University of Nebraska Medical Center DigitalCommons@UNMC

Theses & Dissertations Graduate Studies

Fall 12-15-2017

Effects of GM-CSF on Dendritic Cells and Regulatory T cells in Parkinson’s Disease Patients and Models of Parkinson’s Disease

Charles Schutt University of Nebraska Medical Center

Follow this and additional works at: https://digitalcommons.unmc.edu/etd

Part of the Immunopathology Commons, and the Other Neuroscience and Neurobiology Commons

Recommended Citation Schutt, Charles, "Effects of GM-CSF on Dendritic Cells and Regulatory T cells in Parkinson’s Disease Patients and Models of Parkinson’s Disease" (2017). Theses & Dissertations. 236. https://digitalcommons.unmc.edu/etd/236

This Dissertation is brought to you for free and open access by the Graduate Studies at DigitalCommons@UNMC. It has been accepted for inclusion in Theses & Dissertations by an authorized administrator of DigitalCommons@UNMC. For more information, please contact [email protected]. Effects of GM-CSF on Dendritic Cells and Regulatory T cells in Parkinson’s Disease

Patients and Models of Parkinson’s Disease

By

Charles Schutt

A Dissertation

Presented to the Faculty of

the Graduate College in the University of Nebraska

in Partial Fulfillment of the Requirements

for the Degree of Doctor of Philosophy

Department of Pharmacology and Experimental Neuroscience

Under the Supervision of Professor R. Lee Mosley

University of Nebraska Medical Center

Omaha, Nebraska

November, 2017

Supervisory Committee:

Howard E. Gendelman M.D. Larisa Y. Poluektova, M.D., Ph.D.

Tammy L Kielian Ph. D. Joyce C. Solheim Ph. D.

ii

ACKNOWLEDGMENTS

There are too many people who have contributed to my graduate studies to acknowledge all of them, but several people need to be named. First, I would like to thank my advisor Dr. R. Lee Mosley for giving me the opportunity to pursue my careers goals. I am grateful for his time, patience, support, and guidance over the years spent in his laboratory. Through his efforts, I think I have become a better student, scientist, and person.

I would also like to acknowledge my committee members, Drs. Howard,

Gendelman, Tammy Kielian, Larisa Poluektova, and Joyce Solheim for their time, attention, and guidance. I am grateful for their comments, criticisms, encouragement, and support. This support is instrumental for my development as a student and a scientist. Especially, I would like to thank Dr. Gendelman for his continued financial support for these projects, especially for allowing me to play a role in the clinical trial.

I would like to acknowledge all the past and current members of the Drs. Mosley and Gendelman laboratories. I would like to thank Dr. Kristi Anderson, Bhagya Dyavar

Shetty, Dr. Katherine Estes, Dr. Lisa Kosloski, Dr. Elizabeth Kosmacek, Max Kuenstling,

Yamen Lu, Dr. Jatin Machhi, Krista Namminga, Katherine Olson, Adam Szlachetka

DDS, Rebecca Wilshusen, and Dr. Yuning Zhang for all their help planning, executing and analyzing experiments. I was grateful for the opportunity to learn from and work with each of you. I would also like to thank summer students Anna Miller, Katie Schueth,

Katie Zheng, Sarah Whitmire, Chaoran Ji, Keith Prive, and X. Isabel Heifetz Ament for all their hard work and help generating some of the included data and for the opportunity

I was given to teach them and improve my ability to instruct trainees. I would also like to acknowledge Dr. Shilpa Buch’s laboratory for their help performing luminex assays, Dr.

Steven Bonesera’s laboratory for the use of their thermocycler for all PCR array iii

experiments, the University of Nebraska Medical Center Flow Cytometry Research Core

Facility, and graduate students Sam Johnson, Johnathan Herskovitz, and Nathan Smith for their help characterizing bone marrow dendritic cells.

I would also like to thank the faculty and staff in the Pharmacology and

Experimental Neuroscience Department. I am grateful for their help in navigating my time at the Nebraska Medical Center and for delivering lectures and facilitating my research projects.

There are a great number of people who played a role in my development as a student and a scientist that I would like to acknowledge. I would like to thank Mr. Mark

Stallman and Mr. Jerry Van Dyck, two high school science teachers who were instrumental turning an interest in science into a career path. I would like to thank Dr.

Donald Stratton and Dr. Dean Hoganson, my two advisors at Drake University who guided me toward my goal of going to graduate school. I would like to thank the members of the Forage Additives Research Group at Pioneer Hi-Bred International who gave me my first research experiences which still shape how I plan and execute experiments. I would like to thank Dr. Jason Bartz, my advisor at Creighton University during my Master’s thesis work as well as Dr. Anthony Kincaid, Dr. Ron Shikiya, Dr.

Jacob Ayers, Michelle Kramer, Dr. Sam Saunders, Dr. Qi Yuan, Dr. Katie Langenfeld,

Melissa Clouse, Tom Eckland, and Maria Christensen for their guidance, patience, and support during my time at Creighton University and beyond. The lessons learned at

Creighton University have shaped me as a person and as a scientist. Lastly, I would like to acknowledge the Institute for Environmental Health for giving me a chance to develop as a leader and a manager.

Lastly, I would like to thank my parents, Robert and Kathi Schutt, my sister,

Rebecca Jackson, my brother-in-law, Joshua Jackson, my nephew, Thomas Jackson, iv

my nieces, Tabitha Jackson and Miriam Jackson, and countless other family members and friends for their encouragement, support and prayers. They have meant the world to me.

This dissertation is dedicated to Sue Schutt and Steve Russo.

Your love and support has been missed.

v

ABSTRACT

Parkinson’s disease (PD) is the most common neurodegenerative movement disorder. Pathologically, loss of nigrostriatal neurons and dopamine released by these neurons are responsible for PD motor symptoms. During PD, activation of resident microglia and infiltrating lymphocytes leads to progressive neuroinflammation and reduction in the number and function of regulatory immune cells. Neuroinflammation contributes to progressive neurodegeneration and declining motor function. Reducing neuroinflammation is the target for novel PD therapeutics. Our goal is to increase the number and function of regulatory T cells (Tregs) in PD patients to decrease neuroinflammation and reduce PD symptoms. One potential therapy is granulocyte- macrophage colony stimulating factor (GM-CSF) which induces Tregs, decreases neuroinflammation, and protects dopaminergic neurons in the 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine (MPTP) model of PD.

In a Phase 1 trial, recombinant human GM-CSF (sargramostim) was well- tolerated in PD patients, increased Treg frequency and function, and improved motor function. Expression of helper T cell-related genes in CD4+CD25- cells in blood was determined by PCR array. Sargramostim increased expression of both pro- and anti- inflammatory cytokine genes supporting the notion that sargramostim alters the immune response by increasing the expression of immune mediators, including anti-inflammatory genes.

As T cells do not express GM-CSF receptors and to explore myeloid-mediated

Treg induction, GM-CSF-induced bone marrow-derived dendritic cells were further cultured with GM-CSF and/or stimulated with nitrated α-synuclein. Continued culture with

GM-CSF yielded little change in dendritic cells as determined by surface co-stimulatory molecule expression, and proinflammatory cytokine expression and release; however, vi

their ability to induce Tregs was diminished. In contrast, stimulation with nitrated α- synuclein, regardless of continued culture in GM-CSF, increases proinflammatory gene expression by dendritic cells, but showed variable effects on Treg induction. In the

MPTP model, adoptive transfer of GM-CSF-induced tolerogenic dendritic cells protect dopaminergic neurons in the substantia nigra, decrease neuroinflammation, and increase splenic Tregs in a fashion similar to direct administration of GM-CSF.

In conclusion, GM-CSF induces Tregs in part by acting on dendritic cells to change their response to stimulation. The data suggest that GM-CSF may not suppress neuroinflammation directly, but rather alters the immune response with increased expression of anti-inflammatory mediators and induction of Tregs. Moreover, the introduction of nitrated α-synuclein and possibly other misfolded proteins diminishes homeostasis and Treg induction.

vii

TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... ii

ABSTRACT ...... v

LIST OF FIGURES ...... ix

LIST OF TABLES ...... xi

LIST OF ABBREVIATIONS ...... xii

CHAPTER 1 ...... 1

INTRODUCTION ...... 1

PARKSINSON’S DISEASE ...... 1

NEUROINFLAMATION AND PARKINSON’S DISEASE ...... 11

GRANUCLOCYTE-MACROPHAGE COLONY STIMULATING FACTOR (GM-CSF)

...... 19

MECHANISMS OF TREG INDUCTION ...... 23

SUMMARY AND CONCLUSIONS ...... 24

CHAPTER TWO ...... 26

GENE EXPRESSION IN T RESPONDER CELLS IN PD PATIENTS TREATED WITH

SARGRAMOSTIM ...... 26

ABSTRACT ...... 26

INTRODUCTION ...... 27

METHODS ...... 30

RESULTS ...... 40

DISCUSSION ...... 65 viii

CHAPTER THREE ...... 74

GM-CSF-GENERATED BONE MARROW DERIVED DCs INDUCE REGULATORY T

CELLS AND ARE NEUROPROTECTIVE IN MPTP INTOXICATED NICE ...... 74

ABSTRACT ...... 74

INTRODUCTION ...... 75

METHODS ...... 78

RESULTS ...... 101

DISCUSSION ...... 168

CHAPTER FOUR ...... 176

CYTOKINE ENVIRONMENT IN THE VENTRAL MIDBRAIN AND CERVICAL LYMPH

NODE TWO DAYS AFTER MPTP INTOXICATION ...... 176

ABSTRACT ...... 176

INTRODUCTION ...... 177

MATERIALS and METHODS ...... 180

RESULTS ...... 185

DISCUSSION ...... 196

CHAPTER 5 ...... 199

DISCUSSION AND FUTURE DIRECTIONS ...... 199

DISCUSSION ...... 199

FUTURE DIRECTIONS ...... 209

BIBLIOGRAPHY ...... 214

ix

LIST OF FIGURES

Figure 2.1 Flow cytometric analysis of CD4+ CD25- cells ...... 41

Table 2.2 RNA quality analysis ...... 44

Figure 2.2 Gene expression comparing PD patients to healthy controls pre-treatment with and without CD3/CD28 stimulation ...... 46

Figure 2.6 Gene expression comparing unstimulated CD4+ CD25- T cells from placebo- treated PD patients compared to sargramostim-treated PD patients ...... 59

Figure 2.7 Gene expression related to Treg frequency in placebo- and sargramostim- treated PD patients ...... 63

Figure 2.8 Ingenuity pathway analysis of genes altered by sargramostim in unstimulated and stimulated CD4+ T cells ...... 66

Figure 3.1 Method for BMDC generation ...... 80

Figure 3.2 GM-CSF pre-treatment, but not co-treatment, decrease the LPS- and N-α-

Syn-induced increase in nitrite release in DC2.4 and DC3.2 cells with no decrease in cell viability ...... 103

Figure 3.3 Flow cytometric analysis of DC2.4 cells after GM-CSF pre-treatment and LPS and N-α-Syn stimulation ...... 106

Figure 3.4 Flow cytometric analysis of BMDCs after differentiation in GM-CSF or GM-

CSF and IL-4 prior to stimulation with LPS ...... 109

Figure 3.5 BMDC culture with GM-CSF mitigates the LPS- or N-α-Syn-induced increase in the nitrite concentration in the supernatant, without effecting cell viability ...... 112

Figure 3.6 Surface expression of co-stimulatory molecules on CD11c+ BMDCs ...... 115

Figure 3.7 Gene expression of GM-CSF, N-α-Syn, or GM-CSF+N-α-Syn-treated BMDCs compared to media-cultured, unstimulated BMDCs ...... 118

Figure 3.8 Nitrite concentration in supernatant from BMDCs ...... 122 x

Figure 3.9 Cytokines and chemokines released from BMDCs ...... 125

Figure 3.10 Expression of IDO and kynurenine release in the supernatant ...... 128

Figure 3.11 BMDCs induce functional Tregs in vitro ...... 131

Figure 3.12 Correlation of BMDC surface markers with Treg induction ...... 135

Figure 3.13 BMDCs cause both induction and proliferation of Tregs ...... 138

Figure 3.14 Adoptive transfer of BMDCs before MPTP does not protect dopaminergic neurons ...... 140

Figure 3.15 BMDCs are neuroprotective in the MPTP mouse model ...... 143

Figure 3.16 BMDCs decrease the number of reactive microglia in the MPTP model ... 145

