion at burgeoning synapses in order to stabilize this nascent structure

(Takahashi and Craig 2013). This is accomplished in part through binding with LAR and

Slitrk in a complex as revealed through crystallography (Um et al. 2014). Through interactions with TrkC and related proteins, PTPσ then initiates bi-directional synaptic differentiation that locally recruits the necessary synaptic machinery including scaffolds and other signaling proteins (Dunah et al. 2005; Takahashi and Craig 2013).

Recent studies are beginning to highlight the emerging role of the neutrophin receptor,

TrkC, in synapse organization. Initially found through a cDNA screen for synaptogenesis- promoting proteins, TrkC and PTPσ binding, with the addition of NT-3 modulation, specifically induces pre- and postsynaptic differentiation in a kinase-independent manner to increase vesicle recycling and NMDA receptor and PSD-95 clustering (Takahashi et al.

2011; Naito et al. 2017). Notably, TrkC binds specifically to PTPσ and not to LAR or

PTPδ and requires PTPσ-specific signal transduction to initiate such differentiation

(Takahashi et al. 2011). Additionally, TrkC knockdown through shRNAs has been shown to reduce PSD-95 puncta and altered post-synaptic miniature excitatory frequency (Naito et al. 2017) while PTPσ knockout animals have been found to exhibit deficits in paired pulse facilitation and long term potentiation (Horn et al. 2012).

182

As the cognate receptor of CSPGs, as well as an important synaptogenesis hub receptor, how is PTPσ able to coordinate between both axonal growth and synaptogenesis? From our understanding of CSPGs and this unique receptor’s involvement in both synaptogenesis and axon regeneration, we propose that PTPσ acts as an ON and OFF switch for axonal plasticity that exchanges axonal outgrowth for synaptogenesis or vice versa. Indeed, Coles et al. have found that PTPσ binds to both heparan sulfate and TrkC with widely different results (Coles et al. 2014). When heparan sulfate outcompeted TrkC binding to PTPσ in vitro, for example, neurite outgrowth was favored over synaptogenesis and the opposite occurred when TrkC was bound instead.

This idea of PTPσ as a switch for plasticity between axon extension and synaptogenesis has additional implications following injury in adulthood and may help to explain how regeneration may occur or falter in this context. PTPσ binding to upregulated CSPGs in an

OFF configuration following injury has been shown to promote adhesion of the regenerating growth cone to the substrate gradient to stop axonal outgrowth (Tom et al.

2004; Lang et al. 2015). Growth cones exposed to particularly NG2 proteoglycans have been found to additionally stabilize dystrophic growth cones into irregular synaptic structures (Filous et al. 2014) that may persist for decades (Ruschel et al. 2015). It is this adhesiveness CSPGs impart through PTPσ that may be helping to initially stabilize any burgeoning synapse. Certainly, presynaptic PTPσ binding to postsynaptic adhesion- promoting proteins including NGL-3, NT-3 (Naito et al. 2017), and N-cadherin (Siu et al.

2006) has been previously characterized as necessary to synaptogenesis.

183

However, PTPσ in an ON configuration when axon growth conducive substrates outcompete CSPGs or when chondroitinase ABC is injected results in a switch from stabilizing the growth cone to encouragement of its axon extending program. Selective protease release as characterized in our work may additionally promote this process.

Protease secretion at the growth cone would further encourage axon growth through digestion of inhibitory CSPGs in the immediate surrounding ECM and possibly through digestion of adhesion proteins as well. One may imagine that once a regenerating growth cone meets its postsynaptic target, such as a soma enmeshed in a perineuronal net, the prevalence of CSPGs again switches its receptor to an OFF mode and protease secretion is additionally halted. Otherwise, continuous protease secretion and the resulting digestion of adhesive proteins would potentially interfere with synaptogenesis at this critical stage.

Some level of nuance must be applied to this model to encapsulate a whole spectrum of homeostatic plasticity, however. For instance, enough flexibility would need to be programmed into the system to adjust synaptic strength as governed through molecular and structural changes in order for neuronal networks to remain plastic. Some level of protease secretion through modulation of PTPσ could help in this endeavor as protease secretion has long been shown to remodel synapses (Bajor and Kaczmarek 2013). Valenzuela et al. has additionally identified ADAMTS-4 immunostaining colocalizing within perineuronal nets that correlate with specifically ADAMTS-4 cleaved brevican, a lectican CSPG. In this way, the authors propose that selective secretion of ADAMTS-4 and correlative cleavage of brevican allows for ECM remodeling in perineuronal nets that may allow for greater flexibility of structural plasticity (Valenzuela et al. 2014). Indeed, brevican has recently been found to adjust the tuning of the fast spiking, perineuronal net-enmeshed

184 parvalbumin-positive interneurons in the hippocampal CA1 region through clustering of

AMPA receptors and voltage-gated potassium channel type 3.1b (Favuzzi et al. 2017).

Whether protease secretion and subsequent remodeling of specific CSPGs of the perineuronal net for homeostatic plasticity is related to CSPG-PTPσ interactions will need to be further investigated.

Similarly, PTPσ may be acting as a switch in oligodendrocyte progenitor cells in the post- injury environment. Following traumatic injury, regenerating axons require remyelination to efficiently undergo salutatory conduction. In the presence of upregulated CSPGs where neither PTPσ-bearing oligodendrocyte progenitor cells nor growth cones are mobile or regenerating respectively, it may make sense for oligodendrocyte progenitor cells to refrain from expending energy to differentiate when there is no de novo axonal structure to remyelinate. In such a way, PTPσ may be acting as a switch that coordinates multiple cell types to ultimately orchestrate an effective regeneration program. Modulation of PTPσ to an ON mode could potentially encourage simultaneous regeneration of axons and mobilization of oligodendrocyte progenitor cells that proceed to mature and remyelinate with the aid of protease release. While there still remains many questions relating to how

PTPσ signals to effect so many different outcomes, the idea of PTPσ as switch in multiple cell types coordinating a variety of physiological outcomes seems an attractive theory to further explore.

4.3 Study Limitations

185

As with any study, there are limitations we wished we had additional time and resources to address. To begin, while the use of ISP as a synthetic peptide modulator of PTPσ is arguably the most powerful tool utilized in our studies, it is not free of its own pitfalls.

First, the peptide is reliant on TAT-mediated cell permeation to target the intracellular domain of PTPσ. While TAT efficiently delivers ISP, it has been known to induce some effects of its own such as increased endocytosis and slight protease activity (Ju et al. 2009).

To rule out TAT-mediated effects in our peptide, we have tested a PRR5-ISP construct that utilizes the internalization sequence PRR5 instead of TAT which yielded the same results including enhanced axonal outgrowth and protease digestion in our spot assay

(unpublished). Additionally the scrambled control of our peptide, which comprises the

TAT domain and a scrambled version of the same amino acids used in ISP, did not elicit enhanced axon or oligodendrocyte crossing past a CSPG barrier.

One way by which we have made progress to reduce the reliance of peptide modulation of

PTPσ was to utilize genetic methods. Two constitutive PTPσ knockout mouse lines exist

(Elchebly et al. 1999; Rotin et al. 1999) and one in particular has previously been characterized to display enhanced axon outgrowth past a CSPG gradient (Shen et al. 2009).

We utilized the PTPσ null mutant from Shen et al., and with initial characterization of protease activity, have found that preliminary spot degradation assays utilizing extracted

DRGs from wild type, heterozygous PTPσ, and homozygous PTPσ knockout animals suggested a dose response effect of GAG-CSPG degradation with knockout of PTPσ.

However, experiments with the PTPσ knockout animals include their own drawbacks including difficulty in breeding in this particular mouse line due to multiple health issues

186 limiting perinatal viability as well as fecundity in adulthood. To bypass difficulties inherent in the genetic mouse lines, we could have generated siRNA constructs to knock down

PTPσ. Unfortunately, these genetic methods would have limited the number and types of experiments we could perform considering the paucity of peripheral neurons that could be extracted at each time. Interestingly, however, we did detect increased cathepsin B immunoreactivity in homozygous PTPσ knockout DRG neurons compared to BALB/C wild type controls. The increased presence of cathepsin B, especially along the axons of

DRGs may be important in related enhanced autophagosome-lysosomal fusion (Martin et al. 2011) and the ability of these axons to regenerate better than controls (Shen et al. 2009).

Future studies will be needed to better link cathepsin B activity in lysosomes to the enhanced axon regeneration phenotype typical of this mutant.

Additionally, we have characterized ISP action on PTPσ as peptide “modulation” of the receptor. What, exactly, does modulation denote at the molecular level? ISP was specifically designed to target the intracellular wedge domain of PTPσ that regulates the conformation of PTPσ; however, specific in vitro experiments would be required to fully confirm this hypothesis. To date, we have confirmed that ISP is able to bind to and pull down PTPσ in vitro (Lang et al. 2015) and we have performed in silico ISP-PTPσ simulations that yield favorable Kd values for ISP binding to this wedge domain

(unpublished). To validate that ISP binds to the wedge domain of PTPσ in a cell system, we could potentially produce constructs of PTPσ that are missing the sequences responsible for translating the wedge domain and transfect this construct into a cell line to see if

187 treatment of ISP would still enhance migration across a gradient of CSPGs relative to wild type PTPσ transfected cells.

Unfortunately our experiments also exclude meaningful inquiry into how ISP is regulating

PTPσ. We have previously linked ISP treatment to enhanced ERK signaling (Lang et al.

2015), but how this is achieved at the level of the receptor is still unclear. To begin addressing this issue, it may be fruitful to question the conformation state of PTPσ to help answer whether ISP is promoting HSPG-like axon growth signaling through dimerization of the receptor. HSPG binding to PTPσ has additionally been thought to decrease the receptor’s phosphatase activity (Shen 2014). In contrast, CSPG binding to the receptor promotes monomer conformation of PTPσ and putatively enhances phosphatase activity

(Shen 2014). Recently, the Tremblay lab has developed a split luciferase system that reports the conformation state of PTPσ by synthesizing constructs that have, attached to the intracellular end of the receptor, either the C-terminal or N-terminal ends of the luciferase enzyme (Wu et al. 2017). These constructs are then transfected into an abundantly available cell line. If dimerization of two PTPσ receptors occurs, then the split ends of the luciferase enzyme are also able to bind and produce bioluminescence that could be measured in real time. We can take advantage of this system to help determine whether ISP is promoting signaling of the receptor through promoting dimerization, as when HSPG is present. If this is the case, then luciferase activity should be increased when ISP is applied to the cell line expressing the split luciferase system relative to vehicle control.

188

Determining the conformation state of receptor is useful to help explain how phosphatase signaling may occur. Using the split luciferase system, the Tremblay lab has additionally linked increased phosphatase activity with CSPG binding and monomerization of PTPσ.

When HSPGs were added and dimerization increased, however, phosphatase activity decreased. Interestingly, the level of phosphatase activity responded to either HSPG or

CSPG ligands in a dose-dependent manner. ISP may additionally be signaling through

PTPσ in a similar manner as HSPGs. It may be interesting to assay phosphatase activity in conjunction with observation of its putative dimerization state as we saw a dose-dependent protease activity response with ISP as well. If ISP indeed incurs changes in levels of phosphatase activity similarly to HSPGs, then it would be logical to next identify which effectors may regulate exocytosis, and subsequently protease activity in the ECM. This is also further explored in “Future Directions.”

Aside from PTPσ signaling, studying protease activity is not without its limitations.

Specifically, inherent in the use of primary cells was the difficulty in procuring enough cells required to effectively identify different proteases. While we have positively identified cathepsin B secretion by primary peripheral neurons and MMP-2 by oligodendrocytes, we cannot exclude that other proteases are involved at this time. Especially since our current understanding of proteases suggests that proteases work in concert as a network to orchestrate ECM remodeling, it is highly likely that the proteases we have identified also work to activate other proteases not yet identified through our protein screens. Another caveat in studying proteases is the difficulty in attributing any resulting phenotype with just one of its substrates as any one protease have been widely characterized to include many

189 substrates. Specifically, while we have observed that many of the CSPG-degrading roles of protease activity have resulted in CSPG disinhibition that have been widely noted in the spinal cord injury literature, we cannot exclude that proteases may be encouraging axon outgrowth or oligodendrocyte survival, maturation, and myelination through other means.

For example, proteases have been reported to be involved in a plethora of pathways responsible for cell homeostasis independent of ECM remodeling including growth factor production and processing, receptor processing, and so forth (Bai and Pfaff 2011).

Cathepsin B itself has been reported to degrade adherens and other receptors responsible for cell-cell adhesion that additionally allows for enhanced cell motility (Olson and Joyce

2015). MMP-2 has additionally been shown to excise the pro domain of NGF to enhance sympathetic neuron sprouting (Saygili et al. 2011). Even within the limitations of these caveats, however, we have shown that these proteases are nonetheless important in axon outgrowth and oligodendrocyte progenitor cell homeostasis on a CSPG substrate as inhibition of these proteases resulted in striking phenotypic changes typically seen in

CSPG-induced inhibition.

4.3.1 Limitations of Modulation of Protein Tyrosine Phosphatase Receptor Sigma in

Peripheral Axons

Although we have linked, for the first time, cathepsin B activity following PTPσ modulation to promote axonal outgrowth through CSPG disinhibition, there are other experiments we wished we had also completed. Namely, the bulk of our study was based in in vitro experiments using DRG neurons that would need to be validated in vivo. To begin

190 addressing this problem, we had immunostained spinal cord sections of ISP and vehicle treated spinal cord injured rats with CGRP, a marker for the small caliber subtype of peripheral axons. There was no noticeable difference between ISP-treated and vehicle control spinal cord sections in their CGRP immunostaining; however, increased CGRP+ axonal sprouting may itself be a problem as this is a marker for peripheral axons responsible for nociception. Unfortunately, other markers specific to the different subtypes of DRG neurons were not readily available. Another way to begin exploring whether ISP enhanced DRG sprouting following spinal cord injury in vivo would be to label DRGs in the rat following spinal cord injury and ISP or vehicle treatment with an AAV-GFP injection into the DRG. This would putatively label enough DRGs to allow for visualization of sprouting in the spinal cord in sections near the lesion core. However, while this technique would enhance visualization of potentially sprouting axons into the spinal cord, it would not identify the different subtype of DRGs conducive to sprouting if this were indeed the case. To explore whether there is a differential ability of peripheral axons to sprout into the spinal cord following injury, different genetically labeled mouse lines could be used. For example, a variety of transcription factors have been found to be specific for different subtypes of DRG sensory axons. Specifically, TrkC has been found to label proprioceptors in adulthood, TrkB and Ret (the receptor for the glial-derived growth factor) labels two subtypes of mechanoreceptors, and the transcription factors Runx1,

Mrgrpa3, and Mrgrpb4 have been found to be specific for nociceptors, pruriceptors, and C- mechanoreceptors respectively (Yang Liu and Q. Ma 2011).

191

Ultimately, the goal would be to test the necessity of cathepsin B in the regeneration of

DRG axons into the cord. To address this, a reliable rat or mouse model of DRG labeling would be given a spinal cord injury and treated with ISP or vehicle in conjunction with an

AAV construct overexpressing the cathepsin B inhibitor, cystatin B, injected into DRGs before injury. We predict that these in vivo experiments would correlate with our in vitro experiments to show decreased DRG sprouting into the cord when the AAV-cystatin B construct was injected even in the presence of ISP.

Another limitation of our study was the lack of experimentation of serotonergic neurons in vivo as well as in vitro. While we had first observed that GAG-CSPG digestion may be occurring around serotonergic neurons; unfortunately, due to the difficulty in extracting large quantities of adult serotonergic neurons required for biochemical and in vitro assays, we could not ascertain whether these neurons were enhancing cathepsin B secretion and activity like the DRG neurons. Ideally, we would have liked to perform some of the same biochemical assays to assess protease activity like zymography or EnzChek of the conditioned media of ISP-treated serotonergic neurons. While we have immunostained for cathepsin B and found that it is colocalized with serotonin immunostaining, we would have liked to perform in vivo validation of cathepsin B in enhancing serotonergic sprouting following spinal cord injury. To do this, we would have liked to take advantage of the Cre-

Pet-1 mouse line and the tamoxofin inducible Cre-flox system in conjunction with a cathepsin B flox mutant to conditionally knockout cathepsin B well before spinal cord injury in order to assess the necessity of this protease to serotonergic sprouting following

ISP treatment. We would anticipate that ISP-induced serotonergic sprouting would be, in

192 part, due to cathepsin B secretion and activity. However, there may be issues inherent in this model of knocking out cathepsin B including compensatory increased protease activity from related cathepsins such as L and S. To combat this issue, a more difficult method of injecting an AAV-cystatin B overexpressing construct into the caudal raphe complex which projects to the serotonergic fibers into the spinal cord could be utilized.