Figure 3.17 Gene expression of MPTP and BMDCs + MPTP midbrain compared to PBS control midbrain ...... 148

Figure 3.18 Treg frequency and function after the transfer of BMDCs ...... 152

Figure 3.19 Treg frequency and function after adoptive transfer of BMDCs prior to MPTP

...... 154

Figure 3.20 BMDC-induced Tregs are not neuroprotective ...... 157

Figure 3.21 BMDC supernatant does not protect MES23.5 cells viability after culture

MPP+ and BV2 supernatant ...... 160

Figure 3.22 Release of cytokines and nitrite from BV2 cells treated with BMDC supernatant and LPS ...... 164

Figure 3.23 Frequency of DCs in the spleen after GM-CSF administration ...... 166

Figure 4.1 Flow cytometric profile in lymph nodes after MPTP intoxication ...... 186

Figure 4.2 Changes in cytokines and chemokines in the cervical lymph nodes (cLN) and ventral midbrain (VMB) after MPTP intoxication ...... 189

Figure 4.4 Map of dysregulated genes in the cLN following MPTP intoxication ...... 194

xi

LIST OF TABLES

Table 2.1 Study design ...... 322

Table 2.2 RNA quality analysis ...... 444

xii

LIST OF ABBREVIATIONS 6-OHDA – 6-hydroxydopamine

α-Syn – alpha synuclein

APCs – antigen-presenting cells b/aLN – brachial and axillary lymph nodes

BCA – bicinchoninic acid assay

BBB – blood brain barrier

BMDC – bone marrow-derived dendritic cell

BSA – bovine serum albumin

BDNF – brain derived neurotropic factor

Breg – B regulatory cell

CD – cluster of differentiation

CFSE – carboxyfluorescein succinimidyl ester cLN – cervical lymph nodes

CRC – clinical research center

DCs – dendritic cells

DMSO – dimethyl sulfoxide

DPBS – Dulbecco’s phosphate-buffered saline

ECL – enhanced chemiluminescence reagent

EDTA – ethylenediaminetetraacetic acid

EAE – experimental autoimmune encephalomyelitis xiii

FBS – fetal bovine serum

FDA – United States Food and Drug Administration

FSB – flow cytometry stain buffer

Ga – gauge

GM-CSF – granulocyte-macrophage colony stimulating factor

HBSS – Hanks buffered saline solution

HNE – 4-hydroxy-2-nonenal

HRP – horseradish peroxidase

IDO – indolamine 2,3-deoxygenase

IPA – Ingenuity pathway analysis

IACUC – institutional animal care and use committee

IFNγ – interferon gamma

IL – interleukin iLN – inguinal lymph nodes i.p. – intraperitoneal i.v. - intravenous

Jag-1 – jagged-1

LPS – lipopolysaccharides

LSM – lymphocyte separation media

MAO-B – manoamine oxidase-B

MDSCs – myeloid-derived suppressor cells xiv

MFI – mean fluorescence intensity

MHC – major histocompatibility complex

MPP+ – 1-methyl-4-phenylpyridinium

MPTP – 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

MQ – Milli-Q ultrapure water

MS – multiple sclerosis

NIH – National Institute of Health

NK – natural killer

NO – nitric oxide

N-α-Syn – nitrated alpha synuclein

PAGE – polyacrylamide gel electrophoresis

PBMCs – peripheral blood mononuclear cells

PBS – phosphate-buffered saline

PBST – phosphate-buffered saline with Tween 20

PD – Parkinson’s disease

PET – positron emission tomography

PPAR-γ – peroxisome proliferator-activated receptor gamma

REM – rapid eye movement

RNA-Seq – RNA sequencing

RNS – reactive nitrogen species

ROS – reactive oxygen species xv

RT – room temperature s.c. – subcutaneous injections

TBS – Tris-buffered saline

TGFβ – transforming growth factor beta

TH – tyrosine hydroxylase

TNF – tumor necrosis factor

Teff – effector T cells

Tregs – regulatory T cells

UNMC – University of Nebraska Medical Center

UPDRS – unified Parkinson’s disease rating scale

VIP – vasoactive intestinal peptide

VMB – ventral midbrain

VMAT2 – vesicular monoamine transporter 2 1

CHAPTER 1

INTRODUCTION

PARKSINSON’S DISEASE

Pathology

Parkinson’s disease (PD) is the second most common neurodegenerative disease after Alzheimer’s disease (de Lau and Breteler, 2006; Olanow et al., 2009). PD has several pathological hallmarks. One, is degeneration of the nigrostriatal pathway, the dopaminergic neurons which project rostrally from the substantia nigra pars compacta to the dorsal striatum where they synapse on GABAergic neurons (Jenner,

2008). The loss of nigrostriatal dopaminergic neurons and released dopamine ultimately decrease glutamatergic signaling from the thalamus to the motor cortex resulting in decreased motor function. Dopaminergic neurons in the substantia nigra pars compacta are particularly susceptible to PD degeneration, even compared to other dopaminergic neuron populations in adjacent brain regions. The reasons for increased neuronal susceptibility to damage and loss are not clear. Ideas include, decreased ability of these neurons to scavenge reactive oxygen species (Hornykiewicz and Kish, 1987), decreased expression of neuromelanin (Hirsch et al., 1988), decreased ability to degrade misfolded proteins (Dauer and Przedborski, 2003), increased generation of reactive oxygen species (ROS), decreased mitochondria mass, and decreased myelination which results in these dopaminergic neurons being more sensitive to energy stressors (Surmeier et al., 2011; Haddad and Nakamura, 2015).

Accumulation of Lewy bodies and Lewy neurites, proteinaceous aggregates predominately made up of misfolded, post-translationally modified α-synuclein (α-Syn) is 2

another PD hallmark (Spillantini et al., 1997; Duda et al., 2000; Giasson, 2000; Luk et al., 2009). Lewy bodies were initially identified as eosinophilic inclusions found throughout the basal ganglia and thalamus (Spillantini et al., 1997; Fornai et al., 2003;

Shults, 2006). They are primarily composed of α-Syn, ubiquitin, parkin, microtubule associated protein 1B, phosphorylated IκBα, synphilin-1, amongst other proteins (Jensen et al., 2000; Fornai et al., 2003; Noda et al., 2005). Due to α-Syn abundance at the synaptic terminal, it is putatively involved in vesicular transport and neurotransmitter release (Shults, 2006; Bendor et al., 2013). In the cell, α-Syn is normally found as an unstructured monomer (Uversky et al., 2001) or tetramer (Bartels et al., 2011). When bound to lipid membranes, α-Syn can adopt a more α-helical-rich structure (Davidson et al., 1998; Perrin et al., 2000). In Lewy bodies, α-Syn takes on a fibril structure

(Spillantini et al., 1998) with cross β-sheet conformation (Heise et al., 2005; Vilar et al.,

2008). α-Syn in Lewy bodies also is post-translationally modified, including ubiquitination (Hasegawa et al., 2002; Nonaka et al., 2005), phosphorylation (Fujiwara et al., 2002; Wang et al., 2012), nitration (Duda et al., 2000; Giasson, 2000), and truncation

(Li et al., 2005; Anderson et al., 2006). While post-translational modifications such as nitration increase the propensity of α-Syn to fold into oligomers (Yamin et al., 2003;

Hodara et al., 2004; Uversky et al., 2005), it is unclear in PD if α-Syn becomes modified prior to or after misfolding and forming Lewy bodies. It is also unclear if Lewy body formation results in or causes PD pathology. The Braak hypothesis posits that α-Syn aggregation starts in the enteric nervous system (Braak et al., 2003; Braak et al., 2004).

According to the hypothesis, misfolded α-Syn accumulates and spreads neuron to neuron into and throughout the central nervous system. The resuting Lewy bodies may cause neurodegeneration. Evidence for the Braak hypothesis includes α-Syn expression being highest in regions with the highest accumulation of Lewy bodies, the detection of Lewy body pathology in neuronal transplants, the transfer of α-Syn between 3

neurons and between neurons and astrocytes ex vivo, and pathology induced by transfer of exogenous α-Syn into rodents (McCann et al., 2016; Karampetsou et al.,

2017; Loria et al., 2017; Rietdijk et al., 2017). Despite the above evidence, there is evidence that the Braak hypothesis is not completely correct, such as α-Syn from PD patients being unable to cause pathology in rodents, the severity of PD symptoms not always correlating with the degree of Lewy body pathology, and detection of Lewy body pathology in non-PD patients (Prusiner et al., 2015; McCann et al., 2016). Future studies will rectify the different lines of evidence.

A final hallmark of PD is neuroinflammation which will be discussed in detail in a later section.

Symptoms

PD is generally diagnosed by the symptoms presented. The four cardinal motor symptoms of PD are resting tremor, rigidity, bradykinesia, and postural instability with gait disturbance (Olanow et al., 2009). All four cardinal symptoms may not be present in all PD patients, but the presence of these symptoms forms the basis of a PD diagnosis.

At the time of PD cardinal symptom presentation, 50-60% of nigral dopaminergic neurons are already lost (Agid, 1991). Many non-motor symptoms are associated with the enteric nervous system which present in PD patients prior to the presentation of the cardinal motor symptoms. For example, PD patients may develop olfactory disruptions, drooling, urinary disturbances, sexual dysfunction, orthostatic hypotension, fatigue, rapid eye movement (REM) sleep dysfunction, and insomnia (Olanow et al., 2009; Ou et al.,

2016). These symptoms are all non-descript and may be associated with other disorders, and as a result, these symptoms are not used as PD diagnostic criteria.

These non-motor symptoms may be due to the loss of neurons caused by the accumulation of α-Syn aggregates and act as evidence for the Braak hypothesis. 4

Another major set of non-motor symptoms is neurocognitive symptoms such as depression, delirium, hallucinations, anxiety and dementia (Olanow et al., 2009; Ou et al., 2016). Approximately 50% of PD patients suffer from these cognitive symptoms and about 30% of PD patients are diagnosed with dementia (Aarsland et al., 2005) making treatment of cognitive symptoms an important aspect PD patient care (Hely et al., 2005;

Olanow et al., 2009). The development of these non-motor symptoms may suggest a link between PD and Alzheimer’s disease. In line with this idea, PD patients may display amyloid beta plaques, similar to Alzheimer’s disease patients (Mastaglia et al., 2003;

Lashley et al., 2008), and Alzheimer’s disease patients plaques contain α-Syn (Uéda et al., 1993).

Risk factors

Age is the number one risk factor for PD (de Lau and Breteler, 2006). PD prevalence increases from less than 0.5% at age 60 to around 4% by age 80 (de Lau and Breteler, 2006). There are several occupational and environmental risk factors which increase the risk of developing PD, including, exposure to pesticides and herbicides (de Lau and Breteler, 2006; Noyce et al., 2012; Bellou et al., 2016), heavy metal exposure and welding (de Lau and Breteler, 2006), residing in rural areas, drinking well water (de Lau and Breteler, 2006; Noyce et al., 2012; Bellou et al., 2016), ever having an influenza infection (Harris et al., 2012), or head trauma (de Lau and Breteler,

2006; Noyce et al., 2012). Factors which diminish the risk of developing PD include drinking coffee and caffeine (de Lau and Breteler, 2006; Noyce et al., 2012; Bellou et al.,

2016), smoking cigarettes (de Lau and Breteler, 2006; Noyce et al., 2012; Bellou et al.,

2016), using non-steroidal anti-inflammatory drugs (NSAIDS), especially ibuprofen (Gao et al., 2011; Noyce et al., 2012; Bellou et al., 2016), and red measles infection (Harris et al., 2012). 5

While most cases of PD are classified as sporadic and the underlying cause is not known, a minority of patients have underlying genetic risk factors can increase the risk of PD. For example, having a first degree relative with PD or tremor increases the risk of developing PD (Noyce et al., 2012). The first identified gene associated with PD is SNCA, the gene for α-Syn (Hardy et al., 2003). Duplication, triplication, or mutation of

SNCA are all associated with development of PD (Hardy et al., 2003; Singleton et al.,

2003; Ahn et al., 2008). Other genes involved in familial PD are PARK2 (the gene for parkin, an ubiquitin ligase), UCHL1 (the gene for UCHL-1, an ubiquitin hydrolase),

PINK1 (the gene for Pink1, a PTEN-induced kinase which protects mitochondria),

PARK7 (the gene for DJ-1, an oxidative stress sensor), PARK8 (the gene for LRRK2), as well as others (Hardy et al., 2003; de Lau and Breteler, 2006). Mutations in these genes are causative for PD. Genome-wide associated studies identified several other genes that are associated with PD patients such as CPLX1 (complexin 1 protein involved in synaptic vesicle exocytosis), KAT8 (the gene from lysine acetyltransferase

8), CD38 (a cyclic ADP ribose hydrolase), PYGO2 (a signal transduction protein), HLA-

DR (MHC class II), TOX3 (a gene associated with neuron survival), SATB1 (a gene associated with T cell and Treg development), as well as others (Ahmed et al., 2012;

Wissemann et al., 2013; Dumitriu et al., 2016; Chang et al., 2017). Despite being identified as being associated with PD, it is unclear if these genes are playing a pathogenic role in PD or if these genes are changed because of PD.

Models

Rodents do not develop PD naturally. To mimic PD in rodents, several different intoxicants that mimic the neurodegeneration, neuroinflammation, Lewy body formation, and motor dysfunction to differing degrees. The first of these models is 1-methyl-4- phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced PD, which was first identified in 6

humans who self-administered improperly synthesized desmethylprodine (Gupta, 2011).

These patients presented with rapid on-set PD symptoms. MPTP is a pro-neurotoxin which rapidly diffuses across the blood-brain barrier (Heikkila et al., 1984; Markey et al.,

1984). In the brain, MPTP is metabolized by monoamine oxidase B (MAO-B) to 1- methyl-4-phenylpyridinium (MPP+), the active neurotoxin (Chiba et al., 1984; Heikkila et al., 1984; Markey et al., 1984; Scotcher et al., 1990). MPP+ is structurally similar to dopamine (Klein et al., 1985), as a result, MPP+ is taken up through dopamine transporters into dopaminergic neurons. Within neurons, MPP+ may be sequestered into vesicles (Staal and Sonsalla, 2000) which protects the neurons. MPP+ may enter the mitochondria and block electron transport chain complex I (Singer et al., 1988).

Blocking the electron transport chain results in energy depletion, neurodegeneration, and neuroinflammation (Mizuno et al., 1987; Scotcher et al., 1990; Przedborski et al.,

1996; Niranjan et al., 2010). While MPP+ kills neurons itself, genetic knockout of proinflammatory cytokines such as interleukin (IL)-18 (Sugama et al., 2004) and tumor necrosis factor (TNF)-α (Ferger et al., 2004) leads to decreased dopaminergic neuron loss after MPTP intoxication compared to wild type. This suggests a significant percentage of neuron loss is due to subsequent neuroinflammation and not MPP+ itself.

CD4+ helper T cells are important for neuroinflammation because the absence of this cell type protects most dopaminergic neurons from degeneration (Benner et al., 2008;

Brochard et al., 2009). After neuroinflammation resolution, no additional dopaminergic neurons are lost (Jackson-Lewis et al., 1995). While the MPTP model recapitulates the specific neurodegeneration and neuroinflammation found in PD, few if any Lewy bodies are detected, unless MPTP is administered through an osmotic pump (Fornai et al.,

2005; Shimoji et al., 2005). Also, due to the rapid onset of neurodegeneration, there may (Sundström et al., 1990; Tillerson and Miller, 2003; Fornai et al., 2005) or may not

(Hirst and Ferger, 2008; Hutter-Saunders et al., 2012) be motor impairment. The lesion 7

in the MPTP model is more variable in female mice (Przedborski et al., 2001), meaning that commonly only male mice are used. After MPTP intoxication, female mice express higher GFAP (suggesting greater astrocyte activation), decreased expression of the dopamine transporter in the striatum, and administration of 17β-estradiol in both male and female mice is neuroprotective, all of which may contribute to the increased variability of MPTP lesion in female mice (Ookubo et al., 2008; Ookubo et al., 2009).

This is potentially a problem since female mice are not included in experiments testing efficacy of potential therapies.

Another toxin-based PD model is 6-hydroxydopamine (6-OHDA), an oxidized form of dopamine. Unlike MPTP, which is peripherally administered, 6-OHDA cannot cross the blood brain barrier (Garver et al., 1975). Therefore, 6-OHDA must be injected into the substantia nigra, median forebrain bundle, or striatum (Ungerstedt, 1968; Jeon et al., 1995; Rodrigues et al., 2003; Ma et al., 2014) to induce nigrostriatal degeneration.

6-OHDA is usually injected unilaterally to minimize toxicity, leading to unilateral neuroinflammation and neurodegeneration (Cicchetti et al., 2002; Rodrigues et al., 2003;

Walsh et al., 2011), unlike MPTP where neurodegeneration and neuroinflammation are bilateral. After injection, 6-OHDA is taken up into catecholaminergic neurons through catecholamine transporters (Kostrzewa and Jacobowitz, 1974; Luthman et al., 1989).

Once in neurons, 6-OHDA has two fates, oxidization which forms hydrogen peroxide, which subsequently forms ROS or impairment of complex I and IV of the electron transport chain (Heikkila and Cohen, 1971, 1972; Glinka and Youdim, 1995; Glinka et al., 1996). Both the generation of ROS and inhibition of the electron chain leads to neurodegeneration of catecholaminergic neurons and neuroinflammation. Unlike MPTP, which causes neurodegeneration and neuroinflammation specific for dopaminergic neurons, 6-OHDA affects both dopaminergic and noradrenergic neurons. As with 8

MPTP, 6-OHDA does not induce Lewy body formation (Duty and Jenner, 2011; Bove and Perier, 2012). While MPTP intoxication may or may not lead to motor symptoms, 6-

OHDA intoxication generates noticeable motor deficits. After unilateral administration of

6-OHDA lesion, administration of amphetamine is injected causing ipsilateral rotation in a dose-dependent fashion (Carey, 1992; Robinson et al., 1994).

Because pesticide and herbicide exposure has been linked to increased risk of

PD, pesticides and herbicides were used to model PD symptoms and neurodegeneration in rodent models. Both paraquat (a herbicide) and rotenone (a herbicide and insecticide) can be administered peripherally and taken up into the brain.

Rotenone is lipophilic and rapidly crosses the blood brain barrier and paraquat is actively transported through neutral amino acid transporter (Shimizu et al., 2001). Administration of either toxin causes nigrostriatal pathway degeneration (Betarbet et al., 2000;

McCormack et al., 2002). Rotenone is neurotoxic by inhibiting electro transport chain complex I (Betarbet et al., 2000). Paraquat is neurotoxic by redox cycling and generating reactive oxygen and nitrogen species (Bus et al., 1976; Day et al., 1999).

Unlike the 6-OHDA and MPTP models, injection of rotenone or paraquat increases expression of α-Syn and induces formation of inclusions that are similar to Lewy bodies found in PD patients (Brooks et al., 1999; Betarbet et al., 2000; Manning-Bog et al.,

2002). In addition, administration of both rotenone and paraquat lead to motor symptoms of decreased rearing and spontaneous movement also mimicking PD (Alam and Schmidt, 2004; Cannon et al., 2009). Rotenone and paraquat failed to achieve more acceptance because of variable degrees of degeneration following administration

(Duty and Jenner, 2011; Bove and Perier, 2012).

Familial PD can also be modeled in rodents. As indicated above, several genes have been implicated in the development of familial PD. The gene most associated with 9

PD is SNCA. Mice that overexpress mouse or human α-Syn, or express A30P or A53T mutants of α-Syn develop Lewy body-like inclusions in the nigrostriatal pathway (Kahle et al., 2000; van der Putten et al., 2000; Giasson et al., 2002; Lee et al., 2002; Rieker et al., 2011). However, the degree of neurodegeneration in the nigrostriatal pathway is inconsistent (Masliah et al., 2000; Matsuoka et al., 2001). This lack of dopaminergic neuron degeneration, even in older rodents, is a major reason these models have not gained more prominence. The knockout of other genes associated with PD such as parkin (PARK2), PINK1 (PINK1), and DJ-1 (PARK7) are used as models of PD, however, these models also lack neurodegeneration (Goldberg et al., 2003; Goldberg et al., 2005; Kitada et al., 2007).

Lastly, it should be noted that genetic and toxin models of PD exist in non-rodent organisms such as non-human primates, C. elegans and Drosophila. Non-human primates do not develop PD spontaneously, but can develop PD symptoms after toxin administration (Emborg, 2007). However, due to the cost and specialized facilities needed, the non-human primate models of PD have not been widely used. PD models in C. elegans and Drosophila are beneficial because these organisms contain limited numbers of neurons and other cells in the CNS allowing for easy nerve tracing, quick generation time, and testing on large numbers needed for high throughput screening

(Duty and Jenner, 2011). However, these model organisms are more genetically divergent from PD patients than rodent models, meaning that results generated from these models may not translate into the human population.

Treatments

Since PD symptoms are caused by the loss of dopamine in the striatum, the most clinically useful therapies for PD are replacing lost dopamine. The most common therapy for PD is levodopa given alone or with carbidopa, an inhibitor of aromatic-L- 10

amino acid decarboxylase, the enzyme which decarboxylates levodopa to dopamine

(Calne, 1993; Samii et al., 2004; Dhall and Kreitzman, 2016; Oertel and Schulz, 2016).

While this therapy is beneficial in the short-term, long-term treatment with levodopa often leads to dyskinesia (Calne, 1993; Rascol et al., 2000; Samii et al., 2004; Stocchi et al.,

2010). Other approved PD therapies include inhibition of cathechol-O- methyltransferase, the enzyme that metabolizes dopamine (Muller, 2015; Oertel and

Schulz, 2016), and dopamine agonists (Calne, 1993; Oertel and Schulz, 2016).

Combination therapy of rasagiline, an inhibitor of the dopamine-metabolizing enzyme

MAO-B, and pramipexole, a dopamine receptor 2 agonist, is successful at reducing PD symptoms in a recent clinical trial (Olanow et al., 2017). In addition, selegiline monotherapy, another MAO-B inhibitor, decreased the combined Unified Parkinson’s

Disease Rating Scale (UPDRS) score, showing improvement compared to placebo

(Mizuno et al., 2017).

While these therapies are effective at replacing lost dopamine, these therapies do not address the underlying the cause of disease, the loss of dopaminergic neurons.

One therapeutic strategy is the generation of the antibodies against α-Syn. In practice, passive transfer of anti-α-Syn antibodies increases clearance of α-Syn fibrils (Games et al., 2014; Lindstrom et al., 2014). Passive transfer of antibodies and increased autoantibodies against α-Syn, reduce α-Syn fibrils, and inclusions protect neurons from further degeneration (Masliah et al., 2005; Games et al., 2014; Lindstrom et al., 2014).

While clinical trials using monoclonal antibodies to α-Syn (PRX002) were performed in humans, and these antibodies are safe to administer (Schenk et al., 2017), to date, these antibodies have yet to successfully reduce PD symptoms. Other novel therapeutic approaches that slow or stop dopaminergic neurons loss by decreasing neuroinflammation could be used. Such therapies such as minocycline and pioglitazone 11

were protective in rodent models of PD (Peng et al., 2006; Laloux et al., 2012), but these therapies were not been effective in PD patients (Investigators., 2006; Investigators.,

2015; Simon et al., 2015). Despite advances in therapies for PD, a novel therapy which diminishes neuroinflammation is needed because the progression of neurodegeneration could be slowed.

NEUROINFLAMATION AND PARKINSON’S DISEASE

Inflammatory immune cells in PD

One byproduct of nigrostriatal degeneration and Lewy body formation is inflammation. During PD, several types of immune cells are activated. The first of these immune cell types is microglia, which reside in the brain. Microglia arise from two sources (Chan et al., 2007; Kettenmann et al., 2011). In the postnatal brain, a limited number of blood monocytes migrate into the brain and differentiate into microglia. The second and major source of microglia are myeloid progenitors that arise from the bone marrow then migrate into the brain during embryonic development. These two populations can be differentiated by the expression of CX3CR1 (found on resident microglia) and CCR2 (found on microglia infiltrating in postnatal mice) (Mizutani et al.,

2012). Microglia have several described functions including pruning synapses (Paolicelli et al., 2011; Tremblay et al., 2011), repairing and promoting neuronal survival after damage (Cherry et al., 2014; Michell-Robinson et al., 2015), as well as clearing pathogens in the brain (Michell-Robinson et al., 2015; Orihuela et al., 2016). Microglia primed for repair and neuronal survival are designated M2 microglia. These microglia release anti-inflammatory cytokines IL-4, IL-13, IL-10 and transforming growth factor beta (TGF-β), and release neurotrophic factors such as prostanoids, and arginase

(Minghetti and Levi, 1998; Orihuela et al., 2016). Conversely, M1 microglia are primed 12

to release proinflammatory cytokines such as interferon gamma (IFNγ), TNF-α, IL-12, IL-

1β, and IL-6 as well as the release of reactive oxygen and reactive nitrogen species

(Michell-Robinson et al., 2015; Orihuela et al., 2016). The combination of these M1 and

M2 microglia are responsible for resolution of inflammation and repair of damaged neurons in the central nervous system.