4.3.2 Limitations of Modulation of Protein Tyrosine Phosphatase Receptor Sigma in

Oligodendrocytes

Recently, Ohtake et al. have reported findings that, at first glance, seem as though genetic knockout of PTPσ exacerbates the EAE condition (Ohtake et al. 2017). However, a critical review of this study shows that Ohtake et al. have compared clinical scores between constitutive PTPσ knockout and BALB/C wild type animals, which is a genetic background widely known for its resistance to EAE-induced demyelination (Teuscher et al.

1987). In our study, we have found improvements upon PTPσ modulation in C57BL/6 mice. While it seems as though PTPσ regulates immune cell regulation of EAE resistance in BALB/C mice, it is important to note that this genetic background seems more useful in studying immune-related EAE development instead of functional recovery following demyelination. However, this recent finding may suggest that PTPσ plays a more complicated role in demyelinating pathogenesis than anticipated. PTPσ expression in dendritic cells and macrophages, for example, has been shown to decrease in response to

LPS injections in mice perhaps due to enhanced cytokine signaling (Arimura and Yagi

2010). Additionally, PTPσ has been identified as an inhibitory receptor in plasmacytoid

193 dendritic cells, which produce type I interferon (IFN) in response to infection (Bunin et al.

2015). PTPσ has also been linked to inflammatory processes in ulcerative colitis, a syndrome affected by the immune system (Muise et al. 2007). While these current studies suggest that PTPσ in immune cells may be playing an inhibitory role in response to pro- inflammatory signals and that PTPσ itself may be responsive to pro-inflammatory cytokines, exactly how PTPσ regulates immune cell regulation in the context of EAE is still unclear and will require further study.

Regarding the indirect actions of ISP, recent studies are beginning to suggest that CSPGs promulgate the pro-inflammatory environment by activating glia and immune cells, among them macrophages (Rolls et al. 2008). Specifically, Rolls et al. have shown that CSPGs directly activated microglia and macrophages through the CD44 receptor and increased their secretion of the pro-inflammatory factor, TNFα. Surely, CSPGs play a role in regulating inflammatory processes as its upregulation correlates with the pro-inflammatory phase of multiple different CNS injuries. The Bradbury group first reported a correlative link between the CS-GAG moieties and immune modulation. Specifically, CS-GAG depletion using lentiviral delivery of a chondroitinase ABC construct injected into the spinal cord following contusive injury was seen to decrease the macrophage marker CD68, while increasing the M2 marker, CD206 without affecting the amount of total Iba1-labeled macrophages (Bartus et al. 2014). Typically, following spinal cord injury, M2 markers are upregulated and persist for around three days following injury. However, pro-inflammatory factors and M1 markers overwhelm the injury environment with the influx of systemically circulating macrophages (Donnelly and Popovich 2008). Conditioned media collected

194 from seven-day injured spinal cords of lentiviral chondroitinase ABC treated rats were able to differentiate unpolarized THP-1 monocytes towards an M2 phenotype (Bartus et al.

2014). In a proceeding study, Didangelos et al. linked the enzyme chondroitinase ABC treatment itself with an increase of the anti-inflammatory cytokine, interleukin (IL)-10

(Didangelos et al. 2014). In fact, only western blot analysis from lysate processed from the lesion epicenter, an area normally replete with pro-inflammatory immune cells, that received the chondroitinase ABC injection saw an increased in M2 markers such as Arg1.

Together, these studies suggest that attenuating CS-GAG load itself may modulate inflammation following injury. Indeed, work from Dyck et al. links PTPσ and LAR modulation using ISP and a similarly designed LAR-targeted peptide, ILP (D. Fisher et al.

2011), treatments lead to an increase in anti-inflammatory macrophage phenotype including enhanced M2 marker CD163 and decreased M1 marker CD86 expression (Dyck et al. 2018). Moreover, ISP and ILP treatments decreased pro-inflammatory factors including IL-1β and TNFα while increasing anti-inflammatory factors including IL-10 and arg-1 following spinal cord injury. Given these findings, we hypothesize that the immune cells of EAE-induced spinal cords are also tempered toward an anti-inflammatory bias following ISP treatment and PTPσ modulation. Initial immunostaining of ISP-treated EAE spinal cords saw a gradual decrease in Iba1-labeled cells in demyelinated lesions. While more thorough analysis could have been performed to begin investigating the extent to which PTPσ modulation alters the immune system beyond the decreased iNOS (M1) and increased Arginase-1 (M2) immunostaining we found following EAE, our studies centered on the oligodendrocyte progenitor cell-inhibitory effects wrought by CSPGs. Immediately following EAE, CSPGs were present in similar amounts in demyelinated lesions of both

195 control and ISP-treated groups. Additionally, Dyck et al. found that ILP and ISP treatments did not alter the levels of CS-56 assessed following immunoblotting of 1-14 day injured spinal cord lysates (Dyck et al. 2018). Given that baseline levels of CSPGs are not affected by PTPσ modulation, we maintain that ISP treatment allows for the gradual degradation of

CSPGs overtime by enhanced protease release.

Finally, while we have confirmed that reactive astrocytes secrete versican following demyelinating insult, we have also observed high levels of colocalization of CSPGs with the macrophage marker, Iba1 especially along the white matter in our EAE mouse model.

Unfortunately our studies exclude explorations into how this may occur. Macrophages in this model could be secreting CSPGs as they have been observed to do following pro- inflammatory stimulation in vitro (Makatsori et al. 2003) or they may be phagocytosing

CSPGs previously secreted by reactive glia. How ISP may be affecting these processes is still unclear. Additionally, it is possible that ISP modulation of PTPσ on macrophages could be contributing to the remyelination phenotype we have noticed in our demyelinating models. We can at least rule out that ISP enhances remyelination even in an environment where infiltrating macrophages are precluded as ISP still enhanced remyelination in ex vivo cerebellar slices which lack infiltrating macrophages but where endogenous immune cells are still included. However, to better understand how ISP affects macrophages we would have to assay macrophages’ ability to secrete CSPGs and phagocytosis in vitro. Initial work by the Silver lab suggests that CSPGs may indeed be attenuating the ability of microglia to phagocytose latex beads in vitro (Shaffer et al. 1995; DeWitt et al. 1998). At the heart of this issue is whether ISP may be modulating inflammatory processes in the EAE model

196 directly, that is by modulating PTPσ found on immune cells themselves, or indirectly by decreasing CSPG load. Clearly, additional experiments would be required to better elucidate how ISP may be affecting the immune system after injury.

4.4 Future Directions

Linking protease activity to PTPσ helps to unlock a set of new questions that would inquire how PTPσ is able to regulate axon regeneration, synaptogenesis, and perhaps even basic cellular homeostasis such as autophagy. Here we propose three possible future directions that may help expand a better understanding of how CSPG/PTPσ interactions impact plasticity. These include proposals to help understand the role of protease activity in PTPσ as a switch for plasticity, understanding the molecular mechanism by which PTPσ enhances protease activity, and exploring whether CSPG/PTPσ interactions may be regulating autophagy.

4.4.1 The Role of Protease Activity in Protein Tyrosine Phosphatase Sigma as a

Switch for Plasticity

The most exciting discovery of our work was linking enhanced protease activity with PTPσ modulation. Currently, we are beginning to understand that PTPσ plays a vital role in regulating processes governing both axon outgrowth and synaptogenesis. We would next like to understand whether protease activity plays a role in PTPσ’s regulation of this switch.

As punctate protease release has been implicated in enhancing axon outgrowth and

197 plasticity (Andries et al. 2017), we hypothesize that modulation of PTPσ and subsequent protease activity would similarly promote plasticity, and at high enough concentrations, prevent synaptogenesis from occurring by preventing PTPσ from orchestrating pro- synaptic adhesion between the pre and post-synaptic cells. To test this hypothesis, we could utilize a mixed culture assay to analyze neuronal synapse formation previously developed by Biederer and Scheiffele (Biederer and Scheiffele 2007). In this paradigm, primary rat hippocampal neurons are co-cultured with COS7 cells which express different proteins of interest to synaptogenesis. Following incubation, synapse formation could be assessed through immunostaining for synapse-specific markers such as induced synapsin clustering in the neurons. To begin exploring whether PTPσ serves as a switch to promote either synaptogenesis or axon outgrowth at the expense of the other, we could culture the cells on either low doses of CSPGs or HSPGs, treat with vehicle or ISP, and assess synapsin density and even axon length using Tuj1 immunostaining. We would anticipate that synapsin density would be increased when cells are cultured on CSPGs than HSPGs. However, when cells are cultured on CSPGs and treated with ISP, we would hypothesize that synapsin density would decrease and Tuj1 length would increase.

To begin addressing the role of protease activity in synaptogenesis, we could treat this co- culture system with ISP and observe whether synapsin density decreases. We could then correlate whether this decrease in synapsin density correlates with a rise in protease activity in the conditioned media. To more directly assess the role of protease activity in synaptogenesis, specific proteases such as cathepsin B or perhaps their endogenous inhibitors such as cystatin B could be transfected into the COS7 cell line and treated with

198 vehicle or ISP. If protease activity were useful in PTPσ-modulation to enhance axon outgrowth at the expense of synaptogenesis, we would anticipate that ISP alone would decrease synaptogenesis. Additionally we would predict that overexpressing proteases would decrease synapsin density compared to vehicle and that overexpressing its inhibitor would increase synapsin density even in the presence of ISP treatment.

4.4.2 Identifying the Vesicular Body Linked to Protein Tyrosine Phosphatase Sigma

Regulated Protease Release

As our studies exclude inquiries into the cellular mechanism(s) underlying protease release following modulation of PTPσ, further studies will be needed to better elucidate these processes. For one, the type of vesicular body responsible for delivering cathepsin B into the extracellular space will need to be identified. Then, the possible cellular signaling pathways responsible for exocytosis of cathepsin B by PTPσ modulation will need to be explored.

To begin, the identity of the type of vesicle responsible for the secretion of cathepsin B will need to be characterized, which would help to narrow some possible cellular mechanisms by which PTPσ modulation is inducing cathepsin B secretion. It is possible that cathepsin B is packaged in either exosomes or microvesicles, which have both been implicated in metalloprotease release among other protein or nucleic acid exchange (Camussi et al.

2010). Microvesicles, typically characterized by a diameter of 100-1000nm and formed through budding and fission of the plasma membrane (Al-Nedawi et al. 2014), exhibit a

199 lipid composition similar to the plasma membrane. In contrast, exosomes must be released by the cell through exocytosis and progress through various stages including multivesicular body fusion with the plasma membrane (Kastelowitz and Yin 2014). To fully differentiate between these two types of secretory bodies, electron microscopy coupled with cathepsin B immunolabeling with gold particles may be utilized to better observe the type of vesicle encapsulating this protease following ISP dosing. Using electron microscopy, the size and layers of aminophospholipids present may be used to morphologically differentiate between microvesicles and exosomes (Kastelowitz and Yin 2014).

In addition to microvesicles and exosomes, it is also possible that cathepsin B remains within lysosomes while the contents of this organelle itself is exocytosed into the extracellular space. This has been previously observed in dendritic spines where activity induced fusion of the lysosome with the plasma membrane exocytosed lysosomal cathepsin

B (Padamsey et al. 2017). Lysosomal exocytosis, which includes conventional exocytosis of the lysosome as well as a distinct mechanism known as secretory lysosome, is a highly regulated process that couples the degradative properties of lysosomes with a sorting mechanism responsible for the storage and secretion of various proteins such as ATP, serotonin, or histamine (Blott and Griffiths 2002). Currently, only the protein Lyst has been identified to be uniquely involved in secretory lysosomes as opposed to the fusion and exycytosing processes of other vesicular bodies. Genetically ablating Lyst has been shown to prevent lysosomal fusion with the membrane and to cause enlarged lysosomes in cells

(McVey Ward et al. 2008). Coupling the genetic deletion of Lyst with ISP treatment following analysis of the amount of cathepsin B in the conditioned media would perhaps be

200 one starting point to assess whether cathepsin B release following PTPσ modulation is due to secretory lysosome.

Related to secretory lysosome is also secretory autophagy where the fusion of the lysosome and autophagosome results in an autolysosome that is sometimes exocytosed into the extracellular space (Ponpuak et al. 2015). This process has been cited as an unconventional method of protein secretion like the extracellular expulsion of the degraded contents of the autolysosome as well as lysosomal proteases such as cathepsin B. Unique to this process, however, is its regulatory connection to autophagic flux, which can be largely enhanced by genetic ablation of PTPσ (Martin et al. 2011). In yeast, enhancing autophagy through starvation results in unconventional protein secretion through secretory autophagy

(Ponpuak et al. 2015). It would be of great interest to observe whether this process is evolutionarily conserved in mammalian CNS cells and specifically enhanced in PTPσ null cells. Coupled with a GFP reporter on cathepsin B, PTPσ null cells could be live-imaged to investigate whether this protein’s secretion into the extracellular space is enhanced by knockout of PTPσ. Similarly, wild type cells expressing a cathepsin B-GFP fused construct and ISP treatment can be live-imaged to see if cathepsin B secretion is enhanced.

PTPσ signaling remains somewhat of a molecular black box with some tantalizing connections to broad cellular signaling pathways such as autophagy (Martin et al. 2011),

N-cadherin (Siu et al. 2006), β-catenin (Siu et al. 2006), Akt/ERK (Dyck et al. 2015), and

RhoA/Rock (Ohtake et al. 2016). Exactly how CSPGs perturb these pathways through

PTPσ is still under intense scrutiny. Identification of the type of vesicles responsible for

201 delivering cathepsin B into the extracellular space is one way we could begin to narrow the study of focus among these different pathways.

4.4.3 Protein Tyrosine Phosphatase Sigma, Chondroitin Sulfate Proteoglycans, and

the Regulation of Autophagy

In 2011, PTPσ was identified as a regulator of autophagic flux following an RNAi screen

(Martin et al. 2011). In the same study, Martin et al. found that RNAi knockdown or constitutive knockout of the receptor induced autophagic hyperactivity marked by an increase of cleaved light chain 3 (LC3B) subunit II, an autophagic vesicle-specific protein, as assessed in western blot as well as fluorescent microscopy in an eGFP-LC3B cell line, and increased autophagic vesicle abundance quantified following electron microscopy. In the time since 2011, further research has solidified the impact that CSPGs and PTPσ interactions play in axon regeneration and plasticity. Recently, the Kadomatsu group has presented preliminary evidence that CSPGs, specifically the CS-E sulfated pattern, bind to

PTPσ on axonal growth cones to disrupt autophagic flux at the autophagosome-lysosome fusion step providing tantalizing evidence that links, for the first time, extracellular CSPG imposed regulation of autophagic flux (Kadomatsu 2016). Using the Silver spot assay,

Kadomatsu proposed that axons caught in a CSPG gradient are filled with unfused lysosomes and autophagic vesicles. This calls into mind the bubbling vesicles in dystrophic end balls formed by axons first reported by Tom et al. which were also imaged on a spot gradient of CSPGs (Tom et al. 2004). Work from this thesis further implicates cathepsin B

202 as a protease of interest following PTPσ modulation in dorsal root ganglion axons. This is significant because the regulation of cathepsin B, a lysosomally-bound protease, is vital for regular lysosomal function including protein degradation and turnover (Zhou et al. 2013).

Lysosomes are additionally integral to autophagy as the last step of the pathway involves autophagic vesicles fusing with lysosomes to fully degrade autophagic contents (Sasaki and Yoshida 2015). Moreover, reduced or inhibited cathepsin B activity through Ca074-me or overexpressing its inhibitor, cystatin B, has been reported to impair autophagy (Lamore and Wondrak 2011; Tatti et al. 2012; Soori et al. 2016) and enhance lysosomal dysfunction (Mizunoe et al. 2017). Additionally, our work begins to show that cathepsin B immunoreactivity is enhanced in the axons of PTPσ null neurons compared to wild type control. In all, this burgeoning link between lysosomal-autophagic flux and CSPGs and axon inhibition will require further study.

Following western blotting to probe cleaved LC3B, we have preliminary evidence confirming the discovery by Martin et al. that spinal cord and brain lysates from a PTPσ knockout animal enhances autophagic flux compared to wild type (unpublished). Further preliminary western blots probing for LC3B suggest that oligodendrocyte progenitor cells cultured on CSPGs alone are able to reduce autophagic flux compared to cells grown on laminin. Chondroitinase ABC treatment of CSPGs additionally reversed the inhibitory effects of CSPGs on autophagy, perhaps confirming initial observations by the Kadoka group indicating that the CS-E pattern of CSPG-GAG sulfation plays a major role in autophagy regulation. Intriguingly, ISP treatment of oligodendrocyte progenitor cells grown on CSPGs seems to be able to reverse the CSPG-induced dampening of autophagy.

203

These preliminary observations and experiments are important because they may help to elucidate how CSPGs effect axon regeneration at the molecular level, how axonal dystrophy occurs, and generally how CSPGs affect normal neuronal homeostasis especially since more studies are finding that autophagy impacts axon regeneration following spinal cord injury. For example, inhibition of autophagy has been found to reduce neurite outgrowth in vitro in DRG neurons (Clarke and Mearow 2016). He et al. have additionally observed that induction of autophagy using a synthetic peptide stabilizes microtubules to enhance axon regeneration following spinal cord injury (M. He et al. 2016). Furthermore, initial characterization of autophagy following rat contusive spinal cord injury suggest that autophagic flux is indeed impaired and that lysosomal cathepsins are decreased following injury that may contribute to ER stress and neuronal apoptosis (S. Liu et al. 2015). Given that CSPGs are greatly upregulated following injury (Cregg et al. 2014), it would stand to reason that one of the many ways CSPGs may be potently inhibiting axon regeneration would be through dampening autophagic flux, perhaps at the autophagosome and lysosomal fusion step.

As CSPGs are gaining wider recognition in the detrimental roles they play in other neurodegenerative diseases such as stroke (Huang et al. 2014), Multiple Sclerosis (H.

Mohan et al. 2010), and Alzheimer’s disease (DeWitt et al. 1993) among others, it will be important to fully characterize the effects of CSPG/PTPσ interactions on regulation of autophagy. To begin, CSPG regulation on autophagy can be initially measured by examining the ratio of cleaved LC3B II to uncleaved LC3B on western blots, which helps to qualitatively detect the presence of autophagic vesicles which are enriched with LC3B II.

204

Importantly, it will be vital to determine whether autophagic flux or the complete process of autophagy from the delivery of autophagic vesicles to lysosomes to their subsequent degradation in these structures (Klionsky et al. 2016), are changed by increased extracellular CSPGs. To do this, axons genetically labeled with GFP-tagged LC3B could be imaged in real time and GFP-LC3B puncta may be tracked over time and compared between axons cultured on laminin alone or low concentrations of CSPG. In this way,

LC3B-labeled autophagosomes in the axon tip can be monitored and when performed on a gradient of CSPGs in the Silver spot assay, LC3B dynamics can be qualitatively assessed as the growth cone stalls and end bulbs form. We hypothesize at this juncture that an accumulation of autophagosomes failing to fuse with lysosomes and thus a dampening of autophagic flux would occur as the growth cone develops swelled end bulbs along an increasing gradient of CSPGs. Electron Microscopy can then be implemented on axon tips cultured on laminin or CSPG to qualitatively and quantitatively analyze the morphology of autophagic structures in order to begin exploring how CSPGs are affecting its regulation.

For example, examining axon tips at this level of resolution will help to differentiate different stages of autophagic structures as they progress through the pathway including phagophore, phagosomes, amphisomes, and autolysosomes. As we have previously observed an increase of vesicle accumulation at the dystrophic growth cone (Tom et al.

2004), it is possible that CSPGs may be causing an increase in immature autophagic vesicles that are unable to fuse with lysosomes. Additionally, it will be intriguing to assess whether ISP treatment is able to reverse the autophagy-dampening effects of CSPG on axons using the techniques described above. Ultimately, the relationship between enhanced axon outgrowth and autophagic flux would need to be elucidated. Perhaps enhanced

205 autophagic flux and more efficient autophagosome-lyosomal fusion exhibited by PTPσ null cells (Martin et al. 2011) is in part responsible for the observed ability of these axons to augment axon outgrowth on CSPG-rich environments. Additionally intriguing is whether the colocalized expression of cathepsin B in serotonin axons and correlating enhanced serotonergic sprouting to regions of depleted CS-GAG immunostaining around the ventral motor pool and central canal of the spinal cord following ISP treatment may additionally correlate with an increased ability of this neuronal subtype to fuse autophagosomes and lysosomes. Surely future work will be required to assess whether enhanced cathepsin B expression in serotonin axons underlines their ability to robustly sprout following an upregulation of CSPGs after injury.

4.5 Concluding Remarks

Santiago Ramon y Cajal wrote, “In adult centers the nerve paths are something fixed, ended, immutable. Everything may die, nothing may be regenerated” (Cajal 1959). This long-standing dogma endured until work by David and Aguayo demonstrated through peripheral nerve transplants into the CNS that regeneration in the spinal cord was indeed possible (David and Aguayo 1981). Work continued by others in the spinal cord injury field aims to persistently pursue this ideal: that functional regeneration of the spinal cord is, at the least - possible, following injury.

In this vein, the Silver lab has sought to understand how extracellular matrix proteins, specifically CSPGs, upregulated following an inflammatory storm incited by spinal cord

206 injury contribute to regeneration failure (Tran and Warren et al. 2018). Indeed, intraspinal injections of the CSPG-degrading enzyme, chondroitinase ABC, has been shown to improve functional recovery following injury to the spinal cord (Bradbury et al. 2002;

Alilain et al. 2011). Since the discovery of PTPσ as the cognate receptor of CSPGs, other less invasive treatments, such as ISP, have been developed and shown to improve locomotor, sensory, and urinary function following contusive spinal cord injury (Lang et al. 2015).

Underlined by the two separate studies included in this thesis is the potently prominent effects CSPG inhibition imposes not only in spinal cord injury, but other disease paradigms involving upregulated inflammatory responses following CNS insult such as multiple sclerosis. Provided here is additional work further implicating upregulated CSPGs to not only the failure of axon regeneration, but oligodendrocyte-mediated remyelination as well.

We have sought to better understand how the CSPG and PTPσ interactions can be modulated to improve function following CNS injury. Namely, we have found a novel mechanism by which modulation of PTPσ can lead to punctate protease release that, in turn, degrades CSPGs and in this way relieves CSPG-mediated inhibition on axons and oligodendrocyte progenitor cells. While future work will be necessary to fully understand the mechanism underlying this effect, some intriguing results from this work and previously published studies implicate a tantalizing relationship between lysosomes/autophagosomes and CSPGs/PTPσ. Understanding these interactions may not only elucidate normal CSPG biology, but help to one day explain how CSPGs impact various neurodegenerative diseases.

207

Bibliography

Agrawal SM, Lau L, Yong VW. 2008. MMPs in the central nervous system: where the good guys go bad. Semin. Cell Dev. Biol. 19:42–51.

Al-Nedawi K, Meehan B, Rak J. 2014. Microvesicles: messengers and mediators of tumor progression. - PubMed - NCBI. Cell Cycle 8:2014–2018.

Alilain WJ, Horn KP, Hu H, Dick TE, Silver J. 2011. Functional regeneration of respiratory pathways after spinal cord injury. Nature 475:196–200.

Andersen JN, Elson A, Lammers R, Rømer J, Clausen JT, Møller KB, Møller NP. 2001. Comparative study of protein tyrosine phosphatase-epsilon isoforms: membrane localization confers specificity in cellular signalling. Biochem. J. 354:581–590.

Andrews EM, Richards RJ, Yin FQ, Viapiano MS, Jakeman LB. 2012. Alterations in chondroitin sulfate proteoglycan expression occur both at and far from the site of spinal contusion injury. Experimental Neurology 235:174–187.

Andries L, Van Hove I, Moons L, De Groef L. 2017. Matrix Metalloproteinases During Axonal Regeneration, a Multifactorial Role from Start to Finish. Mol Neurobiol 54:2114–2125.

Aricescu AR, McKinnell IW, Halfter W, Stoker AW. 2002. Heparan sulfate proteoglycans are ligands for receptor protein tyrosine phosphatase sigma. Molecular and Cellular Biology 22:1881–1892.

Arimura Y, Yagi J. 2010. Comprehensive Expression Profiles of Genes for Protein Tyrosine Phosphatases in Immune Cells. Sci Signal 3:rs1–rs1.

Asher RA, Morgenstern DA, Fidler PS, Adcock KH, Oohira A, Braistead JE, Levine JM, Margolis RU, Rogers JH, Fawcett JW. 2000. Neurocan is upregulated in injured brain and in cytokine-treated astrocytes. J. Neurosci. 20:2427–2438.

Asher RA, Morgenstern DA, Properzi F, Nishiyama A, Levine JM, Fawcett JW. 2005. Two separate metalloproteinase activities are responsible for the shedding and processing of the NG2 proteoglycan in vitro. Mol. Cell. Neurosci. 29:82–96.

Asher RA, Morgenstern DA, Shearer MC, Adcock KH, Pesheva P, Fawcett JW. 2002. Versican is upregulated in CNS injury and is a product of oligodendrocyte lineage cells. Journal of Neuroscience 22:2225–2236.

Bahr BA, Wisniewski ML, Butler D. 2012. Positive lysosomal modulation as a unique strategy to treat age-related protein accumulation diseases. Rejuvenation Res 15:189–

208

197.

Bai G, Pfaff SL. 2011. Protease regulation: the Yin and Yang of neural development and disease. Neuron 72:9–21.

Bajor M, Kaczmarek L. 2013. Proteolytic remodeling of the synaptic cell adhesion molecules (CAMs) by metzincins in synaptic plasticity. Neurochem Res 38:1113–1121.

Banerjee SB, Gutzeit VA, Baman J, Aoued HS, Doshi NK, Liu RC, Ressler KJ. 2017. Perineuronal Nets in the Adult Sensory Cortex Are Necessary for Fear Learning. Neuron 95:169–179.e3.

Bardehle S, Krüger M, Buggenthin F, Schwausch J, Ninkovic J, Clevers H, Snippert HJ, Theis FJ, Meyer-Luehmann M, Bechmann I, et al. 2013. Live imaging of astrocyte responses to acute injury reveals selective juxtavascular proliferation. - PubMed - NCBI. Nat Neurosci 16:580–586.

Barnabé-Heider F, Göritz C, Sabelström H, Takebayashi H, Pfrieger FW, Meletis K, Frisén J. 2010. Origin of New Glial Cells in Intact and Injured Adult Spinal Cord. Cell Stem Cell 7:470–482.

Barrett D. 2002. Proteinase and Peptidase Inhibition: Recent Potential Targets for Drug Development. Drug Discovery Today 7:1124.

Bartus K, James ND, Didangelos A, Bosch KD, Verhaagen J, Yanez-Munoz RJ, Rogers JH, Schneider BL, Muir EM, Bradbury EJ. 2014. Large-Scale Chondroitin Sulfate Proteoglycan Digestion with Chondroitinase Gene Therapy Leads to Reduced Pathology and Modulates Macrophage Phenotype following Spinal Cord Contusion Injury. Journal of Neuroscience 34:4822–4836.

Bateman J, Van Vactor D. 2001. The Trio family of guanine-nucleotide-exchange factors: regulators of axon guidance. Journal of Cell Science 114:1973–1980.

Baumann N, Pham-Dinh D. 2001. Biology of oligodendrocyte and myelin in the mammalian central nervous system. Physiological Reviews 81:871–927.

Bauvois B. 2012. New facets of matrix metalloproteinases MMP-2 and MMP-9 as cell surface transducers: Outside-in signaling and relationship to tumor progression. Biochimica et Biophysica Acta (BBA) - Reviews on Cancer 1825:29–36.

Bei F, Lee HHC, Liu X, Gunner G, Jin H, Ma L, Wang C, Hou L, Hensch TK, Frank E, et al. 2016. Restoration of Visual Function by Enhancing Conduction in Regenerated Axons. Cell 164:219–232.

Berdowska I. 2004. Cysteine proteases as disease markers. Clin. Chim. Acta 342:41–69.

209

Bergles DE, Richardson WD. 2015. Oligodendrocyte Development and Plasticity. Cold Spring Harbor Perspectives in Biology 8:a020453.

Bijata M, Labus J, Guseva D, Stawarski M, Butzlaff M, Dzwonek J, Schneeberg J, Böhm K, Michaluk P, Rusakov DA, et al. 2017. Synaptic Remodeling Depends on Signaling between Serotonin Receptors and the Extracellular Matrix. Cell Rep 19:1767–1782.

Bilwes AM, Hertog den J, Hunter T, Noel JP. 1996. Structural basis for inhibition of receptor protein-tyrosine phosphatase-α by dimerization. Nature 382:555–559.

Biniossek ML, Nägler DK, Becker-Pauly C, Schilling O. 2011. Proteomic identification of protease cleavage sites characterizes prime and non-prime specificity of cysteine cathepsins B, L, and S. - PubMed - NCBI. J. Proteome Res. 10:5363–5373.

Bixby JL. 2001. Ligands and Signaling Through Receptor-Type Tyrosine Phosphatases. IUBMB Life (International Union of Biochemistry and Molecular Biology: Life) 51:157– 163.

Blakemore WF, Gilson JM, Crang AJ. 2003. The presence of astrocytes in areas of demyelination influences remyelination following transplantation of oligodendrocyte progenitors. - PubMed - NCBI. Experimental Neurology 184:955–963.

Blott EJ, Griffiths GM. 2002. Secretory lysosomes. - PubMed - NCBI. Nat Rev Mol Cell Biol 3:122–131.

Bolte S, Cordelieres FP. 2006. A guided tour into subcellular colocalization analysis in light microscopy. J Microsc 224:213–232.

Bordi M, Berg MJ, Mohan PS, Peterhoff CM, Alldred MJ, Che S, Ginsberg SD, Nixon RA. 2016. Autophagy flux in CA1 neurons of Alzheimer hippocampus: Increased induction overburdens failing lysosomes to propel neuritic dystrophy. - PubMed - NCBI. Autophagy 12:2467–2483.

Boyd A, Zhang H, Williams A. 2013. Insufficient OPC migration into demyelinated lesions is a cause of poor remyelination in MS and mouse models. - PubMed - NCBI. Acta Neuropathol 125:841–859.

Bradbury EJ, Carter LM. 2011. Manipulating the glial scar: Chondroitinase ABC as a therapy for spinal cord injury. Brain Research Bulletin 84:306–316.

Bradbury EJ, Moon LDF, Popat RJ, King VR, Bennett GS, Patel PN, Fawcett JW, McMahon SB. 2002. Chondroitinase ABC promotes functional recovery after spinal cord injury. Nature 416:636–640.

Brittis PA, Canning DR, Silver J. 1992. Chondroitin sulfate as a regulator of neuronal patterning in the retina. Science 255:733–736.

210

Brkic M, Balusu S, Libert C, Vandenbroucke RE. 2015. Friends or Foes: Matrix Metalloproteinases and Their Multifaceted Roles in Neurodegenerative Diseases. Mediators of Inflammation 2015:620581–27.

Brooks JM, Su J, Levy C, Wang JS, Seabrook TA, Guido W, Fox MA. 2013. A molecular mechanism regulating the timing of corticogeniculate innervation. Cell Rep 5:573–581.

Brooks PC, Silletti S, Schalscha von TL, Friedlander M, Cheresh DA. 1998. Disruption of angiogenesis by PEX, a noncatalytic metalloproteinase fragment with integrin binding activity. Cell 92:391–400.

Brooks PC, Strömblad S, Sanders LC, Schalscha von TL, Aimes RT, Stetler-Stevenson WG, Quigley JP, Cheresh DA. 1996. Localization of matrix metalloproteinase MMP-2 to the surface of invasive cells by interaction with integrin alpha v beta 3. Cell 85:683–693.

Brown JM, Xia J, Zhuang B, Cho K-S, Rogers CJ, Gama CI, Rawat M, Tully SE, Uetani N, Mason DE, et al. 2012. A sulfated carbohydrate epitope inhibits axon regeneration after injury. Proceedings of the National Academy of Sciences 109:4768–4773.

Bundesen LQ, Scheel TA, Bregman BS, Kromer LF. 2003. Ephrin-B2 and EphB2 regulation of astrocyte-meningeal fibroblast interactions in response to spinal cord lesions in adult rats. J. Neurosci. 23:7789–7800.

Bunin A, Sisirak V, Ghosh HS, Grajkowska LT, Hou ZE, Miron M, Yang C, Ceribelli M, Uetani N, Chaperot L, et al. 2015. Protein Tyrosine Phosphatase PTPRS Is an Inhibitory Receptor on Human and Murine Plasmacytoid Dendritic Cells. - PubMed - NCBI. Immunity 43:277–288.

Buttery PC, ffrench-Constant C. 1999. Laminin-2/integrin interactions enhance myelin membrane formation by oligodendrocytes. Mol. Cell. Neurosci. 14:199–212.

Cajal SRY. 1959. Degeneration & Regeneration of the Nervous System.

Camussi G, Deregibus MC, Bruno S, Cantaluppi V, Biancone L. 2010. Exosomes/microvesicles as a mechanism of cell-to-cell communication. Kidney International 78:838–848.

Cao L, Rickenbacher GT, Rodriguez S, Moulia TW, Albers MW. 2012. The precision of axon targeting of mouse olfactory sensory neurons requires the BACE1 protease. Sci. Rep. 2:231.

Carbonetto S, Gruver MM, Turner DC. 1983. Nerve fiber growth in culture on fibronectin, collagen, and glycosaminoglycan substrates. Journal of Neuroscience 3:2324–2335.

211

Carmichael ST, Archibeque I, Luke L, Nolan T, Momiy J, Li S. 2005. Growth-associated gene expression after stroke: evidence for a growth-promoting region in peri-infarct cortex. Experimental Neurology 193:291–311.

Carulli D, Laabs T, Geller HM, Fawcett JW. 2005. Chondroitin sulfate proteoglycans in neural development and regeneration. Current Opinion in Neurobiology 15:116–120.