During PD and other neurodegenerative disease, the debris from dead and dying neurons and the accumulation and release of amyloid aggregates, such as Lewy bodies, activate microglia (Meda et al., 1995; Beyer et al., 2000; Zhang et al., 2005; Kempuraj et al., 2016). Aggregated α-Syn binds to pattern recognition molecules such as CD36 and

Toll-like receptors 1, 2 and 4, to stimulate the release of proinflammatory mediators from microglia (Su et al., 2008; Stefanova et al., 2011; Fellner et al., 2013; Daniele et al.,

2015). In PD, activated microglia accumulate in the striatum, globus pallidus, thalamus, and pons compared to age-matched controls as determined in vivo by positron emission tomography (PET) scanning for activated microglia marker PK11195 (Gerhard et al.,

2006). In the postmortem brain, HLA-DR positive cells increase in the substantia nigra of PD patients compared to non-PD controls (McGeer et al., 1988). Microglia in the post mortem PD patients show increased expression of the adhesion molecules ICAM-1 and

LFA-1 which are important for the migration of microglia and lymphocytes to regions of neuronal loss (Miklossy et al., 2006). PD patients also exhibit increased MHC II-positive cells in several different brain regions including the substantia nigra and putamen

(Imamura et al., 2003). Taken together, M1-activated microglia in PD patients and are in proximity of the degenerating neurons, suggesting that they play a role in degeneration.

Outside the central nervous system, dendritic cells (DCs), macrophages, and monocytes are the most important non-granulocyte myeloid cells. These cells are derived from common myeloid progenitors in the bone marrow (Geissmann et al., 2010). 13

These cells are all part of the innate immune system which play an important role in the activation of the adaptive immune response (Medzhitov and Janeway, 1997; Liu, 2001).

DCs are found in lymphoid tissue such as Peyer’s patches and non-lymphoid tissue such as the skin, lungs, and nasopharynx. DCs arise from two origins, myeloid origin dendritic cells which originate from myeloid progenitor cells, or plasmacytoid dendritic cells which originate from the common dendritic cell progenitor or the lymphoid progenitor (Rutella et al., 2006). After activation, DCs migrate from the peripheral tissue to the T cell zone of the draining lymph nodes to activate T cells (Martín-Fontecha et al.,

2009; Merad et al., 2013). The primary role of DCs is uptake, processing, and presentation of antigens that activate T cells (Merad et al., 2013). DCs can release cytokines and other proinflammatory mediators that influence the immune environment biasing the activation of T cells, leading to activation of the adoptive immune response.

Macrophages are found within the tissues and monocytes are found in the blood

(Gordon and Taylor, 2005). Both cell types play a role in antigen processing and adaptive immune cell activation, but to a lesser extent than DCs (Steinman, 1991;

Banchereau and Steinman, 1998). Primarily, macrophages and monocytes are involved in clearance of pathogens and debris (Gordon and Taylor, 2005). As with DCs, macrophages and monocytes release cytokines and proinflammatory mediators such as

ROS that facilitate clearance of pathogens and wound healing.

In PD, the percentages of B and T cells are increased (see below), suggesting an important role for antigen-presenting cells (APCs) in the progression of disease. Few studies tested the percentage of DCs in PD. One study published a decrease in the percentage of both plasmacytoid and myeloid DCs (Ciaramella et al., 2013). However, another study showed these populations are not changed (Goldeck et al., 2016). In addition, PD patients show increased classical monocytes (CD14+ CD16-), the major 14

circulating monocyte population, and decreased non-classical monocytes (CD14+

CD16+), the major phagocytic macrophages (Grozdanov et al., 2014). Monocytes from

PD patients also exhibit a decreased phagocytic index compared to controls. Classical monocytes in the PD patients display increased expression of CCR2, a chemokine receptor needed for monocyte recruitment to sites of inflammation (Funk et al., 2013).

There is conflicting data concerning the release of IL-1β, TNF-α, and IL-6 from LPS- stimulated monocytes derived from PD patients. One publication indicated these cytokines were increased (Grozdanov et al., 2014), while another publication showed these cytokines were decreased (Hasegawa et al., 2000).

Lymphocytes are immune cells which are derived from a common lymphoid progenitor in the bone marrow (Galy et al., 1995; Kondo et al., 1997). There are three lymphocyte subpopulations: natural killer (NK) cells, B cells and T cells. NK cells can be differentiated by the expression of CD16, CD56, CD27, and CD11b on their surface and release perforin and granzyme-containing granules which kill cancer cells and virus- infected cells (Goh and Huntington, 2017). B cells display immunoglobulins on the surface (Pike and Ratcliffe, 2002). When activated, B cells undergo somatic hypermutation of the immunoglobulin V region, class-switching of the immunoglobulin constant region, and (for some B cells) differentiation into immunoglobulin-secreting plasma cells (Hwang et al., 2015). There are 2 major T cell lineages: CD4+ helper T cells and CD8+ cytotoxic T cells. CD4+ helper T cells are further subdivided into helper

T subtypes (Th1, Th2, Th17), regulatory T cells (Tregs) and several other subsets depending on the cytokines released (Luckheeram et al., 2012; Tripathi and Lahesmaa,

2014). These helper T cells bias the immune response to clear antigens that are activating the immune system (Luckheeram et al., 2012). Helper T cells can also be divided into naïve helper cells (CD4+ CD45RA+), which have yet to be activated in 15

response to antigens, and memory helper T cells (CD4+ CD45RO+) which have previously been activated and are maintained to more rapidly respond to that antigen in the future (Berard and Tough, 2002). CD8+ T cells are cytotoxic, able to bind to and kill tumor or infected cells by release of perforin and granzyme-containing granules or through Fas ligand-Fas interactions (Harty et al., 2000). Unlike B cells which can bind antigens without presentation, both CD8+ and CD4+ T cells need antigens to be presented in the cleft of major histocompatibility complex (MHC) I or MHC II, respectively

(Teh et al., 1988; Bour et al., 1995). T cells can also be divided into αβ or γδ T cells based on which T cell receptor gene segments are utilized by the T cell receptor. In the periphery, most T cells are αβ T cells, but approximately 5% of the T cell population are

γδ T cells (Carding and Egan, 2002). In conclusion, the three lymphocyte subtypes control infection, with B and T cells being the primary components of the cellular adoptive immune response.

In PD patients, the percentage and function of lymphocytes is altered.

Circulating NK cells are increased (Niwa et al., 2012) and circulating B cells are decreased in PD patients compared to healthy controls (Bas et al., 2001; Niwa et al.,

2012; Stevens et al., 2012; Horvath and Ritz, 2015). Interestingly, PD patients have increased autoantibodies (Benkler et al., 2012), possibly including antibodies against α-

Syn (Papachroni et al., 2007). This suggests that even with decreased B cell numbers,

B cell-derived antibody production may be increased. To date, no study has discerned significant changes in the percentages of plasma cells in PD patients. Within the T cell population, PD patients exhibit increased circulating CD8+ (Bas et al., 2001; Baba et al.,

2005) and γδ T cells (Fiszer et al., 1994), but a decreased the percentage of circulating

CD4+ T cells (Bas et al., 2001; Baba et al., 2005; Hutter-Saunders et al., 2012; Niwa et al., 2012; Stevens et al., 2012; Horvath and Ritz, 2015). In PD, the percentage of 16

memory helper T cells (CD4+ CD45RO+) is increased and the percentage of naïve helper T cells (CD4+ CD45RA+) is decreased (Bas et al., 2001; Hutter-Saunders et al.,

2012; Stevens et al., 2012; Horvath and Ritz, 2015). The percentage of proinflammatory helper T cell subsets Th1 and Th17 is increased and anti-inflammatory Th2 and Treg

CD4+ cells are decreased (Chen et al., 2015). The ratio of IFNγ- to IL-4-producing

CD4+ cells is significantly increased in PD patients, also suggesting a shift toward Th1 and away from Th2 (Baba et al., 2005). The ability of Tregs from PD patients to suppress proliferation of CD4+ CD25- cells in culture is diminished compared to Tregs from age-matched controls (Hutter-Saunders et al., 2012). Furthermore, PBMCs from

PD patients show increased basal expression of IFNγ, IL-1β, TNF-α, IL-8, MCP-1 and

MIP-1α, further highlighting that PD patients display increased expression of proinflammatory cytokines and chemokines (Reale et al., 2009). Combined, these results indicate that lymphocyte percentages and functions are dysregulated leading to increased proinflammatory environments with diminished regulatory immune functions.

A final population of immune cells are granulocytes, which are comprised of neutrophils, eosinophils, and basophiles, and differentiated from the common granulocyte progenitor (Akashi et al., 2000). Unlike monocytes, macrophages, DCs, and lymphocytes which are all mononuclear cells, granulocytes possess multiple-lobed nuclei and are called polymorphonuclear cells. As the name in implies, granulocytes also contain large numbers of granules, which contain proteases, cytotoxins, and antimicrobial peptides, and are released to eliminate extracellular pathogens as part of their function in the innate immune system (Geering et al., 2013). Neutrophils are the most abundant immune cell population and are the first cells to respond after infection

(Mestas and Hughes, 2004; Jones et al., 2016). When neutrophils respond to infection, they phagocytize extracellular pathogens that are eliminated by the production of ROS 17

and antimicrobial peptides and enzymes (Mayer-Scholl et al., 2004; Kolaczkowska and

Kubes, 2013). For tumors, helminths, and pathogens too large to phagocytize, neutrophils release proinflammatory mediators such as IL-12, cationic defensins, extracellular enzymes such as elastase, collagenase, and matrix metalloproteinase

(Olsson and Venge, 1974; Olofsson et al., 1976; Cadman and Lawrence, 2010;

Kolaczkowska and Kubes, 2013). Basophils and eosinophils represent less than 5% of the white blood cells in circulation and play roles in helminth infection, asthma, and allergies (Cadman and Lawrence, 2010; Stone et al., 2010; Siracusa et al., 2013).

Basophils and eosinophils function by IgE bound to Fc receptors, which induce release of Th2 cytokines such as IL-4, IL-5, IL-9 and IL-13, leukotrienes, and histamines (Stone et al., 2010; Siracusa et al., 2013).

Even though granulocytes, especially neutrophils, play an important role in inflammation, little research has been performed to evaluate changes in number and function during PD. One study indicated no increase in peripheral neutrophil number

(Atac Ucar et al., 2016). However, the total number of granulocytes are increased in the blood (Horvath and Ritz, 2015), but this study did not differentiate subsets of granulocytes. Release of nitrogen oxygen species from PD patient neutrophils was increased compared to controls, suggesting increased proinflammatory function (Gatto et al., 1996). Combined, these data suggest that the increase in granulocyte number and the increased release of reactive nitrogen species contribute to the inflammation in

PD.