Cataldo AM, Nixon RA. 1990. Enzymatically active lysosomal proteases are associated with amyloid deposits in Alzheimer brain. Proceedings of the National Academy of Sciences 87:3861–3865.

Cavallo-Medved D, Dosescu J, Linebaugh BE, Sameni M, Rudy D, Sloane BF. 2003. Mutant K-ras regulates cathepsin B localization on the surface of human colorectal carcinoma cells. Neoplasia 5:507–519.

Cavallo-Medved D, Rudy D, Blum G, Bogyo M, Caglic D, Sloane BF. 2009. Live-cell imaging demonstrates extracellular matrix degradation in association with active cathepsin B in caveolae of endothelial cells during tube formation. Exp. Cell Res. 315:1234–1246.

Cermak S, Kosicek M, Mladenovic-Djordjevic A, Smiljanic K, Kanazir S, Hecimovic S. 2016. Loss of Cathepsin B and L Leads to Lysosomal Dysfunction, NPC-Like Cholesterol Sequestration and Accumulation of the Key Alzheimer's Proteins. PLoS ONE 11:e0167428.

Chagnon MJ, Uetani N, Tremblay ML. 2004. Functional significance of the LAR receptor protein tyrosine phosphatase family in development and diseases. Biochem. Cell Biol. 82:664–675.

Chang A, Nishiyama A, Peterson J, Prineas J, Trapp BD. 2000. NG2-positive oligodendrocyte progenitor cells in adult human brain and multiple sclerosis lesions. Journal of Neuroscience 20:6404–6412.

Chang A, Staugaitis SM, Dutta R, Batt CE, Easley KE, Chomyk AM, Yong VW, Fox RJ, Kidd GJ, Trapp BD. 2012. Cortical remyelination: A new target for repair therapies in multiple sclerosis. Annals of Neurology 72:918–926.

Chang A, Tourtellotte WW, Rudick R, Trapp BD. 2002. Premyelinating Oligodendrocytes in Chronic Lesions of Multiple Sclerosis. N Engl J Med 346:165–173.

Chang S-H, Kanasaki K, Gocheva V, Blum G, Harper J, Moses MA, Shih S-C, Nagy JA, Joyce J, Bogyo M, et al. 2009. VEGF-A Induces Angiogenesis by Perturbing the Cathepsin-Cysteine Protease Inhibitor Balance in Venules, Causing Basement Membrane Degradation and Mother Vessel Formation. Cancer Res. 69:4537–4544.

Chen S, Dong H, Yang S, Guo H. 2017. Cathepsins in digestive cancers. Oncotarget

212

8:41690–41700.

Chung C, Puthanveetil P, Ory DS, Lieberman AP. 2016. Genetic and pharmacological evidence implicates cathepsins in Niemann-Pick C cerebellar degeneration. Hum. Mol. Genet. 25:1434–1446.

Clandinin TR, Lee CH, Herman T, Lee RC, Yang AY, Ovasapyan S, Zipursky SL. 2001. Drosophila LAR regulates R1-R6 and R7 target specificity in the visual system. Neuron 32:237–248.

Clarke J-P, Mearow K. 2016. Autophagy inhibition in endogenous and nutrient-deprived conditions reduces dorsal root ganglia neuron survival and neurite growth in vitro. J. Neurosci. Res. 94:653–670.

Coles CH, Mitakidis N, Zhang P, Elegheert J, Lu W, Stoker AW, Nakagawa T, Craig AM, Jones EY, Aricescu AR. 2014. Structural basis for extracellular cis and trans RPTPσ signal competition in synaptogenesis. - PubMed - NCBI. Nat Comms 5:5209.

Coles CH, Shen Y, Tenney AP, Siebold C, Sutton GC, Lu W, Gallagher JT, Jones EY, Flanagan JG, Aricescu AR. 2011. Proteoglycan-Specific Molecular Switch for RPTP Clustering and Neuronal Extension. Science 332:484–488.

Colognato H, ffrench-Constant C, Feltri ML. 2005. Human diseases reveal novel roles for neural laminins. Trends in Neurosciences 28:480–486.

Correale J, de los Milagros Bassani Molinas M. 2003. Temporal variations of adhesion molecules and matrix metalloproteinases in the course of MS. Journal of Neuroimmunology 140:198–209.

Court FA, Zambroni D, Pavoni E, Colombelli C, Baragli C, Figlia G, Sorokin L, Ching W, Salzer JL, Wrabetz L, et al. 2011. MMP2-9 cleavage of dystroglycan alters the size and molecular composition of Schwann cell domains. J. Neurosci. 31:12208–12217.

Cregg JM, DePaul MA, Filous AR, Lang BT, Tran A, Silver J. 2014. Functional regeneration beyond the glial scar. Experimental Neurology 253:197–207.

Crespo D, Asher RA, Lin R, Rhodes KE, Fawcett JW. 2007. How does chondroitinase promote functional recovery in the damaged CNS? Experimental Neurology 206:159– 171.

Criste G, Trapp B, Dutta R. 2014. Axonal loss in multiple sclerosis: causes and mechanisms. Handb Clin Neurol 122:101–113.

Crowe MJ, Bresnahan JC, Shuman SL, Masters JN, Beattie MS. 1997. Apoptosis and delayed degeneration after spinal cord injury in rats and monkeys. Nat Med 3:73–76.

213

Cua RC, Lau LW, Keough MB, Midha R, Apte SS, Yong VW. 2013. Overcoming neurite-inhibitory chondroitin sulfate proteoglycans in the astrocyte matrix. Glia 61:972– 984.

Cudrici C, Niculescu T, Niculescu F, Shin ML, Rus H. 2006. Oligodendrocyte cell death in pathogenesis of multiple sclerosis: Protection of oligodendrocytes from apoptosis by complement. J Rehabil Res Dev 43:123–132.

Cuzner ML, Gveric D, Strand C, Loughlin AJ, Paemen L, Opdenakker G, Newcombe J. 1996. The expression of tissue-type plasminogen activator, matrix metalloproteases and endogenous inhibitors in the central nervous system in multiple sclerosis: comparison of stages in lesion evolution. J. Neuropathol. Exp. Neurol. 55:1194–1204.

Czopka T, Holst Von A, Schmidt G, ffrench-Constant C, Faissner A. 2009. Tenascin C and tenascin R similarly prevent the formation of myelin membranes in a RhoA- dependent manner, but antagonistically regulate the expression of myelin basic protein via a separate pathway. Glia 57:1790–1801.

David S, Aguayo AJ. 1981. Axonal elongation into peripheral nervous system “bridges” after central nervous system injury in adult rats. Science 214:931–933.

Davie EW, Ratnoff OD. 1964. Waterfall sequence for intrinsic blood clotting. Science 145:1310–1312.

Davies SJ, Fitch MT, Memberg SP, Hall AK, Raisman G, Silver J. 1997. Regeneration of adult axons in white matter tracts of the central nervous system. Nature 390:680–683.

Davies SJ, Goucher DR, Doller C, Silver J. 1999. Robust regeneration of adult sensory axons in degenerating white matter of the adult rat spinal cord. J. Neurosci. 19:5810– 5822. de Santana Nunes AK, Raposo C, de Oliveira WH, Thomé R, Verinaud L, Tovar-Moll F, Peixoto CA. 2016. Phosphodiesterase-5 inhibition promotes remyelination by MCP- 1/CCR-2 and MMP-9 regulation in a cuprizone-induced demyelination model. Experimental Neurology 275:143–153.

De Wit J, De Winter F, Klooster J, Verhaagen J. 2005. Semaphorin 3A displays a punctate distribution on the surface of neuronal cells and interacts with proteoglycans in the extracellular matrix. Mol. Cell. Neurosci. 29:40–55.

Debant A, Serra-Pagès C, Seipel K, O'Brien S, Tang M, Park SH, Streuli M. 1996. The multidomain protein Trio binds the LAR transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate rac-specific and rho-specific guanine nucleotide exchange factor domains. Proceedings of the National Academy of Sciences 93:5466– 5471.

214

Deguchi K, Takaishi M, Hayashi T, Oohira A, Nagotani S, Li F, Jin G, Nagano I, Shoji M, Miyazaki M, et al. 2005. Expression of neurocan after transient middle cerebral artery occlusion in adult rat brain. Brain Research 1037:194–199.

Dendrou CA, Fugger L, Friese MA. 2015. Immunopathology of multiple sclerosis. Nat. Rev. Immunol. 15:545–558.

DeWitt DA, Perry G, Cohen M, Doller C, Silver J. 1998. Astrocytes regulate microglial phagocytosis of senile plaque cores of Alzheimer's disease. Experimental Neurology 149:329–340.

DeWitt DA, Silver J, Canning DR, Perry G. 1993. Chondroitin sulfate proteoglycans are associated with the lesions of Alzheimer's disease. Experimental Neurology 121:149– 152.

Di Giaimo R, Riccio M, Santi S, Galeotti C, Ambrosetti DC, Melli M. 2002. New insights into the molecular basis of progressive myoclonus epilepsy: a multiprotein complex with cystatin B. Hum. Mol. Genet. 11:2941–2950. di Penta A, Moreno B, Reix S, Fernandez-Diez B, Villanueva M, Errea O, Escala N, Vandenbroeck K, Comella JX, Villoslada P. 2013. Oxidative Stress and Proinflammatory Cytokines Contribute to Demyelination and Axonal Damage in a Cerebellar Culture Model of Neuroinflammation. Stangel M, editor. PLoS ONE 8:e54722.

Dickendesher TL, Baldwin KT, Mironova YA, Koriyama Y, Raiker SJ, Askew KL, Wood A, Geoffroy CG, Zheng B, Liepmann CD, et al. 2012. NgR1 and NgR3 are receptors for chondroitin sulfate proteoglycans. Nature Neuroscience 15:703–712.

Didangelos A, Iberl M, Vinsland E, Bartus K, Bradbury EJ. 2014. Regulation of IL-10 by Chondroitinase ABC Promotes a Distinct Immune Response following Spinal Cord Injury. The Journal of Neuroscience 34:16424–16432.

Djinovic-Carugo K, Gautel M, Ylänne J, Young P. 2002. The spectrin repeat: a structural platform for cytoskeletal protein assemblies. FEBS Letters 513:119–123.

Donnelly DJ, Popovich PG. 2008. Inflammation and its role in neuroprotection, axonal regeneration and functional recovery after spinal cord injury. Experimental Neurology 209:378–388.

Dow KE, Ethell DW, Steeves JD, Riopelle RJ. 1994. Molecular correlates of spinal cord repair in the embryonic chick: heparan sulfate and chondroitin sulfate proteoglycans. - PubMed - NCBI. Experimental Neurology 128:233–238.

Duchossoy Y. 2001. MMP-related gelatinase activity is strongly induced in scar tissue of injured adult spinal cord and forms pathways for ingrowing neurites. Mol. Cell. Neurosci. 17:945–956.

215

Duff D, Long A. 2017. Roles for RACK1 in cancer cell migration and invasion. Cellular Signalling 35:250–255.

Dunah AW, Hueske E, Wyszynski M, Hoogenraad CC, Jaworski J, Pak DT, Simonetta A, Liu G, Sheng M. 2005. LAR receptor protein tyrosine phosphatases in the development and maintenance of excitatory synapses. Nat Neurosci 8:458–467.

Dyck S, Kataria H, Alizadeh A, Santhosh KT, Lang B, Silver J, Karimi-Abdolrezaee S. 2018. Perturbing chondroitin sulfate proteoglycan signaling through LAR and PTPσ receptors promotes a beneficial inflammatory response following spinal c... J Neuroinflammation 15:169.

Dyck SM, Alizadeh A, Santhosh KT, Proulx EH, Wu C-L, Karimi-Abdolrezaee S. 2015. Chondroitin Sulfate Proteoglycans Negatively Modulate Spinal Cord Neural Precursor Cells by Signaling Through LAR and RPTPsigma and Modulation of the Rho/ROCK Pathway. Stem Cells 33:2550–2563.

Elchebly M, Wagner J, Kennedy TE, Lanctôt C, Michaliszyn E, Itié A, Drouin J, Tremblay ML. 1999. Neuroendocrine dysplasia in mice lacking protein tyrosine phosphatase sigma. Nat. Genet. 21:330–333.

Eldridge R, Livanainen M, Stern R, Koerber T, Wilder BJ. 1983. “Baltic” myoclonus epilepsy: hereditary disorder of childhood made worse by phenytoin. The Lancet 322:838–842.

Ellis RC, O'Steen WA, Hayes RL, Nick HS, Wang KKW, Anderson DK. 2005. Cellular localization and enzymatic activity of cathepsin B after spinal cord injury in the rat. Experimental Neurology 193:19–28.

Embury CM, Dyavarshetty B, Lu Y, Wiederin JL, Ciborowski P, Gendelman HE, Kiyota T. 2017. Cathepsin B Improves ß-Amyloidosis and Learning and Memory in Models of Alzheimer's Disease. J Neuroimmune Pharmacol 12:340–352.

Evaniew N, Belley-Côté EP, Fallah N, Noonan VK, Rivers CS, Dvorak MF. 2016. Methylprednisolone for the Treatment of Patients with Acute Spinal Cord Injuries: A Systematic Review and Meta-Analysis. Journal of Neurotrauma 33:468–481.

Farizatto KLG, Ikonne US, Almeida MF, Ferrari MFR, Bahr BA. 2017. Aβ42-mediated proteasome inhibition and associated tau pathology in hippocampus are governed by a lysosomal response involving cathepsin B: Evidence for protective crosstalk between protein clearance pathways. Padmanabhan J, editor. PLoS ONE 12:e0182895.

Favuzzi E, Marques-Smith A, Deogracias R, Winterflood CM, Sánchez-Aguilera A, Mantoan L, Maeso P, Fernandes C, Ewers H, Rico B. 2017. Activity-Dependent Gating of Parvalbumin Interneuron Function by the Perineuronal Net Protein Brevican. - PubMed - NCBI. Neuron 95:639–655.e10.

216

Fawcett JW. 2015. The extracellular matrix in plasticity and regeneration after CNS injury and neurodegenerative disease. Progress in brain research 218:213–226.

Fehlings MG, Wilson JR, Cho N. 2014. Methylprednisolone for the Treatment of Acute Spinal Cord Injury: Counterpoint. Neurosurgery 61:36–42.

Feng Y, Yu S, Lasell TKR, Jadhav AP, Macia E, Chardin P, Melancon P, Roth M, Mitchison T, Kirchhausen T. 2003. Exo1: a new chemical inhibitor of the exocytic pathway. Proceedings of the National Academy of Sciences 100:6469–6474.

Filous AR, Tran A, Howell CJ, Busch SA, Evans TA, Stallcup WB, Kang SH, Bergles DE, Lee S-I, Levine JM, et al. 2014. Entrapment via Synaptic-Like Connections between NG2 Proteoglycan+ Cells and Dystrophic Axons in the Lesion Plays a Role in Regeneration Failure after Spinal Cord Injury. The Journal of Neuroscience 34:16369– 16384.

Fisher D, Xing B, Dill J, Li H, Hoang HH, Zhao Z, Yang X-L, Bachoo R, Cannon S, Longo FM, et al. 2011. Leukocyte common antigen-related phosphatase is a functional receptor for chondroitin sulfate proteoglycan axon growth inhibitors. J. Neurosci. 31:14051–14066.

Fitch MT, Doller C, Combs CK, Landreth GE, Silver J. 1999. Cellular and molecular mechanisms of glial scarring and progressive cavitation: in vivo and in vitro analysis of inflammation-induced secondary injury after CNS trauma. J. Neurosci. 19:8182–8198.

Forgione N, Fehlings MG. 2014. Rho-ROCK Inhibition in the Treatment of Spinal Cord Injury. World Neurosurgery 82:e535–e539.

Fosang AJ, Last K, Gardiner P, Jackson DC, Brown L. 1995. Development of a cleavage- site-specific monoclonal antibody for detecting metalloproteinase-derived aggrecan fragments: detection of fragments in human synovial fluids. Biochem. J. 310 ( Pt 1):337– 343.

Fosang AJ, Neame PJ, Last K, Hardingham TE, MURPHY G, Hamilton JA. 1992. The interglobular domain of cartilage aggrecan is cleaved by PUMP, gelatinases, and cathepsin B. J Biol Chem 267:19470–19474.

Fox AN, Zinn K. 2005. The heparan sulfate proteoglycan syndecan is an in vivo ligand for the Drosophila LAR receptor tyrosine phosphatase. Curr. Biol. 15:1701–1711.

Franklin RJM, ffrench-Constant C. 2008. Remyelination in the CNS: from biology to therapy. Nat Rev Neurosci 9:839–855.

Freria CM, Hall JCE, Wei P, Guan Z, McTigue DM, Popovich PG. 2017. Deletion of the Fractalkine Receptor, CX3CR1, Improves Endogenous Repair, Axon Sprouting, and Synaptogenesis after Spinal Cord Injury in Mice. - PubMed - NCBI. Journal of

217

Neuroscience 37:3568–3587.