Released mediators in PD

In addition to immune cells, the immune system requires the release of mediators to carry out its function. These mediators may be cytokines, chemokines, reactive oxygen and nitrogen species, prostaglandins, histamines, leukotrienes, anti-microbial 18

peptides and extracellular enzymes which bias the immune system, regulate inflammation, control regulatory functions of the immune system, act as chemoattractants for immune cells, have intrinsic capacity to remove pathogens, and regulate the functions of non-immune cells (Bogdan et al., 2000; Vaday and Lider, 2000;

Borish and Steinke, 2003; Peters-Golden et al., 2005; Zhang and An, 2007; Azad et al.,

2008; O'Mahony et al., 2011; Ricciotti and FitzGerald, 2011; Mangoni et al., 2016). In

PD, the concentration of proinflammatory mediators are increased compared to healthy controls. This includes increased concentration of proinflammatory cytokines such as IL-

6, IL-1β and nitrate (a reactive nitrogen species) in the cerebrospinal fluid (CSF) (Mogi et al., 1994; Blum-Degen et al., 1995; Qureshi et al., 1995; Scalzo et al., 2010). Counter to the expectation, an increase of IL-6 concentration in CSF correlated with diminished motor symptoms of PD (Müller et al., 1998). The concentration of serum proinflammatory mediators appears to be more variable in some studies showing increased IL6, IL-1β, or RANTES (Rentzos et al., 2007; Lindqvist et al., 2012; Dursun et al., 2015) or decreased IL-6, IL-1α, IL-8, TNF-α and nitrite concentrations (Dursun et al.,

2015; Cubukcu et al., 2016; Gupta et al., 2016). PD patients also exhibit increased serum IL-10, an anti-inflammatory cytokine (Rentzos et al., 2009), perhaps indicating a general increase in cytokine production or a compensatory response to the proinflammatory environment. In the postmortem brain, PD patients also exhibit increased reactivity for proteins modified with 4-hydroxy-2-nonenal, an oxidized lipid

(HNE) (Yoritaka et al., 1996), protein containing carbonyls (Alam et al., 1997a), and 8- hydroxyguanine (an oxidized nucleotide DNA base) (Alam et al., 1997b), all are markers of an oxidatively stressed environment. In the CSF, PD patients show increased 8- hydroxyguanine compared to controls, suggesting increased oxidative damage (Abe et al., 2003). PD lymphocytes also show increased micronuclei and more single strand breaks in the chromosomal DNA, suggesting that the oxidative damage detected in the 19

brain is also in peripheral immune cells (Migliore et al., 2001; Petrozzi et al., 2001;

Migliore et al., 2002).

In summary, there is a sustained proinflammatory environment in affected brain tissue. This is demonstrated by increased activation of microglia (McGeer et al., 1988;

Imamura et al., 2003), dysregulation of peripheral immune cells (Bas et al., 2001; Niwa et al., 2012), and increased concentrations of proinflammatory mediators in serum and

CSF (Mogi et al., 1994; Blum-Degen et al., 1995; Qureshi et al., 1995; Rentzos et al.,

2007; Scalzo et al., 2010; Lindqvist et al., 2012; Dursun et al., 2015). The presence of pro-inflammatory mediators in regions of neurodegeneration suggests a role for neuroinflammation in PD pathogenesis. Mitigating neuroinflammation could diminish disease severity or slow degeneration in PD and may create an environment where repair can be initiated.

GRANUCLOCYTE-MACROPHAGE COLONY STIMULATING FACTOR (GM-

CSF)

Structure, function and signaling of GM-CSF to its receptor

One of the therapeutics that may be used to mitigate neuroinflammation and immune system dysfunction in PD is the cytokine GM-CSF (colony stimulating factor-2,

CSF2). In the body, GM-CSF is a monomer with four helical domains (Shearer, 2003).

GM-CSF binds to its receptor (CSF2R), which is composed of two subunits. The α subunit that contains the site of GM-CSF binding and the β subunit is a common chain shared by GM-CSF, IL-3 and IL-5 receptors, and is important for intracellular signaling

(Hercus et al., 2009). The crystal structure of the GM-CSF receptor bound to GM-CSF portrays a hexamer complex of 2 α subunits, 2 β subunits of the GM-CSF receptor 20

binding 2 GM-CSF monomers (Hansen et al., 2008). After GM-CSF binding, the kinase

JAK2 phosphorylates tyrosine amino acids in the cytoplasmic tail of both subunits of the

GM-CSF receptor (Shearer, 2003; Hansen et al., 2008; Hercus et al., 2009).

Phosphorylated tyrosines recruit the signal transducing STAT family members such as

STAT1 and STAT5, MAPK pathway members such as ERK1/2, and PI3-K pathway members for downstream signaling (Shearer, 2003; Hercus et al., 2009). As expected from activating of several pathways in several different cell types, GM-CSF displays several functions, the most important of which is that GM-CSF acts on CD34+ immune stem cells to suppress lymphocyte progenitor differentiation and promote differentiation into macrophages, IL-12-producing DCs, and granulocytes (Shearer, 2003).

Therapeutic application of GM-CSF in rodent models and humans

GM-CSF has been used in two opposing ways to therapeutically modulate the immune system. First, GM-CSF can be used as an adjuvant to improve the immune response to an antigen. Recombinant GM-CSF, when combined with the hepatitis B vaccine, increased anti-hepatitis B antibody titer suggesting a more protective response

(Cruciani et al., 2007). GM-CSF has also been and is continuing to be used as an adjuvant for vaccines in melanoma, prostate cancer, pancreatic cancer, colon cancer, breast cancer, and renal cancer (Jones et al., 1996; Ryan et al., 2000; Parmiani et al.,

2007; Spitler et al., 2009; Garcia et al., 2014) (NCT00064129, NCT00669734,

NCT01134614, NCT02466906, NCT01479244, NCT00458536). GM-CSF is used to reduce myelosuppression in these cancers. However, GM-CSF does not reliably increase T cell or antibody responses to antigens of these tumors, potentially due to differences in the dose of GM-CSF used.

In contrast to its role as an adjuvant, GM-CSF can also act to suppress immune responses in models of autoimmunity. In the experimental myasthenia gravis mouse 21

model, there is progressive muscle weakness due to an immune response to muscle acetylcholine receptor (Sheng et al., 2011). Administration of GM-CSF decreases severity of muscle weakness and production of antibodies reactive to acetylcholine receptor, decreases CD4+ T cell expression of proinflammatory cytokines, and increases

Treg percentage. The experimental model of thyroiditis is induced by immunization with thyroglobulin, leading to destruction of thyroid follicles (Vasu et al., 2003).

Administration of GM-CSF diminishes thyroid damage, decreases expression of disease-exacerbating IL-12, and increases the percentage of Tregs (Vasu et al., 2003;

Ganesh et al., 2009). Additional research determined that GM-CSF acts on DCs that can induce Tregs and the suppression of autoimmunity (Ganesh et al., 2009;

Bhattacharya et al., 2011). NOD mice are a model of type 1 diabetes that develop insulitis and is mitigated by GM-CSF treatment (Cheatem et al., 2009). GM-CSF administration increases the percentages of Tregs, Treg expression of IL-10 and TGF-β, and percentages of dendritic cells. In addition to the aforementioned autoimmunity models, GM-CSF can also diminishes the colitis in the dextran sulfate sodium model

(Sainathan et al., 2008). In this model, GM-CSF treatment diminished expression of proinflammatory cytokines IL-1α, IL-1-β, and TNFα, and increased indolamine 2,3 dioxygenase (IDO) expression in plasmacytoid DCs in the gut. In a model of traumatic brain injury, GM-CSF administration diminished the number Iba-1+ microglia, decreased lesion size, and spared more damage to the cortex (Kelso et al., 2015). Lastly, GM-CSF was protective in the MPTP PD model. Administration of GM-CSF protects dopaminergic neurons and diminishes the number of activated microglia following MPTP

(Kosloski et al., 2013). This neuroprotective effect is mediated by an increase in Tregs, as the transfer of the induced Tregs was sufficient for neuroprotection and diminished neuroinflammation. In conclusion, GM-CSF can act to diminish the inflammatory response in autoimmunity by acting on DCs to induce Tregs. 22

One of the benefits of using GM-CSF as a therapy is that it is an FDA approved drug called sargramostim (Leukine), which is recombinant human GM-CSF produced in yeast with a leucine to proline substitution at position 23 (Dorr, 1993; Waller, 2007).

Sargramostim decreased neutrophil recovery time and increased survival times in acute myelogenous leukemia (Rowe et al., 1995) and lymphoid neoplasia (Nemunaitis et al.,

1991). Administration of sargramostim also diminishes the recovery time of immune cells following bone marrow replacement (Nemunaitis et al., 1991; Hussein et al., 1995).

Sargramostim and filgrastim (recombinant GM-CSF produced in E. coli) display similar therapeutic effects (Beveridge et al., 2009), but sargramostim-treated patients reported fewer adverse events (Dorr, 1993).

Because of the benefit of GM-CSF in models of autoimmunity, ongoing research is determining whether sargramostim can act as an adjuvant or as an anti-inflammatory mediator in human patients. As indicated above, ongoing research is testing the efficacy of sargramostim as an adjuvant for the hepatitis B vaccine and to increase immune responses to various tumors. In addition, sargramostim is also being investigated for its ability to diminish clinical symptoms of Crohn’s disease, a chronic inflammatory disease of the gastrointestinal tract. In several trials (Korzenik et al., 2005; Takazoe et al., 2009;

Valentine et al., 2009), but not all (Roth et al., 2012), sargramostim diminishes clinical severity compared to placebo control. Additional clinical trials have been completed to better investigate the clinical efficacy of sargramostim, but results have not been published (NCT00206674, NCT00206596). To date, none of these trials addressed if the changes in clinical score were related to changes in immune cell, especially Treg, number or function. The ability of sargramostim to mitigate the cognitive decline in

Alzheimer’s disease is also being tested in ongoing clinical trials (NCT01409915 and

NCT02667496). In PD, sargramostim decreased the Unified Parkinson’s Disease 23

Rating Scale (UPDRS) part III clinical score in patients, mitigated the loss of beta ERD in the precentral gyrus, and increased Treg percentage and function (Gendelman et al.,

2017). These data suggest that sargramostim mitigates PD symptoms which may be related to Treg number and/or function. In total, GM-CSF and sargramostim have immune modulatory function related to increases in immune cell number. Depending on which cells are increased, GM-CSF may display an immune-stimulating or immune- diminishing function.

MECHANISMS OF TREG INDUCTION

Tregs are derived from two sources. The first is thymic-derived or natural Tregs, the second is induced or peripheral Tregs. Natural Tregs are derived from the thymus and are part of the mechanism of central tolerance (Bluestone and Abbas, 2003;

Povoleri et al., 2013). These CD4+ T cells recognize self-antigens presented in the thymus with intermediate strength as part of CD4+ T cell differentiation. Natural Tregs express high CD25, Foxp3, GITR, CTLA-4, and Helios. Induced Tregs are derived from mature, naïve CD4+ T cells that interact with APCs in the periphery. The absence of sufficient antigen stimulation, the presence of co-stimulatory molecules and cytokines such as IL-10 and TGF-β bias the activation of CD4+ T cells to induced Tregs. Induced

Tregs express variable amounts of CD25, Foxp3, GITR, and CTLA-4, but not Helios.

Despite different origins, natural Tregs and induced Tregs possess similar suppressive functions such as release of immunosuppressive cytokines including IL-10 and TGF-β, cell to cell suppression with co-stimulatory molecules like CTLA-4, competition for growth factors such as IL-2 and tryptophan, and modulation of the functions of APCs.

Dendritic cells are the primary immune cells that induce Tregs in the periphery.

As described above, two populations of dendritic cells, plasmacytoid and myeloid, arise 24

from different progenitors (Rutella et al., 2006). Both dendritic cell populations display two maturation states, immature cells which actively take up antigens, but exhibit low antigen presentation capability, and mature dendritic cells which exhibit less antigen uptake and increased antigen presentation capabilities (Maldonado and von Andrian,

2010; Raker et al., 2015). Dendritic cells are functionally characterized as immunogenic or tolerogenic dendritic cells, which induce naive helper T cells to differentiate effector or regulatory helper T cells, respectively (Maldonado and von Andrian, 2010). Tolerogenic dendritic cells induce Tregs by multiple mechanisms including release of anti- inflammatory cytokines IL-10 and TGF-β, increased expression of IDO and its by-product kynurenine, and altered surface co-stimulatory molecule expression (Steinman et al.,

2003; Maldonado and von Andrian, 2010; Li and Shi, 2015; Raker et al., 2015). The percentage of tolerogenic dendritic cells can be increased naturally by increased expression of anti-inflammatory cytokines, increased concentration of glucocorticoids, other anti-inflammatory modulators such as VIP and vitamin D3, and pharmacological inducers such as GM-CSF (Maldonado and von Andrian, 2010). As indicated above,

GM-CSF induces tolerogenic dendritic cells which can be productive in models of autoimmunity. GM-CSF-induced tolerogenic dendritic cells induce Treg formation through the co-stimulatory molecules OX40L and Jagged-1 (Gopisetty et al., 2013;

Haddad et al., 2016; Kumar et al., 2017). Because of the neuroprotective effects of GM-

CSF in PD patients and in the MPTP model, research is warranted to determine whether the induction of tolerogenic dendritic cells is a mechanism of neuroprotection.