Friedman GC, Seeds NW. 1994. Tissue plasminogen activator expression in the embryonic nervous system. Brain Res. Dev. Brain Res. 81:41–49.

Fry EJ, Chagnon MJ, López-Vales R, Tremblay ML, David S. 2010. Corticospinal tract regeneration after spinal cord injury in receptor protein tyrosine phosphatase sigma deficient mice. Glia 58:423–433.

Fryer HJ, Kelly GM, Molinaro L, Hockfield S. 1992. The high molecular weight Cat-301 chondroitin sulfate proteoglycan from brain is related to the large aggregating proteoglycan from cartilage, aggrecan. J Biol Chem 267:9874–9883.

Fuller ML, DeChant AK, Rothstein B, Caprariello A, Wang R, Hall AK, Miller RH. 2007. Bone morphogenetic proteins promote gliosis in demyelinating spinal cord lesions. 62:288–300.

Gadani SP, Walsh JT, Lukens JR, Kipnis J. 2015. Dealing with Danger in the CNS: The Response of the Immune System to Injury. Neuron 87:47–62.

Gadani SP, Walsh JT, Smirnov I, Zheng J, Kipnis J. 2015. The Glia-Derived Alarmin IL- 33 Orchestrates the Immune Response and Promotes Recovery following CNS Injury. Neuron 85:703–709.

Gama CI, Tully SE, Sotogaku N, Clark PM, Rawat M, Vaidehi N, Goddard WA III, Nishi A, Hsieh-Wilson LC. 2006. Sulfation patterns of glycosaminoglycans encode molecular recognition and activity. Nature chemical biology 2:nchembio810–473.

Garcia E, Aguilar-Cevallos J, Silva-Garcia R, Ibarra A. 2016. Cytokine and Growth Factor Activation In Vivo and In Vitro after Spinal Cord Injury. Mediators of Inflammation 2016:1–21.

Gardner RT, Habecker BA. 2013. Infarct-derived chondroitin sulfate proteoglycans prevent sympathetic reinnervation after cardiac ischemia-reperfusion injury. J. Neurosci. 33:7175–7183.

Gardner RT, Wang L, Lang BT, Cregg JM, Dunbar CL, Woodward WR, Silver J, Ripplinger CM, Habecker BA. 2015. Targeting protein tyrosine phosphatase σ after myocardial infarction restores cardiac sympathetic innervation and prevents arrhythmias. Nat Comms 6:ncomms7235.

Gaspar P, Cases O, Maroteaux L. 2003. The developmental role of serotonin: news from mouse molecular genetics. Nat Rev Neurosci 4:1002–1012.

Geurts N, Becker-Pauly C, Martens E, Proost P, Van den Steen PE, Stöcker W, Opdenakker G. 2012. Meprins process matrix metalloproteinase‐9 (MMP‐9)/gelatinase B

218 and enhance the activation kinetics by MMP‐3. FEBS Letters 586:4264–4269.

Gocheva V, Wang H-W, Gadea BB, Shree T, Hunter KE, Garfall AL, Berman T, Joyce JA. 2010. IL-4 induces cathepsin protease activity in tumor-associated macrophages to promote cancer growth and invasion. Genes & Development 24:241–255.

Gogolla N, Caroni P, Luthi A, Herry C. 2009. Perineuronal nets protect fear memories from erasure. Science 325:1258–1261.

Gondi CS, Rao JS. 2013. Cathepsin B as a cancer target. Expert Opin. Ther. Targets 17:281–291.

Gonzalez-Rothi EJ, Lee K-Z, Dale EA, Reier PJ, Mitchell GS, Fuller DD. 2015. Intermittent hypoxia and neurorehabilitation. J Appl Physiol 119:1455–1465.

Göritz C, Dias DO, Tomilin N, Barbacid M, Shupliakov O, Frisén J. 2011. A pericyte origin of spinal cord scar tissue. Science (New York, N.Y.) 333:238–242.

Grimpe B, Silver J. 2004. A novel DNA enzyme reduces glycosaminoglycan chains in the glial scar and allows microtransplanted dorsal root ganglia axons to regenerate beyond lesions in the spinal cord. J. Neurosci. 24:1393–1397.

Gris P, Tighe A, Levin D, Sharma R, Brown A. 2007. Transcriptional regulation of scar gene expression in primary astrocytes. - PubMed - NCBI. Glia 55:1145–1155.

Gross J, Lapiere CM. 1962. Collagenolytic activity in amphibian tissues: a tissue culture assay. Proceedings of the National Academy of Sciences 48:1014–1022.

Gueye Y, Ferhat L, Sbai O, Bianco J, Ould-yahoui A, Bernard A, Charrat E, Chauvin JP, Risso J-J, Féron F, et al. 2011. Trafficking and secretion of matrix metalloproteinase-2... [Glia. 2011] - PubMed - NCBI. Glia 59:750–770.

Han KA, Woo D, Kim S, Choii G, Jeon S, Won SY, Kim HM, Do Heo W, Um JW, Ko J. 2016. Neurotrophin-3 Regulates Synapse Development by Modulating TrkC-PTPσ Synaptic Adhesion and Intracellular Signaling Pathways. Journal of Neuroscience 36:4816–4831.

Harlow DE, Macklin WB. 2014. Inhibitors of myelination: ECM changes, CSPGs and PTPs. Experimental Neurology 251:39–46.

Hawthorne AL, Hu H, Kundu B, Steinmetz MP, Wylie CJ, Deneris ES, Silver J. 2011. The Unusual Response of Serotonergic Neurons after CNS Injury: Lack of Axonal Dieback and Enhanced Sprouting within the Inhibitory Environment of the Glial Scar. Journal of Neuroscience 31:5605–5616.

Hayden SM, Seeds NW. 1996. Modulated expression of plasminogen activator system

219 components in cultured cells from dissociated mouse dorsal root ganglia. Journal of Neuroscience 16:2307–2317.

He M, Ding Y, Chu C, Tang J, Xiao Q, Luo Z-G. 2016. Autophagy induction stabilizes microtubules and promotes axon regeneration after spinal cord injury. Proc. Natl. Acad. Sci. U.S.A. 113:11324–11329.

He X, Knepper M, Ding C, Li J, Castro S, Siddiqui M, Schachner M. 2012. Promotion of Spinal Cord Regeneration by Neural Stem Cell-Secreted Trimerized Cell Adhesion Molecule L1. Callaerts P, editor. PLoS ONE 7:e46223.

Hermelinda S-C, Guizar-Sahagun G, Feria-Velasco A, Grijalva I, Espitia L, Ibarra A, Madrazo I. 1998. Spontaneous long-term remyelination after traumatic spinal cord injury in rats. Brain Research 782:126–135.

Hoffmann KM, Tonks NK, Barford D. 1997. The crystal structure of domain 1 of receptor protein-tyrosine phosphatase mu. J Biol Chem 272:27505–27508.

Hofmeyer K, Treisman JE. 2009. The receptor protein tyrosine phosphatase LAR promotes R7 photoreceptor axon targeting by a phosphatase-independent signaling mechanism. Proc. Natl. Acad. Sci. U.S.A. 106:19399–19404.

Horn KE, Xu B, Gobert D, Hamam BN, Thompson KM, Wu C-L, Bouchard J-F, Uetani N, Racine RJ, Tremblay ML, et al. 2012. Receptor protein tyrosine phosphatase sigma regulates synapse structure, function and plasticity. J. Neurochem. 122:147–161.

Hostenbach S, Cambron M, D’haeseleer M, Kooijman R, De Keyser J. 2014. Astrocyte loss and astrogliosis in neuroinflammatory disorders. Neurosci. Lett. 565:39–41.

Hsu JYC, McKeon R, Goussev S, Werb Z, Lee JU, Trivedi A, Noble-Haeusslein LJ. 2006. Matrix Metalloproteinase-2 Facilitates Wound Healing Events That Promote Functional Recovery after Spinal Cord Injury. J. Neurosci. 26:9841–9850.

Hu J, Ni S, Cao Y, Zhang T, Wu T, Yin X, Lang Y, Lu H. 2016. The Angiogenic Effect of microRNA-21 Targeting TIMP3 through the Regulation of MMP2 and MMP9. Zhao C, editor. PLoS ONE 11:e0149537.

Hu X, Hu J, Dai L, Trapp B, Yan R. 2015. Axonal and Schwann cell BACE1 is equally required for remyelination of peripheral nerves. J. Neurosci. 35:3806–3814.

Huang L, Wu Z-B, Zhuge Q, Zheng W, Shao B, Wang B, Sun F, Jin K. 2014. Glial scar formation occurs in the human brain after ischemic stroke. Int. J. Med. Sci. 11:344–348.

Hughes CE, Caterson B, Fosang AJ, Roughley PJ, Mort JS. 1995. Monoclonal antibodies that specifically recognize neoepitope sequences generated by “aggrecanase” and matrix metalloproteinase cleavage of aggrecan: application to catabolism in situ and in vitro.

220

Biochem. J. 305 ( Pt 3):799–804.

Jedeszko C, Sloane BF. 2004. Cysteine cathepsins in human cancer. Biol. Chem. 385:1017–1027.

Ji R-R, Xu Z-Z, Wang X, Lo EH. 2009. Matrix metalloprotease regulation of neuropathic pain. Trends Pharmacol. Sci. 30:336–340.

Jin Y, Dougherty SE, Wood K, Sun L, Cudmore RH, Abdalla A, Kannan G, Pletnikov M, Hashemi P, Linden DJ. 2016. Regrowth of Serotonin Axons in the Adult Mouse Brain Following Injury. Neuron 91:748–762.

Joensuu T, Tegelberg S, Reinmaa E, Segerstråle M, Hakala P, Pehkonen H, Korpi ER, Tyynelä J, Taira T, Hovatta I, et al. 2014. Gene expression alterations in the cerebellum and granule neurons of Cstb(-/-) mouse are associated with early synaptic changes and inflammation. PLoS ONE 9:e89321.

Jones LL, Sajed D, Tuszynski MH. 2003. Axonal Regeneration through Regions of Chondroitin Sulfate Proteoglycan Deposition after Spinal Cord Injury: A Balance of Permissiveness and Inhibition. Journal of Neuroscience 23:9276–9288.

Ju SM, Song HY, Lee JA, Lee SJ, Choi SY, Park J. 2009. Extracellular HIV-1 Tat upregulates expression of matrix metalloproteinase-9 via a MAPK-NF-kappaB dependent pathway in human astrocytes. - PubMed - NCBI. Experimental & Molecular Medicine 41:86–93.

Kadomatsu K. 2016. Sulfated glycans regulate autophagy and axon regeneration. In: Geller HM; Silver J, editors. San Diego.

Kajstura TJ, Dougherty SE, Linden DJ. 2017. Serotonin axons in the neocortex of the adult female mouse regrow after traumatic brain injury. J. Neurosci. Res. 38:358.

Kaplan A, Ong Tone S, Fournier AE. 2015. Extrinsic and intrinsic regulation of axon regeneration at a crossroads. Front Mol Neurosci 8:27.

Karus M, Ulc A, Ehrlich M, Czopka T, Hennen E, Fischer J, Mizhorova M, Qamar N, Brüstle O, Faissner A. 2015. Regulation of oligodendrocyte precursor maintenance by chondroitin sulphate glycosaminoglycans. Glia 64:270–286.

Kastelowitz N, Yin H. 2014. Exosomes and microvesicles: identification and targeting by particle size and lipid chemical probes. - PubMed - NCBI. ChemBioChem 15:923–928.

Kawabata H, Setoguchi T, Yone K, Souda M, Yoshida H, Kawahara K-I, Maruyama I, Komiya S. 2010. High mobility group box 1 is upregulated after spinal cord injury and is associated with neuronal cell apoptosis. Spine 35:1109–1115.

221

Keough MB, Rogers JA, Zhang P, Jensen SK, Stephenson EL, Chen T, Hurlbert MG, Lau LW, Rawji KS, Plemel JR, et al. 2016. An inhibitor of chondroitin sulfate proteoglycan synthesis promotes central nervous system remyelination. Nat Comms 7:11312.

Kieseier BC, Clements JM, Pischel HB, Wells GM, Miller K, Gearing AJ, Hartung HP. 1998. Matrix metalloproteinases MMP-9 and MMP-7 are expressed in experimental autoimmune neuritis and the Guillain-Barré syndrome. Annals of Neurology 43:427–434.

Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG. 2009. Identification of Two Distinct Macrophage Subsets with Divergent Effects Causing either Neurotoxicity or Regeneration in the Injured Mouse Spinal Cord. Journal of Neuroscience 29:13435–13444.

Kikuchi H, Yamada T, Furuya H, Doh-ura K, Ohyagi Y, Iwaki T, Kira J-I. 2003. Involvement of cathepsin B in the motor neuron degeneration of amyotrophic lateral sclerosis. Acta Neuropathol 105:462–468.

Kim D, Zai L, Liang P, Schaffling C, Ahlborn D, Benowitz LI. 2013a. Inosine Enhances Axon Sprouting and Motor Recovery after Spinal Cord Injury. PLoS ONE 8:e81948.

Kim D, Zai L, Liang P, Schaffling C, Ahlborn D, Benowitz LI. 2013b. Inosine enhances axon sprouting and motor recovery after spinal cord injury. PLoS ONE 8:e81948.

Kim Y, Remacle AG, Chernov AV, Liu H, Shubayev I, Lai C, Dolkas J, Shiryaev SA, Golubkov VS, Mizisin AP, et al. 2012. The MMP-9/TIMP-1 axis controls the status of differentiation and function of myelin-forming Schwann cells in nerve regeneration. - PubMed - NCBI. Singh SR, editor. PLoS ONE 7:e33664.

Kinnunen A, Kinnunen T, Kaksonen M, Nolo R, Panula P, Rauvala H. 1998. N-syndecan and HB-GAM (heparin-binding growth-associated molecule) associate with early axonal tracts in the rat brain. Eur. J. Neurosci. 10:635–648.

Kippert A, Fitzner D, Helenius J, Simons M. 2009. Actomyosin contractility controls cell surface area of oligodendrocytes. BMC Cell Biology 2009 10:1 10:71.

Kjellen L, Lindahl U. 1991. Proteoglycans: structures and interactions. Annual Review of Biochemistry 60:443–475.

Klionsky DJ, Abdelmohsen K, Abe A, Abedin MJ, Abeliovich H, Acevedo Arozena A, Adachi H, Adams CM, Adams PD, Adeli K, et al. 2016. Guidelines for the use and interpretation of assays for monitoring autophagy (3rd edition). Autophagy 12:1–222.

Ko JS, Pramanik G, Um JW, Shim JS, Lee D, Kim KH, Chung G-Y, Condomitti G, Kim HM, Kim H, et al. 2015. PTPσ functions as a presynaptic receptor for the glypican- 4/LRRTM4 complex and is essential for excitatory synaptic transmission. - PubMed -

222

NCBI. Proceedings of the National Academy of Sciences 112:1874–1879.

Kostoulas G, Lang A, Nagase H, Baici A. 1999. Stimulation of angiogenesis through cathepsin B inactivation of the tissue inhibitors of matrix metalloproteinases. FEBS Letters 455:286–290.

Krystosek A, Seeds NW. 1981. Plasminogen activator release at the neuronal growth cone. Science 213:1532–1534.

Krystosek A, Seeds NW. 1984. Peripheral neurons and Schwann cells secrete plasminogen activator. The Journal of Cell Biology 98:773–776.

Kuhlmann T, Miron V, Cui Q, Cuo Q, Wegner C, Antel J, Bruck W. 2008. Differentiation block of oligodendroglial progenitor cells as a cause for remyelination failure in chronic multiple sclerosis. Brain 131:1749–1758.

Kurazono S, Okamoto M, Sakiyama J, Mori S, Nakata Y, Fukuoka J, Amano S, Oohira A, Matsui H. 2001. Expression of brain specific chondroitin sulfate proteoglycans, neurocan and phosphacan, in the developing and adult hippocampus of Ihara's epileptic rats. Brain Research 898:36–48.

Kvachnina E, Liu G, Dityatev A, Renner U, Dumuis A, Richter DW, Dityateva G, Schachner M, Voyno-Yasenetskaya TA, Ponimaskin EG. 2005. 5-HT7 receptor is coupled to G alpha subunits of heterotrimeric G12-protein to regulate gene transcription and neuronal morphology. J. Neurosci. 25:7821–7830.

Kwok JCF, Dick G, Wang D, Fawcett JW. 2011. Extracellular matrix and perineuronal nets in CNS repair. Dev Neurobiol 71:1073–1089.

Kwon SK, Woo J, Kim SY, Kim H, Kim E. 2010. Trans-synaptic Adhesions between Netrin-G Ligand-3 (NGL-3) and Receptor Tyrosine Phosphatases LAR, Protein-tyrosine Phosphatase (PTP ), and PTP via Specific Domains Regulate Excitatory Synapse Formation. J Biol Chem 285:13966–13978.

Lamore SD, Wondrak GT. 2011. Autophagic-lysosomal dysregulation downstream of cathepsin B inactivation in human skin fibroblasts exposed to UVA. Photochem. Photobiol. Sci. 11:163–172.

Lang BT, Cregg JM, DePaul MA, Tran AP, Xu K, Dyck SM, Madalena KM, Brown BP, Weng Y-L, Li S, et al. 2015. Modulation of the proteoglycan receptor PTPσ promotes recovery after spinal cord injury. Nature 518:404–408.

Larsen PH, Wells JE, Stallcup WB, Opdenakker G, Yong VW. 2003. Matrix metalloproteinase-9 facilitates remyelination in part by processing the inhibitory NG2 proteoglycan. J. Neurosci. 23:11127–11135.

223

Larsen PH, Yong VW. 2004. The expression of matrix metalloproteinase-12 by oligodendrocytes regulates their maturation and morphological differentiation. J. Neurosci. 24:7597–7603.

Lau LW, Keough MB, Haylock-Jacobs S, Cua R, Döring A, Sloka S, Stirling DP, Rivest S, Yong VW. 2012. Chondroitin sulfate proteoglycans in demyelinated lesions impair remyelination. 72:419–432.

Lee JY, Choi HY, Yune TY. 2015. MMP-3 secreted from endothelial cells of blood vessels after spinal cord injury activates microglia, leading to oligodendrocyte cell death. Neurobiology of Disease 82:141–151.

Lee S, Sato Y, Nixon RA. 2011. Lysosomal proteolysis inhibition selectively disrupts axonal transport of degradative organelles and causes an Alzheimer's-like axonal dystrophy. J. Neurosci. 31:7817–7830.

Lehmann HC, Köhne A, Bernal F, Jangouk P, Meyer Zu Hörste G, Dehmel T, Hartung HP, Previtali SC, Kieseier BC. 2009. Matrix metalloproteinase-2 is involved in myelination of dorsal root ganglia neurons. Glia 57:479–489.

Lemke AK, Sandy JD, Voigt H, Dreier R, Lee JH, Grodzinsky AJ, Mentlein R, Fay J, Schunke M, Kurz B. 2010. Interleukin-1alpha treatment of meniscal explants stimulates the production and release of aggrecanase-generated, GAG-substituted aggrecan products and also the release of pre-formed, aggrecanase-generated G1 and m-calpain-generated G1-G2. Cell and Tissue Research 340:179–188.

Leonetti C, Macrez R, Pruvost M, Hommet Y, Bronsard J, Fournier A, Perrigault M, Machin I, Vivien D, Clemente D, et al. 2017. Tissue-type plasminogen activator exerts EGF-like chemokinetic effects on oligodendrocytes in white matter (re)myelination. Mol Neurodegeneration 12:20.

Leppert D, Ford J, Stabler G, Grygar C, Lienert C, Huber S, Miller KM, Hauser SL, Kappos L. 1998. Matrix metalloproteinase-9 (gelatinase B) is selectively elevated in CSF during relapses and stable phases of multiple sclerosis. Brain 121 ( Pt 12):2327–2334.

Leppert D, Waubant E, Bürk MR, Oksenberg JR, Hauser SL. 1996. Interferon beta-1b inhibits gelatinase secretion and in vitro migration of human T cells: a possible mechanism for treatment efficacy in multiple sclerosis. Annals of Neurology 40:846–852.

Li H, Wong C, Li W, Ruven C, He L, Wu X, Lang BT, Silver J, Wu W. 2015. Enhanced regeneration and functional recovery after spinal root avulsion by manipulation of the proteoglycan receptor PTPσ. Sci. Rep. 5:14923.

Li Y, Raisman G. 1995. Sprouts from cut corticospinal axons persist in the presence of astrocytic scarring in long-term lesions of the adult rat spinal cord. Experimental Neurology 134:102–111.

224

Lindberg RL, De Groot CJ, Montagne L, Freitag P, van der Valk P, Kappos L, Leppert D. 2001. The expression profile of matrix metalloproteinases (MMPs) and their inhibitors (TIMPs) in lesions and normal appearing white matter of multiple sclerosis. Brain 124:1743–1753.

Lindholm MW, Nilsson J, Moses J. 2005. Low density lipoprotein stimulation of human macrophage proteoglycan secretion. Biochemical and Biophysical Research Communications 328:455–460.

Liu H, Shubayev VI. 2011. Matrix metalloproteinase-9 controls proliferation of NG2+ progenitor cells immediately after spinal cord injury. - PubMed - NCBI. Experimental Neurology 231:236–246.

Liu P, Liu X, Liou AK-F, Xing J, Jing Z, Ji X, Liu X, Zhao H, Yan F, Chen J, et al. 2014. The neuroprotective mechanism of erythropoietin-TAT fusion protein against neurodegeneration from ischemic brain injury. CNS Neurol Disord Drug Targets 13:1465–1474.

Liu S, Sarkar C, Dinizo M, Faden AI, Koh EY, Lipinski MM, Wu J. 2015. Disrupted autophagy after spinal cord injury is associated with ER stress and neuronal cell death. Cell Death Dis 6:e1582–e1582.

Liu Yang, Ma Q. 2011. Generation of somatic sensory neuron diversity and implications on sensory coding. Current Opinion in Neurobiology 21:52–60.

Liu Yuanyuan, Wang X, Li W, Zhang Q, Li Y, Zhang Z, Zhu J, Chen B, Williams PR, Zhang Y, et al. 2017. A Sensitized IGF1 Treatment Restores Corticospinal Axon- Dependent Functions. Neuron 95:817–833.e4.

Lord-Fontaine S, Yang F, Diep Q, Dergham P, Munzer S, Tremblay P, McKerracher L. 2008. Local Inhibition of Rho Signaling by Cell-Permeable Recombinant Protein BA-210 Prevents Secondary Damage and Promotes Functional Recovery following Acute Spinal Cord Injury. Journal of Neurotrauma 25:1309–1322.

Lucchinetti C, Bruck W, Parisi J, Scheithauer B, Rodriguez M, Lassmann H. 1999. A quantitative analysis of oligodendrocytes in multiple sclerosis lesions. A study of 113 cases. Brain 122 ( Pt 12):2279–2295.

Luo F, Burke K, Kantor C, Miller RH, Yang Y. 2014. Cyclin-dependent kinase 5 mediates adult OPC maturation and myelin repair through modulation of Akt and GsK- 3β signaling. J. Neurosci. 34:10415–10429.

Luo F, Zhang J, Burke K, Miller RH, Yang Y. 2016. The Activators of Cyclin-Dependent Kinase 5 p35 and p39 Are Essential for Oligodendrocyte Maturation, Process Formation, and Myelination. J. Neurosci. 36:3024–3037.

225

Lutz D, Wolters-Eisfeld G, Schachner M, Kleene R. 2013. Cathepsin E generates a sumoylated intracellular fragment of the cell adhesion molecule L1 to promote neuronal and Schwann cell migration as well as myelination. Journal of Neurochemistry 128:713– 724.

Lv Q-L, Huang Y-T, Wang G-H, Liu Y-L, Huang J, Qu Q, Sun B, Hu L, Cheng L, Chen S-H, et al. 2016. Overexpression of RACK1 Promotes Metastasis by Enhancing Epithelial-Mesenchymal Transition and Predicts Poor Prognosis in Human Glioma. IJERPH 13:1021.

Ma Y, Chen Y, Petersen I. 2017. Expression and epigenetic regulation of cystatin B in lung cancer and colorectal cancer. Pathology - Research and Practice 213:1568–1574.

Mahad DH, Trapp BD, Lassmann H. 2015. Pathological mechanisms in progressive multiple sclerosis. Lancet Neurol 14:183–193.

Makatsori E, Lamari FN, Theocharis AD, Anagnostides S, Hjerpe A, Tsegenidis T, Karamanos NK. 2003. Large matrix proteoglycans, versican and perlecan, are expressed and secreted by human leukemic monocytes. Anticancer Res. 23:3303–3309.

Marcon RM, Barros Filho TEP de, Oliveira RP, Cristante AF, Taricco MA, Colares G, Barbarini AF, Teixeira WGJ, Souza FI de. 2010. Experimental study on the action of methylprednisolone on Wistar rats before spinal cord injury. Acta Ortopédica Brasileira 18:26–30.

Martin KR, Xu Y, Looyenga BD, Davis RJ, Wu CL, Tremblay ML, Xu HE, MacKeigan JP. 2011. Identification of PTP as an autophagic phosphatase. Journal of Cell Science 124:812–819.

Martinez P, Denys A, Delos M, Sikora A-S, Carpentier M, Julien S, Pestel J, Allain F. 2015. Macrophage polarization alters the expression and sulfation pattern of glycosaminoglycans. Glycobiology 25:502–513.

Mason SD, Joyce JA. 2011. Proteolytic networks in cancer. Trends in Cell Biology 21:228–237.

Massey JM, Hubscher CH, Wagoner MR, Decker JA, Amps J, Silver J, Onifer SM. 2006. Chondroitinase ABC digestion of the perineuronal net promotes functional collateral sprouting in the cuneate nucleus after cervical spinal cord injury. J. Neurosci. 26:4406– 4414.

Mataga N, Mizuguchi Y, Hensch TK. 2004. Experience-dependent pruning of dendritic spines in visual cortex by tissue plasminogen activator. Neuron 44:1031–1041.

McFarlane S. 2003. Metalloproteases. Neuron 37:559–562.

226

McGlinchey RP, Lee JC. 2015. Cysteine cathepsins are essential in lysosomal degradation of α-synuclein. Proc. Natl. Acad. Sci. U.S.A. 112:9322–9327.

McGuire PG, Seeds NW. 1990. Degradation of underlying extracellular matrix by sensory neurons during neurite outgrowth. Neuron 4:633–642.

McKeon RJ, Höke A, Silver J. 1995. Injury-induced proteoglycans inhibit the potential for laminin-mediated axon growth on astrocytic scars. Experimental Neurology 136:32– 43.

McKeon RJ, Schreiber RC, Rudge JS, Silver J. 1991. Reduction of neurite outgrowth in a model of glial scarring following CNS injury is correlated with the expression of inhibitory molecules on reactive astrocytes. Journal of Neuroscience 11:3398–3411.

McLean J, Batt J, Doering LC, Rotin D, Bain JR. 2002. Enhanced rate of nerve regeneration and directional errors after sciatic nerve injury in receptor protein tyrosine phosphatase sigma knock-out mice. J. Neurosci. 22:5481–5491.

McVey Ward D, Griffiths GM, Stinchcombe JC, Kaplan J. 2008. Analysis of the Lysosomal Storage Disease Chediak-Higashi Syndrome. Traffic 1:816–822.

Miller BA, Crum JM, Tovar CA, Ferguson AR, Bresnahan JC, Beattie MS. 2007. Developmental stage of oligodendrocytes determines their response to activated microglia in vitro. J Neuroinflammation 4:28.

Minor K, Phillips J, Seeds NW. 2009. Tissue plasminogen activator promotes axonal outgrowth on CNS myelin after conditioned injury. J. Neurochem. 109:706–715.

Miranpuri GS, Schomberg DT, Alrfaei B, King KC, Rynearson B, Wesley VS, Khan N, Obiakor K, Wesley UV, Resnick DK. 2016. Role of Matrix Metalloproteinases 2 in Spinal Cord Injury-Induced Neuropathic Pain. Ann Neurosci 23:25–32.

Mironova YA, Giger RJ. 2013. Where no synapses go: gatekeepers of circuit remodeling and synaptic strength. Trends in Neurosciences 36:363–373.

Miyata S, Komatsu Y, Yoshimura Y, Taya C, Kitagawa H. 2012. Persistent cortical plasticity by upregulation of chondroitin 6-sulfation. Nature Neuroscience 15:414–22– S1–2.

Mizuno H, Warita H, Aoki M, Itoyama Y. 2008. Accumulation of chondroitin sulfate proteoglycans in the microenvironment of spinal motor neurons in amyotrophic lateral sclerosis transgenic rats. J. Neurosci. Res. 86:2512–2523.

Mizunoe Y, Sudo Y, Okita N, Hiraoka H, Mikami K, Narahara T, Negishi A, Yoshida M, Higashibata R, Watanabe S, et al. 2017. Involvement of lysosomal dysfunction in autophagosome accumulation and early pathologies in adipose tissue of obese mice. -

227

PubMed - NCBI. Autophagy 13:642–653.

Mohamed MM, Sloane BF. 2006. Cysteine cathepsins: multifunctional enzymes in cancer. - PubMed - NCBI. Nature Reviews Cancer 2015 15:12 6:764–775.

Mohan H, Krumbholz M, Sharma R, Eisele S, Junker A, Sixt M, Newcombe J, Wekerle H, Hohlfeld R, Lassmann H, et al. 2010. Extracellular Matrix in Multiple Sclerosis Lesions: Fibrillar Collagens, Biglycan and Decorin are Upregulated and Associated with Infiltrating Immune Cells. 20:966–975.

Moon HY, Becke A, Berron D, Becker B, Sah N, Benoni G, Janke E, Lubejko ST, Greig NH, Mattison JA, et al. 2016. Running-Induced Systemic Cathepsin B Secretion Is Associated with Memory Function. - PubMed - NCBI. Cell Metabolism 24:332–340.

Morawski M, Brückner G, Arendt T, Matthews RT. 2012. Aggrecan: Beyond cartilage and into the brain. The International Journal of Biochemistry & Cell Biology 44:690– 693.

Moreno M, Guo F, Mills Ko E, Bannerman P, Soulika A, Pleasure D. 2013. Origins and significance of astrogliosis in the multiple sclerosis model, MOG peptide EAE. Journal of the Neurological Sciences 333:55–59.

Mort JS, Magny MC, Lee ER. 1998. Cathepsin B: an alternative protease for the generation of an aggrecan “metalloproteinase” cleavage neoepitope. Biochem. J. 335 ( Pt 3):491–494.

Muise AM, Walters T, Wine E, Griffiths AM, Turner D, Duerr RH, Regueiro MD, Ngan B-Y, Xu W, Sherman PM, et al. 2007. Protein-Tyrosine Phosphatase Sigma Is Associated with Ulcerative Colitis. Current Biology 17:1212–1218.

Naito Y, Lee AK, Takahashi H. 2017. Emerging roles of the neurotrophin receptor TrkC in synapse organization. Neuroscience Research 116:10–17.

Nash B, Ioannidou K, Barnett SC. 2010. Astrocyte phenotypes and their relationship to myelination. J Anatomy 219:44–52.

Nash B, Thomson CE, Linington C, Arthur AT, McClure JD, McBride MW, Barnett SC. 2011. Functional duality of astrocytes in myelination. J. Neurosci. 31:13028–13038.

Neha Aggarwal BFS. 2014. Cathepsin B: Multiple roles in cancer. Schilling O; Findeisen P, editors. Proteomics. Clinical applications 8:427–437.

Neumann S, Woolf CJ. 1999. Regeneration of dorsal column fibers into and beyond the lesion site following adult spinal cord injury. Neuron 23:83–91.

Noble LJ, Donovan F, Igarashi T, Goussev S, Werb Z. 2002. Matrix metalloproteinases

228 limit functional recovery after spinal cord injury by modulation of early vascular events. J. Neurosci. 22:7526–7535.

Oblander SA, Brady-Kalnay SM. 2010. Distinct PTPmu-associated signaling molecules differentially regulate neurite outgrowth on E-, N-, and R-cadherin. Molecular and Cellular Neuroscience 44:78–93.

Oh LYS, Larsen PH, Krekoski CA, Edwards DR, Donovan F, Werb Z, Yong VW. 1999. Matrix Metalloproteinase-9/Gelatinase B Is Required for Process Outgrowth by Oligodendrocytes. Journal of Neuroscience 19:8464–8475.

Ohtake Y, Kong W, Hussain R, Horiuchi M, Tremblay ML, Ganea D, Li S. 2017. Protein tyrosine phosphatase σ regulates autoimmune encephalomyelitis development. Brain Behav. Immun. 65:111–124.

Ohtake Y, Wong D, Abdul-Muneer PM, Selzer ME, Li S. 2016. Two PTP receptors mediate CSPG inhibition by convergent and divergent signaling pathways in neurons. Sci. Rep. 6:399.

Okada S, Nakamura M, Katoh H, Miyao T, Shimazaki T, Ishii K, Yamane J, Yoshimura A, Iwamoto Y, Toyama Y, et al. 2006. Conditional ablation of Stat3 or Socs3 discloses a dual role for reactive astrocytes after spinal cord injury. Nat Med 12:829–834.

Okamoto M, Sakiyama J, Mori S, Kurazono S, Usui S, Hasegawa M, Oohira A. 2003. Kainic acid-induced convulsions cause prolonged changes in the chondroitin sulfate proteoglycans neurocan and phosphacan in the limbic structures. Experimental Neurology 184:179–195.

Olson OC, Joyce JA. 2015. Cysteine cathepsin proteases: regulators of cancer progression and therapeutic response. Nature Reviews Cancer 2015 15:12 15:nrc4027– 729.