SUMMARY AND CONCLUSIONS

PD is a neurodegenerative disease characterized by the loss of dopaminergic neurons in the pars compacta substantia nigra and accumulation of α-Syn-containing 25

Lewy bodies. During PD disease, microglia become activated and the peripheral immune system becomes dysregulated. Included in this peripheral dysfunction is a decrease in Treg percentage and function and increased expression of proinflammatory mediators, suggesting a bias toward inflammation. Correcting this imbalance would be a novel therapeutic for PD. GM-CSF is a cytokine which, among other functions, increases the number and function of Tregs. Administration of GM-CSF increases the percentage of Tregs, diminishes neuroinflammation and protectes dopaminergic neurons in the MPTP model. Sargramostim diminishes UPDRS part III scores and increases

Treg percentage and function in PD patients. In models of autoimmunity, GM-CSF diminishes autoimmunity by inducing tolerogenic dendritic cells which induce Tregs and suppress inflammation. This project investigates whether tolerogenic dendritic cells are protective in the MPTP model of PD.

26

CHAPTER TWO

GENE EXPRESSION IN T RESPONDER CELLS IN PD PATIENTS

TREATED WITH SARGRAMOSTIM

ABSTRACT

In Parkinson’s disease (PD) the percentage of peripheral immune cells and mediators is altered. The increase in proinflammatory cytokines and helper T cell subsets and the decrease in anti-inflammatory cytokines and helper T cell subsets may contribute to the pathology and progression of PD. A novel therapeutic agent that could restore balance between proinflammatory and anti-inflammatory cells and cytokines may decrease the symptoms of disease, slow the progression of disease, and create an environment to recover lost neurons. Sargramostim is recombinant human GM-CSF which is an FDA approved drug. From a phase I clinical trial, we isolated CD4+ CD25- T cells from healthy controls, PD placebo controls, or sargramostim-treated PD patients and isolated RNA for PCR arrays to test expression of genes for helper T cell differentiation. We hypothesized that sargramostim-treated PD patients would show increased expression of anti-inflammatory genes and decreased expression of proinflammatory genes. We found that sargramostim increased the expression of both pro- and anti-inflammatory genes as well as genes involved in helper T cell differentiation and proliferation. These data demonstrate that sargramostim increases the expression of genes associated with all helper T cell subsets, and in general expands the immune response. This suggests that GM-CSF alters the immune response in PD patients, including increasing Tregs and anti-inflammatory immune cells and cytokines. 27

INTRODUCTION

Part of PD pathology of PD is inflammation in the substantia nigra and striatum.

This neuroinflammation is mediated by activated resident microglia as well as infiltrating immune cells, including lymphocytes (Qian et al., 2010; Gonzalez and Pacheco, 2014).

These activated cells release reactive oxygen and nitrogen species as well as proinflammatory cytokines and chemokines (Whitton, 2007). This oxidizing environment leads to the post-translational modification of proteins such as nitration of α-synuclein

(Duda et al., 2000; Giasson, 2000). The resulting inflammatory and oxidizing environment promotes the loss of more dopaminergic neurons which, in turn, leads to more neuroinflammation and increased neurodegeneration (Mosley et al., 2012).

Not only is there inflammation in the brains of PD patients, the peripheral immune system shows signs of dysregulation. Peripheral blood of PD patients contain decreased numbers of plasmacytoid and myeloid dendritic cells in the blood (Ciaramella et al., 2013), and increased numbers of total granulocytes (Horvath and Ritz, 2015), but neutrophil number is not increased (Atac Ucar et al., 2016). Total numbers of monocytes are not changed in PD patients’ blood; however, classical monocytes

(CD14+ CD16-) are increased in PD patients while non-classical monocytes (CD14-

CD16+) are decreased in PD patients (Grozdanov et al., 2014). In addition to changes in the myeloid cells, changes within the lymphocyte population have also been described. PD patients display increased γδ T cells (Fiszer et al., 1994) and CD8+ T cells (Baba et al., 2005), but decreased B cells (Niwa et al., 2012; Stevens et al., 2012;

Horvath and Ritz, 2015) and CD4+ T cells (Baba et al., 2005; Hutter-Saunders et al.,

2012; Niwa et al., 2012; Stevens et al., 2012; Horvath and Ritz, 2015). More specifically,

PD patients exhibit decreased naïve CD4+ T cells (CD4+ CD45RA+) and increased 28

memory CD4+ T cells (CD4+ CD45RO+) (Hutter-Saunders et al., 2012; Stevens et al.,

2012; Horvath and Ritz, 2015). In addition, the percentage of different helper T cell subsets is also changed within PD patients. Percentages of proinflammatory Th1 and

Th17 helper T cells are increased, but the percentage of the anti-inflammatory Th2 helper T cells and Tregs are decreased (Chen et al., 2015). Tregs also possess decreased ability to suppress the proliferation of CD4+ CD25- responder T cells in culture (Hutter-Saunders et al., 2012), suggesting that Tregs from PD patients are less functional. In addition to these alterations in immune cell number and function, serum proinflammatory cytokines including IL-6, MCP-1, MIP-1α, IL-1β, IFNγ, and TNF-α are increased in serum (Reale et al., 2009; Scalzo et al., 2010; Lindqvist et al., 2012).

Reestablishing immune homeostasis is a novel target for PD treatment. By increasing the percentages and concentration of anti-inflammatory cells and concentrations of cytokines, thus decreasing the percentage of proinflammatory cells and concentration of cytokines, neuroinflammation and neuroprotection can be mitigated. Several potential therapies target diminution of neuroinflammation and are neuroprotective in models of PD; some increase Treg percentage and function. One potential therapeutic is vasoactive intestinal peptide (VIP). This short peptide is neuroprotective in the MPTP model (Delgado and Ganea, 2003; Reynolds et al., 2010;

Olson et al., 2015) and induces Tregs (Gonzalez-Rey and Delgado, 2007). Adoptive transfer of splenocytes from VIP-treated mice is neuroprotective (Olson et al., 2015).

VIP decreases reactive microglia numbers in the substantia nigra following MPTP intoxication (Delgado and Ganea, 2003; Reynolds et al., 2010; Olson et al., 2015). This effect of VIP is mediated by binding to VIP receptor 2, since a specific agonist to this receptor, but not VIP receptor 1, has the same neuroprotective effect as VIP (Olson et al., 2015). Bee venom, and its active component phospholipase A, protects 29

dopaminergic neurons and decreases neuroinflammation in the MPTP of PD (Doo et al.,

2010; Chung et al., 2012; Chung et al., 2015; Kim et al., 2016). This neuroprotection was mitigated by Treg depletion, suggesting that bee venom acts through Tregs (Chung et al., 2012; Chung et al., 2015). However, in a clinical trial, bee venom was unable to decrease Unified Parkinson’s Disease Rating Scale (UPDRS) scores of PD patients

(Hartmann et al., 2016). Minocycline is another anti-inflammatory drug that is protective in the MPTP model of PD (Du et al., 2001; Wu et al., 2002). However, like bee venom, minocycline was unable to diminish PD symptoms in a clinical trial (Investigators, 2006).

Lastly, pioglitazone, a ligand for peroxisome proliferator-activated receptor gamma

(PPAR-γ), treatment results in protected neurons and reduced neuroinflammation in

MPTP-intoxicated mice (Laloux et al., 2012). However, pioglitazone was unable to improve outcomes in PD patents (Investigators., 2015; Simon et al., 2015). Despite neuroprotection and decreased neuroinflammation from the use of these and other therapeutics in PD models, no therapy to date has proven effective in PD patients.

GM-CSF represents a new potential therapy for PD. Pre-treatment with GM-CSF protects dopaminergic neurons and decreases neuroinflammation in the model of MPTP

(Kosloski et al., 2013). GM-CSF also increases the percentage of splenic Tregs in a dose-dependent manner. Adoptive transfer of Tregs from GM-CSF-treated mice are into

MPTP-intoxicated mice decrease neuroinflammation and neurodegeneration, suggesting that the neuroprotective benefits of GM-CSF are mediated, at least in part, by the induced Tregs. GM-CSF is a readily translatable drug since it is already FDA approved for humans. Sargramostim (Leukine) is a yeast-produced recombinant human protein which is FDA approved for reconstitution of the immune system post-chemotherapy

(Waller, 2007). Sargramostim decreased the symptoms of Crohn’s disease, an inflammatory gastrointestinal disorder (Korzenik et al., 2005; Takazoe et al., 2009; 30

Valentine et al., 2009), but no research has been done to determine if gut inflammation is decreased or Treg numbers are increased in these patients. The Nebraska Medical

Center initiated a phase I clinical trial to test the benefit of sargramostim in PD patients.

This trial tested Treg percentage and function as well as neurological changes, motor functions and metabolomic changes. If Tregs are having an in vivo effect, we hypothesized that anti-inflammatory cytokines would be increased with decreased proinflammatory cytokines expression by CD4+ CD25- T cells, which are suppressed by

Tregs. To test this, we isolated CD4+ CD25- T cells and used PCR arrays to test the expression of genes related to CD4+ T cells expression and differentiation. Here, we present the expression of these genes in healthy controls, and PD patients treated with either placebo or sargramostim.

METHODS

Study Design

A single-center, randomized, double-blind phase 1 clinical trial was performed at the University of Nebraska Medical Center (UNMC) and was approved by the UNMC

Institutional Review Board (Gendelman et al., 2017). Briefly, PD patients were recruited according to the following inclusion criteria: age 35-85, onset and persistence of symptoms ≥3 years and a Hoehn and Yahr disease scale ≤ stage 4. Exclusion criteria include diagnosis of multiple system atrophy, corticobasal degeneration, prior head injury, stroke, brain surgery, mental illness, cognitive impairment, autoimmune disease, systemic inflammatory disorder, hematologic disorders, PD symptoms lasting less than 3 years, and more than one blood relative diagnosed with PD. This trial also excluded individuals who had taken lithium, neuroleptics, or immunomodulatory treatment within 31

90 days of starting the study. Allergies to benzyl alcohol, colony-stimulating factors or yeast-derived products and ferrous metal implants were also grounds for exclusion.

Age-matched, non-PD controls were also recruited. After referral to the clinical research center (CRC), written informed consent was obtained in accordance with Good Clinical

Practice guidelines.

Healthy, non-PD controls, and PD patients had 3 pretreatment appointments to establish neurological and immunological baselines (visits 1-3). The 3rd appoint was the final appointment for the healthy controls and the PD patients were randomized in equal numbers to receive sargramostim or placebo. PD patients self-administered a subcutaneous injection of saline (placebo) or 6 µg/kg/day sargramostim every day for 56 days. Every two weeks during this treatment phase, patients continued doctor visits for blood draws and to monitor health and motor function (visits 4-7). Patients returned for a post-treatment visit one month after ceasing treatment (visit 8). During each visit, blood was drawn, the patient was examined by a physician, and the UPDRS part III clinical assessment was performed. Patients remained on anti-PD medications and UPDRS III evaluation was performed while patients were in the "ON" state. Table 2.1 is a summary of the study design and the number of patients from each treatment group included for the gene expression analysis.

A subset of healthy controls, placebo-treated PD patients, and sargramostim- treated PD patients were used for the following studies. Multiple methods were used to

32

Table 2.1 Study design

33

Table 2.1 Study design

The table describes of the study design for this clinical trial. The treatment phase, visit, and weeks indicate when each visit occurs relative to starting treatment with placebo or sargramostim and which visit belongs to each treatment phase. For the treatment groups, the number of patients from which RNA was isolated out of the total number of patients is indicated.

34 isolate CD4+ CD25- cells, which were isolated from blood and combined with CD4+

CD127low CD25- cells remaining after Treg isolation. The combination was used for flow cytometric analysis and RNA isolation.

Isolation of CD4+ cells

CD4+ cells were isolated by isolating lymphocytes using lymphocyte separation media (LSM, MP Biomedicals LLC Santa Ana, CA) prior to positive CD4+ isolation.

Briefly, the whole volume of a single tube of blood was diluted by the addition of an equal volume of 1x HBSS (Gibco, Carlsbad, CA). The lymphocytes and monocytes were separated by layering diluted blood on top of LSM before centrifugation at 400 xg for 20 min without breaking. The buffy coat was transferred to a new tube and was washed in 3 volumes of 1x HBSS. Cells were concentrated by centrifugation at 200 xg for 10 min. To the cell pellet, 80 μl of Miltenyi CD4+ buffer (0.5% bovine serum albumin

(BSA), 2 mM EDTA in 1x DPBS (Gibco)) and 20 μl anti-CD4 magnetic beads (Miltenyi

Biotec Auburn, CA) were added for every 1x107 cells. After incubation at 4°C for 15 min, cells were washed in 2 ml Miltenyi CD4+ buffer for every 1x107 cells and cells were concentrated by centrifugation at 300 xg for 10 min. The cell pellet was resuspended in

500 μl Miltenyi CD4+ buffer and the whole volume was pipetted into a Miltenyi LS column which was placed in a magnet. The column was washed three times with 2 ml

Miltenyi CD4+ buffer. Five milliliters of Miltenyi CD4+ buffer was added to the column, the column was removed from the magnet, and the buffer was forced through the column into a 15 ml conical tube using a plunger. This procedure was used to isolate cells for XIL visits 1 and 2, YEA visits 1 and 2, ZNB visits 1 and 2, BED visit 1, CAE visit

1, DBF visit 1, and ECG visit 1.