Oluich L-J, Stratton JAS, Xing YL, Ng SW, Cate HS, Sah P, Windels F, Kilpatrick TJ, Merson TD. 2012. Targeted ablation of oligodendrocytes induces axonal pathology independent of overt demyelination. J. Neurosci. 32:8317–8330.

Ould-yahoui A, Sbai O, Baranger K, Bernard A, Gueye Y, Charrat E, Clément B, Gigmes D, Dive V, Girard SD, et al. 2013. Role of matrix metalloproteinases in migration and neurotrophic properties of nasal olfactory stem and ensheathing cells. Cell Transplant 22:993–1010.

Padamsey Z, McGuinness L, Bardo SJ, Reinhart M, Tong R, Hedegaard A, Hart ML, Emptage NJ. 2017. Activity-Dependent Exocytosis of Lysosomes Regulates the Structural Plasticity of Dendritic Spines. Neuron 93:132–146.

Page-McCaw A, Ewald AJ, Werb Z. 2007. Matrix metalloproteinases and the regulation

229 of tissue remodelling. Nat Rev Mol Cell Biol 8:221–233.

Pan D, Rubin GM. 1997. Kuzbanian Controls Proteolytic Processing of Notch and Mediates Lateral Inhibition during Drosophila and Vertebrate Neurogenesis. Cell 90:271–280.

Pastrana E, Moreno-Flores MT, Gurzov EN, Avila J, Wandosell F, Diaz-Nido J. 2006. Genes associated with adult axon regeneration promoted by olfactory ensheathing cells: a new role for matrix metalloproteinase 2. J. Neurosci. 26:5347–5359.

Patrikios P, Stadelmann C, Kutzelnigg A, Rauschka H, Schmidbauer M, Laursen H, Sorensen PS, Brück W, Lucchinetti C, Lassmann H. 2006. Remyelination is extensive in a subset of multiple sclerosis patients. Brain 129:3165–3172.

Pearse DD, Pereira FC, MARCILLO AE, Bates ML, Berrocal YA, Filbin MT, Bunge MB. 2004. cAMP and Schwann cells promote axonal growth and functional recovery after spinal cord injury. Nat Med 10:610–616.

Pendleton JC, Shamblott MJ, Gary DS, Belegu V, Hurtado A, Malone ML, McDonald JW. 2013. Chondroitin sulfate proteoglycans inhibit oligodendrocyte myelination through PTPσ. 247:113–121.

Pereira JE, Costa LM, Cabrita AM, Couto PA, Filipe VM, Magalhães LG, Fornaro M, Di Scipio F, Geuna S, Maurício AC, et al. 2009. Methylprednisolone fails to improve functional and histological outcome following spinal cord injury in rats. Experimental Neurology 220:71–81.

Petanceska S, Burke S, Watson SJ, Devi L. 1994. Differential distribution of messenger RNAs for cathepsins B, L and S in adult rat brain: an in situ hybridization study. Neuroscience 59:729–738.

Pizzorusso T. 2009. Erasing fear memories. Science (New York, N.Y.) 325:1214–1215.

Pizzorusso T, Medini P, Berardi N, Chierzi S, Fawcett JW, Maffei L. 2002. Reactivation of ocular dominance plasticity in the adult visual cortex. Science 298:1248–1251.

Ponpuak M, Mandell MA, Kimura T, Chauhan S, Cleyrat C, Deretic V. 2015. Secretory autophagy. Current Opinion in Cell Biology 35:106–116.

Properzi F, Carulli D, Asher RA, Muir E, Camargo LM, van Kuppevelt TH, Dam ten GB, Furukawa Y, Mikami T, Sugahara K, et al. 2005. Chondroitin 6-sulphate synthesis is upregulated in injured CNS, induced by injury-related cytokines and enhanced in axon- growth inhibitory glia. - PubMed - NCBI. European Journal of Neuroscience 21:378– 390.

Pruvost M, Lépine M, Leonetti C, Etard O, Naveau M, Agin V, Docagne F, Maubert E,

230

Ali C, Emery E, et al. 2017. ADAMTS-4 in oligodendrocytes contributes to myelination with an impact on motor function. Glia 65:1961–1975.

Puente XS, Sánchez LM, Overall CM, López-Otín C. 2003. Human and mouse proteases: a comparative genomic approach. Nat Rev Genet 4:544–558.

Pulido R, Serra-Pagès C, Tang M, Streuli M. 1995. The LAR/PTP delta/PTP sigma subfamily of transmembrane protein-tyrosine-phosphatases: multiple human LAR, PTP delta, and PTP sigma isoforms are expressed in a tissue-specific manner and associate with the LAR-interacting protein LIP.1. Proceedings of the National Academy of Sciences 92:11686–11690.

Pyka M, Wetzel C, Aguado A, Geissler M, Hatt H, Faissner A. 2011. Chondroitin sulfate proteoglycans regulate astrocyte-dependent synaptogenesis and modulate synaptic activity in primary embryonic hippocampal neurons. - PubMed - NCBI. European Journal of Neuroscience 33:2187–2202.

Rajapaksha TW, Eimer WA, Bozza TC, Vassar R. 2011. The Alzheimer's β-secretase enzyme BACE1 is required for accurate axon guidance of olfactory sensory neurons and normal glomerulus formation in the... - PubMed - NCBI. Mol Neurodegeneration 6:88.

Rashid-Doubell F, McKinnell I, Aricescu AR, Sajnani G, Stoker A. 2002. Chick PTPsigma regulates the targeting of retinal axons within the optic tectum. J. Neurosci. 22:5024–5033.

Rawlings ND, Tolle DP, Barrett AJ. 2004. MEROPS: the peptidase database. Nucleic Acids Research 32:D160–4.

Riccio M, Santi S, Dembic M, Digiaimo R, Cipollini E, Costantinoceccarini E, Ambrosetti D, Maraldi N, Melli M. 2005. Cell-specific expression of the epm1 (cystatin B) gene in developing rat cerebellum. Neurobiology of disease 20:104–114.

Robinson S, Tani M, Strieter RM, Ransohoff RM, Miller RH. 1998. The chemokine growth-regulated oncogene-alpha promotes spinal cord oligodendrocyte precursor proliferation. Journal of Neuroscience 18:10457–10463.

Rodriguez JP, Coulter M, Miotke J, Meyer RL, Takemaru K-I, Levine JM. 2014. Abrogation of β-catenin signaling in oligodendrocyte precursor cells reduces glial scarring and promotes axon regeneration after CNS injury. J. Neurosci. 34:10285–10297.

Rojas PS, Neira D, Muñoz M, Lavandero S, Fiedler JL. 2014. Serotonin (5‐HT) regulates neurite outgrowth through 5‐HT1A and 5‐HT7 receptors in cultured hippocampal neurons. J. Neurosci. Res. 92:1000–1009.

Rolls A, Shechter R, London A, Segev Y, Jacob-Hirsch J, Amariglio N, Rechavi G, Schwartz M. 2008. Two faces of chondroitin sulfate proteoglycan in spinal cord repair: a

231 role in microglia/macrophage activation. Streit W, editor. PLOS Medicine 5:e171.

Rosdahl JA, Mourton TL, Brady-Kalnay SM. 2002. Protein kinase C delta (PKCdelta) is required for protein tyrosine phosphatase mu (PTPmu)-dependent neurite outgrowth. - PubMed - NCBI. Molecular and Cellular Neuroscience 19:292–306.

Rosen CL, Bunge RP, Ard MD, Wood PM. 1989. Type 1 astrocytes inhibit myelination by adult rat oligodendrocytes in vitro. Journal of Neuroscience 9:3371–3379.

Rotin D, Wallace MJ, Batt J, Fladd CA, Henderson JT, Skarnes W. 1999. Neuronal defects and posterior pituitary hypoplasia in mice lacking the receptor tyrosine phosphatase PTP|[sigma]|. Nat. Genet. 21:334–338.

Ruschel J, Hellal F, Flynn KC, Dupraz S, Elliott DA, Tedeschi A, Bates M, Sliwinski C, Brook G, Dobrindt K, et al. 2015. Systemic administration of epothilone B promotes axon regeneration after spinal cord injury. Science (New York, N.Y.) 348:347–352.

Sajnani-Perez G, Chilton JK, Aricescu AR, Haj F, Stoker AW. 2003. Isoform-specific binding of the tyrosine phosphatase ptpσ to a ligand in developing muscle. Molecular and Cellular Neuroscience 22:37–48.

Sakry D, Neitz A, Singh J, Frischknecht R, Marongiu D, Binamé F, Perera SS, Endres K, Lutz B, Radyushkin K, et al. 2014. Oligodendrocyte Precursor Cells Modulate the Neuronal Network by Activity-Dependent Ectodomain Cleavage of Glial NG2. Barres BA, editor. PLoS Biol 12:e1001993.

Sapieha PS, Duplan L, Uetani N, Joly S, Tremblay ML, Kennedy TE, Di Polo A. 2005. Receptor protein tyrosine phosphatase sigma inhibits axon regrowth in the adult injured CNS. - PubMed - NCBI. Molecular and Cellular Neuroscience 28:625–635.

Sasaki K, Yoshida H. 2015. Organelle autoregulation-stress responses in the ER, Golgi, mitochondria and lysosome. - PubMed - NCBI. Journal of Biochemistry 157:185–195.

Saygili E, Schauerte P, Pekassa M, Saygili E, Rackauskas G, Schwinger RHG, Weis J, Weber C, Marx N, Rana OR. 2011. Sympathetic neurons express and secrete MMP-2 and MT1-MMP to control nerve sprouting via pro-NGF conversion. Cell. Mol. Neurobiol. 31:17–25.

Schaapveld RQ, Schepens JT, Bächner D, Attema J, Wieringa B, Jap PH, Hendriks WJ. 1998. Developmental expression of the cell adhesion molecule-like protein tyrosine phosphatases LAR, RPTPdelta and RPTPsigma in the mouse. Mech. Dev. 77:59–62.

Schachtrup C, Lu P, Jones LL, Lee JK, Lu J, Sachs BD, Zheng B, Akassoglou K. 2007. Fibrinogen inhibits neurite outgrowth via beta 3 integrin-mediated phosphorylation of the EGF receptor. Proceedings of the National Academy of Sciences 104:11814–11819.

232

Schachtrup C, Ryu JK, Helmrick MJ, Vagena E, Galanakis DK, Degen JL, Margolis RU, Akassoglou K. 2010. Fibrinogen triggers astrocyte scar formation by promoting the availability of active TGF-beta after vascular damage. J. Neurosci. 30:5843–5854.

Schmalfeldt M, Bandtlow CE, Dours-Zimmermann MT, Winterhalter KH, Zimmermann DR. 2000. Brain derived versican V2 is a potent inhibitor of axonal growth. Journal of Cell Science 113 ( Pt 5):807–816.

Schwab ME, Strittmatter SM. 2014. Nogo limits neural plasticity and recovery from injury. Current Opinion in Neurobiology 27:53–60.

Schwartz NB. 1977. Regulation of chondroitin sulfate synthesis. Effect of beta-xylosides on synthesis of chondroitin sulfate proteoglycan, chondroitin sulfate chains, and core protein. J Biol Chem 252:6316–6321.

Seeds N, Mikesell S, Vest R, Bugge T, Schaller K, Minor K. 2011. Cathepsin B: an alternative protease for the generation of an aggrecan “metalloproteinase” cleavage neoepitope. - PubMed - NCBI. Cell. Mol. Neurobiol. 31:961–967.

Seeds NW, Basham ME, Haffke SP. 1999. Neuronal migration is retarded in mice lacking the tissue plasminogen activator gene. Proceedings of the National Academy of Sciences 96:14118–14123.

Seeds NW, Williams BL, Bickford PC. 1995. Tissue plasminogen activator induction in Purkinje neurons after cerebellar motor learning. Science 270:1992–1994.

Sendler M, Maertin S, John D, Persike M, Weiss FU, Krüger B, Wartmann T, Wagh P, Halangk W, Schaschke N, et al. 2016. Cathepsin B Activity Initiates Apoptosis via Digestive Protease Activation in Pancreatic Acinar Cells and Experimental Pancreatitis. J Biol Chem 291:14717–14731.

Sevenich L, Joyce JA. 2014. Pericellular proteolysis in cancer. Genes & Development 28:2331–2347.

Shaffer LM, Dority MD, Gupta-Bansal R, Frederickson RC, Younkin SG, Brunden KR. 1995. Amyloid beta protein (A beta) removal by neuroglial cells in culture. Neurobiol. Aging 16:737–745.

Shen Y. 2014. Traffic lights for axon growth: proteoglycans and their neuronal receptors. Neural Regen Res 9:356–361.

Shijo T, Warita H, Suzuki N, Kitajima Y, Ikeda K, Akiyama T, Ono H, Mitsuzawa S, Nishiyama A, Izumi R, et al. 2017. Aberrant astrocytic expression of chondroitin sulfate proteoglycan receptors in a rat model of amyotrophic lateral sclerosis. J. Neurosci. Res. 5:323.

233

Shimoda M, Khokha R. 2013. Proteolytic factors in exosomes. Proteomics 13:1624– 1636.

Siconolfi LB, Seeds NW. 2001. Mice lacking tPA, uPA, or plasminogen genes showed delayed functional recovery after sciatic nerve crush. J. Neurosci. 21:4348–4355.

Siconolfi LB, Seeds NW. 2003. Mice lacking tissue plasminogen activator and urokinase plasminogen activator genes show attenuated matrix metalloproteases activity after sciatic nerve crush. J. Neurosci. Res. 74:430–434.

Siebert JR, Osterhout DJ. 2011. The inhibitory effects of chondroitin sulfate proteoglycans on oligodendrocytes. Journal of Neurochemistry 119:176–188.

Siebert JR, Stelzner DJ, Osterhout DJ. 2011. Chondroitinase treatment following spinal contusion injury increases migration of oligodendrocyte progenitor cells. Experimental Neurology 231:19–29.

Silver DJ, Silver J. 2014. Contributions of chondroitin sulfate proteoglycans to neurodevelopment, injury, and cancer. Current Opinion in Neurobiology 27:171–178.

Siu R, Fladd C, Rotin D. 2006. N-Cadherin Is an In Vivo Substrate for Protein Tyrosine Phosphatase Sigma (PTP ) and Participates in PTP -Mediated Inhibition of Axon Growth. Molecular and Cellular Biology 27:208–219.

Skuljec J, Gudi V, Ulrich R, Frichert K, Yildiz O, Pul R, Voss EV, Wissel K, Baumgärtner W, Stangel M. 2011. Matrix metalloproteinases and their tissue inhibitors in cuprizone-induced demyelination and remyelination of brain white and gray matter. J. Neuropathol. Exp. Neurol. 70:758–769.

Sloane B, Dunn, Honn K. 1981. Lysosomal cathepsin B: correlation with metastatic potential. Science 212:1151–1153.

Snow DM, Lemmon V, Carrino DA, Caplan AI, Silver J. 1990. Sulfated proteoglycans in astroglial barriers inhibit neurite outgrowth in vitro. Experimental Neurology 109:111– 130.

Sobel RA, Ahmed AS. 2001. White Matter Extracellular Matrix Chondroitin Sulfate/Dermatan Sulfate Proteoglycans in Multiple Sclerosis. J. Neuropathol. Exp. Neurol. 60:1198–1207.

Sofroniew MV, Vinters HV. 2009. Astrocytes: biology and pathology. Acta Neuropathol 119:7–35.

Son Y-J. 2015. Synapsing with NG2 cells (polydendrocytes), unappreciated barrier to axon regeneration? Neural Regen Res 10:346–348.

234

Soori M, Lu G, Mason RW. 2016. Cathepsin Inhibition Prevents Autophagic Protein Turnover and Downregulates Insulin Growth Factor-1 Receptor–Mediated Signaling in Neuroblastoma. J Pharmacol Exp Ther 356:375–386.

Starckx S, Van den Steen PE, Verbeek R, van Noort JM, Opdenakker G. 2003. A novel rationale for inhibition of gelatinase B in multiple sclerosis: MMP-9 destroys αB- crystallin and generates a promiscuous T cell epitope. Journal of Neuroimmunology 141:47–57.

Stryker E, Johnson KG. 2007. LAR, liprin α and the regulation of active zone morphogenesis. Journal of Cell Science 120:3723–3728.

Sypecka J, Ziemka Nalecz M, Dragun Szymczak P, Zalewska T. 2017. A simple, xeno- free method for oligodendrocyte generation from human neural stem cells derived from umbilical cord: engagement of gelatinases in cell commitment and differentiation. J Tissue Eng Regen Med 11:1442–1455.

Takahashi H, Arstikaitis P, Prasad T, Bartlett TE, Wang YT, Murphy TH, Craig AM. 2011. Postsynaptic TrkC and presynaptic PTPsigma function as a bidirectional excitatory synaptic organizing complex. 69:287–303.

Takahashi H, Craig AM. 2013 Jul 5. Protein tyrosine phosphatases PTP. Trends in Neurosciences:1–13.

Takesian AE, Hensch TK. 2013. Balancing Plasticity/Stability Across Brain Development. Progress in brain research 207:3–34.

Tang X, Davies JE, Davies SJA. 2003. Changes in distribution, cell associations, and protein expression levels of NG2, neurocan, phosphacan, brevican, versican V2, and tenascin-C during acute to chronic maturation of spinal cord scar tissue. J. Neurosci. Res. 71:427–444.

Tatti M, Motta M, Di Bartolomeo S, Scarpa S, Cianfanelli V, Cecconi F, Salvioli R. 2012. Reduced cathepsins B and D cause impaired autophagic degradation that can be almost completely restored by overexpression of these two proteases in Sap C-deficient fibroblasts. Hum. Mol. Genet. 21:5159–5173.

Taylor AR, Welsh CJ, Young C, Spoor E, Kerwin SC, Griffin JF, Levine GJ, Cohen ND, Levine JM. 2014. Cerebrospinal fluid inflammatory cytokines and chemokines in naturally occurring canine spinal cord injury. Journal of Neurotrauma 31:1561–1569.

Taylor DL, Pirianov G, Holland S, McGinnity CJ, Norman AL, Reali C, Diemel LT, Gveric D, Yeung D, Mehmet H. 2010. Attenuation of proliferation in oligodendrocyte precursor cells by activated microglia. J. Neurosci. Res. 88:1632–1644.

Teuscher C, Blankenhorn EP, Hickey WF. 1987. Differential susceptibility to actively

235 induced experimental allergic encephalomyelitis and experimental allergic orchitis among BALB/c substrains. Cell. Immunol. 110:294–304.

Tholen S, Biniossek ML, Gansz M, Ahrens TD, Schlimpert M, Kizhakkedathu JN, Reinheckel T, Schilling O. 2014. Double deficiency of cathepsins B and L results in massive secretome alterations and suggests a degradative cathepsin-MMP axis. Cell. Mol. Life Sci. 71:899–916.

Thompson KM, Uetani N, Manitt C, Elchebly M, Tremblay ML, Kennedy TE. 2003. Receptor protein tyrosine phosphatase sigma inhibits axonal regeneration and the rate of axon extension. Molecular and Cellular Neuroscience 23:681–692.

Tom VJ, Steinmetz MP, Miller JH, Doller CM, Silver J. 2004. Studies on the Development and Behavior of the Dystrophic Growth Cone, the Hallmark of Regeneration Failure, in an In Vitro Model of the Glial Scar and after Spinal Cord Injury. Journal of Neuroscience 24:6531–6539.

Tom VJ, Steinmetz MP, Miller JH, Doller CM, Silver J. Studies on the Development and Behavior of the Dystrophic Growth Cone, the Hallmark of Regeneration Failure, in an In Vitro Model of the Glial Scar and after Spinal Cord Injury.

Tran AP, Warren PM, Silver J. 2018. The Biology of Regeneration Failure and Success After Spinal Cord Injury. Physiological Reviews 98:881–917.

Tremblay ML. 2009. The PTP family photo album. Cell 136:213–214.

Trentini A, Castellazzi M, Cervellati C, Manfrinato MC, Tamborino C, Hanau S, Volta CA, Baldi E, Kostic V, Drulovic J, et al. 2016. Interplay between Matrix Metalloproteinase-9, Matrix Metalloproteinase-2, and Interleukins in Multiple Sclerosis Patients. Dis. Markers 2016:3672353–9.

Trivedi A, Olivas AD, Noble-Haeusslein LJ. 2006. Inflammation and Spinal Cord Injury: Infiltrating Leukocytes as Determinants of Injury and Repair Processes. Clinical Neuroscience Research 6:283–292.

Tsai H-H, Niu J, Munji R, Davalos D, Chang J, Zhang H, Tien A-C, Kuo CJ, Chan JR, Daneman R, et al. 2016. Oligodendrocyte precursors migrate along vasculature in the developing nervous system. Science (New York, N.Y.) 351:379–384.

Tsai HH, Li H, Fuentealba LC, Molofsky AV, Taveira-Marques R, Zhuang H, Tenney A, Murnen AT, Fancy SPJ, Merkle F, et al. 2012. Regional astrocyte allocation regulates CNS synaptogenesis and repair. Science 337:358–362.

Turk V, Bode W. 2001. The cystatins: Protein inhibitors of cysteine proteinases. FEBS Letters 285:213–219.

236

Turk V, Stoka V, Vasiljeva O, Renko M, Sun T, Turk B, Turk D. 2012. Cysteine cathepsins: from structure, function and regulation to new frontiers. Biochim. Biophys. Acta 1824:68–88.

Uhm JH, Dooley NP, Oh LY, Yong VW. 1998. Oligodendrocytes utilize a matrix metalloproteinase, MMP-9, to extend processes along an astrocyte extracellular matrix. Glia 22:53–63.

Um JW, Kim KH, Park BS, Choi Y, Kim D, Kim CY, Kim SJ, Kim M, Ko JS, Lee S-G, et al. 2014. Structural basis for LAR-RPTP/Slitrk complex-mediated synaptic adhesion. Nat Comms 5:5423.

Um JW, Ko J. 2013. LAR-RPTPs: synaptic adhesion molecules that shape synapse development. Trends in Cell Biology 23:465–475.

Valente MM, Allen M, Bortolotto V, Lim ST, Conant K, Grilli M. 2015. The MMP- 1/PAR-1 Axis Enhances Proliferation and Neuronal Differentiation of Adult Hippocampal Neural Progenitor Cells. Neural Plasticity 2015:646595–10.

Valenzuela JC, Heise C, Franken G, Singh J, Schweitzer B, Seidenbecher CI, Frischknecht R. 2014. Hyaluronan-based extracellular matrix under conditions of homeostatic plasticity. Phil. Trans. R. Soc. B 369:20130606–896.

Veeravalli KK, Dasari VR, Rao JS. 2012. Regulation of proteases after spinal cord injury. Journal of Neurotrauma 29:2251–2262.

Veeravalli KK, Dasari VR, Tsung AJ, Dinh DH, Gujrati M, Fassett D, Rao JS. 2009. Human umbilical cord blood stem cells upregulate matrix metalloproteinase-2 in rats after spinal cord injury. Neurobiology of Disease 36:200–212.

Vela JM, Molina-Holgado E, Arévalo-Martín A, Almazán G, Guaza C. 2002. Interleukin- 1 regulates proliferation and differentiation of oligodendrocyte progenitor cells. Mol. Cell. Neurosci. 20:489–502.

Verma S, Dixit R, Pandey KC. 2016. Cysteine Proteases: Modes of Activation and Future Prospects as Pharmacological Targets. Frontiers in Pharmacology 7:30433.

Végh MJ, Heldring CM, Kamphuis W, Hijazi S, Timmerman AJ, Li KW, van Nierop P, Mansvelder HD, Hol EM, Smit AB, et al. 2014. Reducing hippocampal extracellular matrix reverses early memory deficits in a mouse model of Alzheimer's disease. Acta Neuropathol Commun 2:76.

Viapiano MS, Matthews RT. 2006. From barriers to bridges: chondroitin sulfate proteoglycans in neuropathology. Trends in Molecular Medicine 12:488–496.

Victor BC, Anbalagan A, Mohamed MM, Sloane BF, Cavallo-Medved D. 2011.

237

Inhibition of cathepsin B activity attenuates extracellular matrix degradation and inflammatory breast cancer invasion. Breast Cancer Res. 13:R115.

Wagner J, Boerboom D, Tremblay ML. 1994. Molecular Cloning and Tissue-Specific RNA Processing of a Murine Receptor-Type Protein Tyrosine Phosphatase. Eur J Biochem 226:773–782.

Wallace MJ, Fladd C, Batt J, Rotin D. 1998. The Second Catalytic Domain of Protein Tyrosine Phosphatase δ (PTPδ) Binds to and Inhibits the First Catalytic Domain of PTPς. Molecular and Cellular Biology 18:2608–2616.

Wang C, Sun B, Zhou Y, Grubb A, Gan L. 2012. Cathepsin B Degrades Amyloid-β in Mice Expressing Wild-type Human Amyloid Precursor Protein. J Biol Chem 287:39834– 39841.

Wang G, Shi Y, Jiang X, Leak RK, Hu X, Wu Y, Pu H, Li W-W, Tang B, Wang Y, et al. 2015. HDAC inhibition prevents white matter injury by modulating microglia/macrophage polarization through the GSK3β/PTEN/Akt axis. Proc. Natl. Acad. Sci. U.S.A. 112:2853–2858.

Wang H-F, Liu X-K, Li R, Zhang P, Chu Z, Wang C-L, Liu H-R, Qi J, Lv G-Y, Wang G- Y, et al. 2017. Effect of glial cells on remyelination after spinal cord injury. Neural Regen Res 12:1724–1732.

Wells JEA, Rice TK, Nuttall RK, Edwards DR, Zekki H, Rivest S, Yong VW. 2003. An adverse role for matrix metalloproteinase 12 after spinal cord injury in mice. J. Neurosci. 23:10107–10115.

Wiese S, Faissner A. 2015. The role of extracellular matrix in spinal cord development. - PubMed - NCBI. Experimental Neurology 274:90–99.

Williams R, Yao H, Peng F, Yang Y, Bethel-Brown C, Buch S. 2010. Cooperative induction of CXCL10 involves NADPH oxidase: Implications for HIV dementia. Glia 58:611–621.

Wilson MT, Snow DM. 2000. Chondroitin sulfate proteoglycan expression pattern in hippocampal development: Potential regulation of axon tract formation. Journal of Comparative Neurology 424:532–546.

Wolswijk G. 1998. Chronic stage multiple sclerosis lesions contain a relatively quiescent population of oligodendrocyte precursor cells. Journal of Neuroscience 18:601–609.

Woo J, Kwon S-K, Choi S, Kim S, Lee J-R, Dunah AW, Sheng M, Kim E. 2009. Trans- synaptic adhesion between NGL-3 and LAR regulates the formation of excitatory synapses. Nat Neurosci 12:428–437.

238

Wu C-L, Hardy S, Aubry I, Landry M, Haggarty A, Saragovi HU, Tremblay ML. 2017. Identification of function-regulating antibodies targeting the receptor protein tyrosine phosphatase sigma ectodomain. - PubMed - NCBI. Sato M, editor. PLoS ONE 12:e0178489.

Xu B, Park D, Ohtake Y, Li H, Hayat U, Liu J, Selzer ME, Longo FM, Li S. 2015. Role of CSPG receptor LAR phosphatase in restricting axon regeneration after CNS injury. Neurobiology of disease 73:36–48.

Xu Y, Fisher GJ. 2012. Receptor type protein tyrosine phosphatases (RPTPs) - roles in signal transduction and human disease. J Cell Commun Signal 6:125–138.

Yan H, Grossman A, Wang H, D'Eustachio P, Mossie K, Musacchio JM, Silvennoinen O, Schlessinger J. 1993. A novel receptor tyrosine phosphatase-sigma that is highly expressed in the nervous system. J Biol Chem 268:24880–24886.

Yang D-S, Stavrides P, Mohan PS, Kaushik S, Kumar A, Ohno M, Schmidt SD, Wesson D, Bandyopadhyay U, Jiang Y, Pawlik M, Peterhoff CM, Yang AJ, Wilson DA, St George-Hyslop P, Westaway D, Mathews PM, Levy E, Cuervo AM, Nixon RA. 2011a. Reversal of autophagy dysfunction in the TgCRND8 mouse model of Alzheimer's disease ameliorates amyloid pathologies and memory deficits. Brain 134:258–277.

Yang D-S, Stavrides P, Mohan PS, Kaushik S, Kumar A, Ohno M, Schmidt SD, Wesson DW, Bandyopadhyay U, Jiang Y, Pawlik M, Peterhoff CM, Yang AJ, Wilson DA, St George-Hyslop P, Westaway D, Mathews PM, Levy E, Cuervo AM, Nixon RA. 2011b. Therapeutic effects of remediating autophagy failure in a mouse model of Alzheimer disease by enhancing lysosomal proteolysis. Autophagy 7:788–789.

Yang S, Cacquevel M, Saksida LM, Bussey TJ, Schneider BL, Aebischer P, Melani R, Pizzorusso T, Fawcett JW, Spillantini MG. 2015. Perineuronal net digestion with chondroitinase restores memory in mice with tau pathology. Experimental Neurology 265:48–58.

Yang T, Yin W, Derevyanny VD, Moore LA, Longo FM. 2005. Identification of an ectodomain within the LAR protein tyrosine phosphatase receptor that binds homophilically and activates signalling pathways promoting neurite outgrowth. European Journal of Neuroscience 22:2159–2170.

Yang Y, Jalal FY, Thompson JF, Walker EJ, Candelario-Jalil E, Li L, Reichard RR, Ben C, Sang Q-X, Cunningham LA, et al. 2011. Tissue inhibitor of metalloproteinases-3 mediates the death of immature oligodendrocytes via TNF-α/TACE in focal cerebral ischemia in mice. J Neuroinflammation 8:108.

Yi J-H, Katagiri Y, Susarla B, Figge D, Symes AJ, Geller HM. 2012. Alterations in sulfated chondroitin glycosaminoglycans following controlled cortical impact injury in mice. Journal of Comparative Neurology 520:3295–3313.

239

Yi J-H, Katagiri Y, Yu P, Lourie J, Bangayan NJ, Symes AJ, Geller HM. 2014. Receptor protein tyrosine phosphatase σ binds to neurons in the adult mouse brain. Experimental Neurology 255:12–18.

Shen Y, Tenney AP, Busch SA, Horn KP, Cuascut FX, Liu K, He A, Silver, J, Flanagan JG. 2009. PTP Is a Receptor for Chondroitin Sulfate Proteoglycan, an Inhibitor of Neural Regeneration. Science 326:592–596.

Yong VW, Agrawal SM, Stirling DP. 2007. Targeting MMPs in acute and chronic neurological conditions. Neurotherapeutics 4:580–589.

Youn GS, Ju SM, Choi SY, Park J. 2015. HDAC6 mediates HIV‐1 tat‐induced proinflammatory responses by regulating MAPK‐NF‐kappaB/AP‐1 pathways in astrocytes. Glia 63:1953–1965.

Yu Q, Stamenkovic I. 1999. Localization of matrix metalloproteinase 9 to the cell surface provides a mechanism for CD44-mediated tumor invasion. Genes & Development 13:35– 48.

Yui S, Fujita N, Chung C-S, Morita M, Nishimura R. 2014. Olfactory ensheathing cells (OECs) degrade neurocan in injured spinal cord by secreting matrix metalloproteinase-2 in a rat contusion model. Jpn. J. Vet. Res. 62:151–162.

Yunker PJ, Still T, Lohr MA, Yodh AG. 2011. Suppression of the coffee-ring effect by shape-dependent capillary interactions. Nature 476:308–311.

Zaiden M, Beit-Yannai E. 2015. Receptor Protein Tyrosine Phosphatase Sigma (RPTP-σ) Increases pro-MMP Activity in a Trabecular Meshwork Cell Line Following Oxidative Stress Conditions. Invest. Ophthalmol. Vis. Sci. 56:5720–11.

Zhang Haoqian, Adwanikar H, Werb Z, Noble-Haeusslein LJ. 2010. Matrix metalloproteinases and neurotrauma: evolving roles in injury and reparative processes. Neuroscientist 16:156–170.

Zhang Haoqian, Chang M, Hansen CN, Basso DM, Noble-Haeusslein LJ. 2011. Role of Matrix Metalloproteinases and Therapeutic Benefits of Their Inhibition in Spinal Cord Injury. Neurotherapeutics 8:206–220.

Zhang Hui, Jarjour AA, Boyd A, Williams A. 2011. Central nervous system remyelination in culture — A tool for multiple sclerosis research. Experimental Neurology 230:138–148.

Zhang Rui, Zhang C, Zhao Q, Li D. 2013. Spectrin: Structure, function and disease. Sci. China Life Sci. 56:1076–1085.

Zhang Ye, Chen K, Sloan SA, Bennett ML, Scholze AR, O'Keeffe S, Phatnani HP,

240

Guarnieri P, Caneda C, Ruderisch N, et al. 2014. An RNA-sequencing transcriptome and splicing database of glia, neurons, and vascular cells of the cerebral cortex. J. Neurosci. 34:11929–11947.

Zhou J, Tan S-H, Nicolas V, Bauvy C, Yang N-D, Zhang J, Xue Y, Codogno P, Shen H- M. 2013. Activation of lysosomal function in the course of autophagy via mTORC1 suppression and autophagosome-lysosome fusion. Cell Research 2013 23:4 23:cr201311– 523.

Zuo J, Ferguson TA, Hernandez YJ, Stetler-Stevenson WG, Muir D. 1998. Neuronal matrix metalloproteinase-2 degrades and inactivates a neurite-inhibiting chondroitin sulfate proteoglycan. Journal of Neuroscience 18:5203–5211.

241