Isolation of Responder T cells 35

Responder T cells were isolated according to the protocol for the R&D CD4+ enrichment kit as follows. Briefly, lymphocytes were isolated by LSM as described above. The resulting lymphocytes were resuspended in 1 ml 1x R&D CD4+ enrichment kit column buffer (R&D systems, Minneapolis, MN). In addition to antibodies for the selection of CD4+ CD25- cells, 2 μg/1x106 cells of anti-CD16 (clone 3G8 BD, Franklin

Lakes, NJ), anti-CD8 (clone HIT8a BD), and anti-CD25 (clone M-A251 BD) and 1

μg/1x106 cells anti-CD16 (clone 245536 R&D Systems) was added to enhance purity.

Cells were incubated with antibodies at RT for 15 min mixing the tube every 2 min, washed 2 times in 1x column buffer, and resuspended in 2 ml R&D column buffer before adding to the column. After incubating the cells on the column for 10 min, the cells were washed through the column with R&D column buffer and cells were concentrated by centrifugation at 400 xg for 5 min. These cells were combined with the cells isolated following the cell rosetting protocol as described below. This method was used for isolating cells from the following patient visits: XIL visits 3, 4 and 5, YEA visit 3, ACC visit

3, ZNB visits 3, 4 and 5, BED visit 2, CAE visit 2, DBF visit 2, ECG visit 2, FDH visits 1 and 2, HFJ visits 1 and 2, and IGK visits 1 and 2.

Isolation of CD4+ CD25- cells

As with the responder T cells, lymphocytes and monocytes were isolated by LSM and these buffy coat cells were washed in 1x HBSS. To the cell pellet, 80 μl of Miltenyi

CD4+ buffer (0.5% BSA, 2 mM EDTA in 1x DPBS) and 20 μl anti-CD25 magnetic beads

(Miltenyi Biotec) were added for every 1x107 cells. After incubation at 4°C for 15 min, cells were washed in 2 ml Miltenyi CD4+ buffer for every 1x107 cells, concentrated by centrifugation at 300 xg for 10 min, resuspended in 500 μl Miltenyi CD4+ buffer, and the whole volume was added into a Miltenyi LD column. The column was washed 3 times with 2 ml Miltenyi CD4+ buffer. The CD25-depleted cells were concentrated by 36

centrifugation at 300 xg for 10 min, and 40 μl Miltenyi CD4+ buffer and 10 μl CD4+ antibody cocktail was added for every 1x107 cells. After incubation at 4°C for 5 min, 30

μl Miltenyi CD4+ buffer and 20 μl of Miltenyi microbeads were added for every 1x107 cells. After incubation at 4°C for 10 min, 500 μl Miltenyi CD4+ buffer was added and the whole volume was added into a Miltenyi LS column. Cell effluents through the column were collected as CD4+ CD25- T cells. Cells were counted and combined with T cells isolated from the rosetted procedure for flow cytometry and RNA isolation. This method was used for T cell isolation for the following patient visits: XIL visits 7 and 8, ZNB visits

6, 7 and 8, BED visits 3-8, CAE visits 3, DBF visits 3-8, ECG visits 3, HFJ visits 3-8, IGK visit 3, JHL visits 1-8, KIM visits 1-3, LJN visits 1-8, MKO visits 1-8, and NLP visits 1-8.

Isolation of CD4+ CD127low CD25- cells

Tregs (CD4+ CD127low CD25+) were isolated from 6 tubes of blood using the

RosetteSep human CD4+ CD127low CD25+ isolation kit (Stemcell Technologies,

Vancouver, BC) as follows. Briefly, 50 μL/mL RosetteSep human CD4+ CD127low pre- enrichment cocktail was added to whole blood and the mixture was incubated for 20 min at RT. Whole blood (25 ml) was layered onto 15 ml density media in a Sep-Mate-50 tube and cells were separated by centrifugation at 1,200 xg for 10 min. The top layer, which contains mononuclear cells, was decanted and washed two times in Treg buffer

(1x DPBS with 2% fetal bovine serum (FBS, Sigma, St. Louis, MO)). To isolate CD25+ cells, EasySep positive selection cocktail was added at 50 µl/ml, incubated at RT for 15 min, 50 µl nanoparticles/ml was added, and incubated at RT for 10 min. The tube was placed in a magnet field and incubated for 5 min. While remaining in the magnetic field, cells were washed 9 times by decanting supernatant and 2.5 ml of Treg buffer was added. The negatively selected CD25- T cells decanted in the washes were concentrated by centrifugation at 400 xg for 5 min. Red blood cells in the pellet were 37

osmolysed by incubating in 9 ml autoclaved ultrapure water for 30 sec and 1 ml 10x PBS was added to restore osmolarity and cells were concentrated by centrifugation at 400 xg for 5 min. The cells were counted and combined with the CD4+, responder T cells or

CD4+ C25- cells isolated above for flow cytometry and RNA isolation. This method was performed for all patient visits and these cells combined with cells isolated using the above methods.

Flow cytometry

Flow cytometric analysis was performed at the UNMC flow cytometry research facility. A volume corresponding to 250,000 cells was brought to 100 μl by the addition of flow cytometry stain buffer (FSB, 0.5% BSA and 0.1% sodium azide in 1x DPBS) and the cells were stained by the addition of anti-CD4-FITC (clone RPA-T4 BD), anti-CD25-

PE (clone M-A251 BD), anti-CD127-PERCPCy5.5 (clone HIL-7R-M21BD), anti-

CD45RA-AF700 (clone HI100, BD), anti-CD45RO-APC (clone UCHL1, BD) and anti-

CCR7-PECy7 (clone 3D12 BD) at 4°C for 20 min. Cells were washed twice by the addition of 2 ml FSB per wash, concentrated by centrifugation at 400 xg for 5 min, resuspended in 500 μl flow cytometry fix (1% formaldehyde in 1x DPBS), and cells were incubated at RT for 10 min. Fixed cells were concentrated by centrifugation at 400 xg for 5 min. The cell pellet was resuspended in 300 μl FSB prior to analysis with a BD

LSR II flow cytometer interfaced with FACSDiva analytical software (version 8.0) (BD

Biosciences, San Jose, CA).

RNA isolation, cDNA conversion and PCR arrays

With fewer than 3x106 total cells, then all cells were pelleted at 400 xg for 10 min and frozen at -80°C for RNA isolation when remaining samples were collected. With more than 3x106 cells, half of the cells were pelleted and frozen as above. The other 38

half of the cells were brought to ~1x106 cells/ml with the addition of complete media

(RPMI 1640 supplemented with L-glutamine (Gibco), 10% FBS (Sigma), 1 mM sodium pyruvate (Gibco), 1x MEM non-essential amino acids (Hyclone, Logan, Utah), 1x

Penicillin-Streptomycin (Gibco), 10 mM HEPES (Sigma), 2 mM L-glutamine (Gibco) and

55 nM 2-mercaptoethanol (Gibco)) and 100 μl was added to each well of a U-bottom 96- well plate. An additional 100 μl media containing ~5x105 anti-CD3/CD28 magnetic, co- stimulatory dynabeads was added to each well (Gibco). After 6 hours of stimulation, the cells were removed from the 96-well plate, combined and concentrated at 400 xg for 10 min, and cell pellets were frozen at -80°C.

RNA was isolated according to the directions in the RNeasy mini kit (Qiagen,

Valencia, CA). Briefly, cell pellets were resuspended in Buffer RLT (350 µl RLT buffer/~5x106 cells) supplemented with 10 μl β-mercaptoethaol/ml RLT and cells were lysed by passing through a 20 Ga needle 10 times. The cell lysate was clarified by centrifugation at 20,800 xg for 3 min, supernatant added to an equal volume of 70% ethanol, added to a spin column, and passed through at 10,800 xg for 1 min. The flow through was discarded and the filter was washed in RW1 buffer. DNA was digested using the RNase-free DNase kit (Qiagen) by adding 10 μl DNase I and 70 μl RDD buffer per column, and the digestions were incubated at RT for 15 min. The column was washed in RW1 a second time and then in RDD buffer two times. RNA was eluted by the addition of water. The eluted RNA was passed through the column a second time to maximize RNA yield and RNA concentration was determined by UV spectrometry at

260, 280, and 230 nm (NanoDrop spectrophotometer, ThermoScientific, Waltham, MA).

Single-stranded cDNA was made as directed using the RevertAID single strand cDNA synthesis kit (ThermoScientific). Briefly, depending on the RNA concentration,

100, 125 or 250 ng RNA was brought to 11 μl with water provided in the kit and 1 μl 39

poly(T) primers and the secondary structure of RNA was disrupted by heating at 65°C for 5 min. A master mix of 4 μl reaction buffer, 2 μl dNTPs, 1 μl ribolock RNase inhibitor and 1 μl reverse transcriptase/sample was added and synthesis was performed at 42°C for 1 hr. The reaction was terminated by heating at 70°C for 5 min. The multiple reactions of each sample were combined and frozen at -20°C until the PCR arrays were performed.

The expression of genes related to helper T cell differentiation was determined using a Helper T cell differentiation array which was performed according to the manufacturer’s protocol SABiosciences (Frederick, MD). Briefly, water, 2x RT2 ROX master mix, and the cDNA was combined and 25 μl was added to each well of the array.

PCR was performed using an Realplex S2 thermocycler (Eppendorf, Hamburg,

Germany) with a 10 min hot start at 95°C, then 40 cycles of a two-steps of 95°C for 15 sec and 60°C for 1 min followed by a melting curve. GAPDH was used as the housekeeping gene. Data analysis was performed using the Qiagen software and fold changes were determined using the ΔΔCt method.

Statistical analysis

For flow cytometry samples, the Mann-Whitney U test was used to determine significant differences between the healthy controls, placebo-treated PD patients and sargramostim-treated PD patients at each visit. Statistics for flow cytometry data were performed by GraphPad (La Jolla, CA) Prism version 6. For the PCR arrays, the RT2

Profiler PCR array data analysis software version 3.5 was used to calculate the fold change between samples. For downregulated genes, fold changes are the negative inverse of fold change.

40

RESULTS

Flow cytometry characterization of CD4+ CD25-

Flow cytometric analysis was performed on CD4+ CD25- T cells isolated from participants to identify various subpopulations of T cells over the trial in the treatment groups. In Figure 2.1A, the percentage of CD4+ T cells in the total population for each visit was graphed. Consistently, around 40% of the total population was CD4+ T cells for the healthy controls, the placebo-treated PD patients and the sargramostim-treated

PD patients. The identity of the non-CD4 cells was not determined, though they appeared to be the same size and granularity as the CD4+ cells suggesting they were lymphocytes and/or monocytes. In Figure 2.1B, no significant differences in the percentage of naïve CD4+ T cells (CD4+ CD45RA+ CCR7+) were discerned between placebo-treated and sargramostim-treated PD patients, with the exception at visit 7 where sargramostim-treated patients have a significant decrease in these T cells.

Healthy controls were not different from either PD patient group at visits 1-3. No differences were detected between the percentage of central memory (CD4+ CD45RO+

CCR7+) or effector memory (CD4+ CD45RO+ CCR7-) helper T cells between healthy controls, placebo-treated, and sargramostim-treated PD patients at any visit (Figure

2.1C and 2.1D respectively). Interestingly, sargramostim-treated PD patients tend to possess increased effector memory helper T cells compared to the placebo-control patients at visits 6 and 7 (Figure 2.1D). In Figure 2.1E, the percentages of Tregs (CD4+

CD127- CD25+) were not different between the healthy controls, placebo-treated PD patients, and sargramostim-treated PD patients at any visit. At visit 6, activated effector

T cells (CD4+ CD127+ CD25+) in sargramostim-treated PD patients were elevated compared to placebo-treated PD patients, but this was the only visit with a significant 41

Figure 2.1 Flow cytometric analysis of CD4+ CD25- cells

42

Figure 2.1 Flow cytometric analysis of CD4+ CD25- cells

CD4+ CD25- cells were isolated from the blood of healthy controls (black lines), placebo- treated (blue lines) and sargramostim-treated (red lines) PD patients at each visit. Flow cytometric analysis was used to determine the surface expression of CD4, CD127,

CD25, CD45RO, CD45RA and CCR7. Within the whole blood, we determined the percentage of total CD4+ T cells (A). The percentage of naïve helper T cells (CD4+

CD45RA+ CCR7+) (B), central memory helper T cells (CD4+ CD45RO+ CCR7+) (C), effector memory helper T cells (CD4+ CD45RO+ CCR7-) (D), regulatory T cells (CD4+

CD127- CD25+) (E) activated effector helper T cells (CD4+ CD127+ CD25+) (F) within the total CD4+ T cell population was determined. Significance was determined by the

Mann-Whitney U test at each visit. a-p < 0.05.

43 difference and there were no differences compared to the healthy controls (Figure 2.1F).

These data demonstrate that the isolated T cells population showed roughly equal percentages of these T cell subpopulations over time. As such, changes in gene expression were not due to differences in cell type being analyzed.

Gene expression comparing to PD patients to controls

After isolating RNA from each sample, we determined the A260/A80 and

A260/A230 by UV spectrometry. Each PCR array contains a genomic DNA contamination control. The result is included in Table 2.2 to give an indication of

RNA/cDNA quality. The first gene expression analysis we performed was to compare all

PD patients to healthy controls pre-treatment (visits 1-3). In unstimulated CD4+ CD25- cells (Figure 2.2A), PD patients upregulated more than 2-fold several genes associated with proinflammatory cell types. For instance, STAT4 is a transcription factor induced by

IL-12 (Watford et al., 2004), EOMES, a transcription factor upregulated in activated Th1 helper T cells (Lupar et al., 2015), RORC is the master transcription factor for Th17 helper T cells (Martinez et al., 2008; Jetten, 2009), and IL18RAP is the beta chain for the

IL-18 receptor (Fiszer et al., 2007). PD patients also downregulated by more than 2-fold anti-inflammatory genes such as CCR4, a chemokine receptor selectively expressed on

Th2 helper T cells (Yoshie and Matsushima, 2015), and ICOS, a co-stimulatory molecule the favors development of Th2 helper T cells and proliferation and survival of Tregs

(Simpson et al., 2010).

We also analyzed the gene expression of CD4+ CD25- cells stimulated with

CD3/CD28 beads for 6 hours from healthy control and all PD patients prior to treatment

(Figure 2.2B). PD patients increase the expression of IRF4, a gene involved in the differentiation and function of Th2 and Tregs (Zheng et al., 2009), by more than 2-fold 44

Table 2.2 RNA quality analysis

45

Table 2.2 RNA quality analysis

A table listing the ratio of the RNA absorbance at 260 and 280 nm (A260/A280) and the ratio of RNA absorbance at 260 and 230 nm (A260/A230). Lastly, we tested each sample for genomic contamination of the cDNA. There was no genomic contamination when the ct value for the included control well on each PCR array was >35. When the ct value was in the cycle 30-35 range, genomic contamination is possible. Genomic DNA was confirmed when the ct value was less than cycle 30. The PCR array for patient

DBF at visit 2 unstimulated could not be determined because an improper PCR cycle program was used and there was insufficient sample to repeat the analysis. This sample was excluded from all analyses.

46

Figure 2.2 Gene expression comparing PD patients to healthy controls pre-treatment with and without CD3/CD28 stimulation

47

Figure 2.2 Gene expression comparing PD patients to healthy controls pre- treatment with and without CD3/CD28 stimulation

RNA was extracted from the isolated CD4+ CD25-, and cDNA copies were made from it prior to PCR using an array for human CD4+ T cell differentiation. The fold change in expression from each gene was compared for the PD patients compared to healthy controls using the ΔΔCt method. Heat maps were generated for all genes more than 2- fold up- or downregulated for (A) unstimulated and (B) CD3/CD28 bead-stimulated CD4+

CD25- cells. The numbers are the fold change. Red shaded cells are upregulated genes and green shaded cells are downregulated genes. The shade of color denotes the degree of upregulation.

48

compared to healthy controls. However, PD patients exhibited 2-fold decreased expression of anti-inflammatory genes such as IL5, TNFSF11, CCR4, CCL7, IL9,

PTGDR2 and ICOS. Expression of proinflammatory genes such as SOCS5, IL17RE and IL17A were decreased in PD patients. Combined, these data suggest PD patients express more proinflammatory genes at rest compared to healthy controls. However, after stimulating CD4+ CD25- cells with CD3/CD28 beads, PD patients display lower gene expression compared to healthy controls.

Gene expression compared to baseline for placebo or sargramostim PD patients

We next tested the expression of genes in the placebo-treated and sargramostim-treated PD patients at visits 4, 5, 6, 7 and 8 compared to the baseline

(visits 1-3) in unstimulated CD4+ CD25-. Figure 2.3A is a heatmap of all 2-fold gene expression changes in placebo-treated PD patients for each visit compared to the baseline. The gene changes in the placebo-treated PD patients were smaller compared to sargramostim-treated patients and no discernible peak in the expression was evident over time.

Figure 2.3B is a heat map of all 2-fold gene changes at each visit for the sargramostim-treated patients compared to baseline. Some increased genes were non- associated genes; those involved in T cell differentiation and proliferation and included

GATA4, TNFRSF9, IL2, KIF2C, and HOXA3. Expression of proinflammatory genes were increased including IL21, IL12B, IL17A, IL1R1, IL17RE, SOCS5, and HAVCR2.

Increased anti-inflammatory gene expression included IL4, CCL7, IL1RL1, IL9, PPARG,

LRRC32, TNFSF11, IL13, IL5, PTGDR2, CCR4, and ICOS. Expression of non- associated genes such as STAT1, MAF, and RUNX3 were also decreased. Several proinflammatory genes displayed decreased expression including STAT4, RORA, IRF1, 49

Figure 2.3 Gene expression comparing unstimulated CD4+ CD25- on-treatment and post-treatment visits to baseline for placebo- or sargramostim-treated PD patients 50

Figure 2.3 Gene expression comparing unstimulated CD4+ CD25- on-treatment and post-treatment visits to baseline for placebo- or sargramostim-treated PD patients

RNA was extracted from the isolated CD4+ CD25-, and cDNA copies were made from it prior to PCR using an array for human CD4+ T cell differentiation. The fold change of expression from each gene was compared for the placebo and sargramostim-treated PD patients on-treatment (visits 4-7) and post-treatment (visits 4-8) relative to baseline

(visits 1-3) using the ΔΔCt method. Heat maps were generated for all genes more than

2-fold up- or downregulated for the unstimulated CD4+ CD25- from (A) placebo-treated

PD patients and (B) sargramostim-treated PD patients. The numbers are the fold change. Red cells are upregulated genes and green cells are downregulated genes.

The color shade denotes the degree of upregulation. Pro-inflammatory genes are those related to Th1 and Th17 cells. Anti-inflammatory genes are related to Th2 and Tregs.

The non-associated genes are found in all T cells, regardless of subset.

51

and IL18RAP. Anti-inflammatory genes including CCL5, ID2, Jak1, STAT6, and GATA3 were also decreased. In general, the peak in gene expression was at visit 5. At visits 6 and 7, expression remained elevated in sargramostim-treated PD patients compared to baseline, but the expression was decreased compared to visit 5. In general, there is no peak in decreased expression of genes. Interestingly, there was a trend to decreasing expression of CCL5 (the gene for the chemokine RANTES) over treatment, with expression lowest at visit 7. Post-treatment (visit 8), most genes returned to near baseline, though some genes remained increased or decreased.

Counter to our hypothesis, sargramostim increased the expression of both pro- and anti-inflammatory genes compared to baseline in CD4+ CD25- T cells. It is not surprising that sargramostim increased the expression of genes associated with proliferation, since sargramostim increases the number and percentage of CD4+ T cells compared to placebo control (unpublished data). The increase in anti-inflammatory genes is in line with increased Treg number and function from peripheral blood

(Gendelman et al., 2017). From the data presented here, proinflammatory genes were also increased. Interestingly, sargramostim does not change in the number of Teffs

(CD4+ CD127+ CD25+) in whole blood and the T cells isolates for these analyses compared to placebo controls. These data demonstrate that sargramostim can stimulate the expression of proinflammatory and anti-inflammatory genes in helper T cells.

However, to what degree these changes in gene expression promote the differentiation of helper T cells to different subsets was not determined.

Because the identity of helper T cells is often revealed after activation, we stimulated cells with CD3/CD28 beads. Figure 2.4A is a heat map of the 2-fold gene changes in CD3/CD28 bead-stimulated CD4+ CD25- cells from placebo on- or post- 52

Figure 2.4 Gene expression comparing CD3/CD28-stimulated CD4+ CD25- at baseline to on- and post-treatment visits for placebo-or sargramostim-treated PD patients 53

Figure 2.4 Gene expression comparing CD3/CD28-stimulated CD4+ CD25- at baseline to on- and post-treatment visits for placebo- or sargramostim-treated PD patients

RNA was isolated from the isolated CD4+ CD25-, and copied to generate cDNA prior to

PCR using an array for human CD4+ T cell differentiation. The fold change of expression from each gene by PD patients was compared to healthy controls using the

ΔΔCt method. Heat maps were generated for all genes that were more than 2-fold up- or downregulated for the stimulated CD4+ CD25- from (A) placebo-treated PD patients and (B) sargramostim-treated PD patients. The numbers are fold change. Red shaded cells are upregulated genes and green shaded cells are downregulated genes. The shade of color denotes the degree of upregulation. Pro-inflammatory genes are those related to Th1 and Th17 cells. Anti-inflammatory genes are related to Th2 and Tregs.

The non-associated genes are found in all T cells and are not related to only subset.

54

treatment compared to baseline. We were unable to isolate sufficient cells for stimulation at each visit for each patient. Due to lack of samples, data were grouped by treatment phase as opposed to grouping by visit. As in Figure 2.3, gene changes are smaller in the placebo-treated PD patients compared to sargramostim-treated PD patients. In addition, there were no apparent changes were discerned in the pattern of gene expression on treatment or post treatment compared to baseline.

Figure 2.4B is a heat map of all 2-fold gene changes in CD3/CD28 bead- stimulated CD4+ CD25- T cells from sargramostim-treated PD patients on- or post- treatment compared to baseline. CD3/CD28 bead-stimulated CD4+ CD25- T cells from the sargramostim-treated patients on treatment showed upregulated several non- associated genes such as GATA4, HOXA10, PERP, HOXA3, and KIF2C and a downregulated IL2, STAT1, MAF, and IL2RA. Several proinflammatory genes were upregulated such as IL17A, IL17RE, IL18, IL12B, and RORC, while several were downregulated such as IFNG, TNF, STAT4, and IL18RAP. Anti-inflammatory genes that were upregulated included PTGDR2, ASB2, UTS2, PPARG, IL4, ICOS, and LRRC32 were increased. But the expression CCL5, FOXP3, IRF4, STAT6, REL, and JAK1 were downregulated. Many of the genes changed in the CD3/CD28-stimulated cells are similar to the genes changed in unstimulated cells (Figure 2.3). For example, non- associated genes like GATA4 and PERP, pro-inflammatory genes like IL17A and

IL17RE, and anti-inflammatory genes like IL4 and PTGDR2 are upregulated in both unstimulated- and CD3/CD28-stimulated CD4+ CD25- cells. Interestingly, the fold change is smaller after CD3/CD28-stimulation compared to unstimulated cells. In general, the fold change on-treatment is higher and returns to near baseline at the post- treatment visit, however only one sample in the sargramostim-treated post-treatment group. 55

Combined, Figures 2.3 and 2.4 demonstrate sargramostim increases the expression of many genes, both pro- and anti-inflammatory, in unstimulated CD4+

CD25- T cells compared to pre-treatment baseline. However, stimulating this cell population with CD3/CD28 beads also increased the expression of these genes in all treatment phases, but decreases the effect of sargramostim treatment on gene expression.

Gene expression comparing the placebo to sargramostim PD patients

Figures 2.3 and 2.4 demonstrated changes in gene expression relative to baseline for the placebo controls and sargramostim-treated group. Another way of analyzing these data is to test sargramostim-treated gene expression relative to the placebo control at each visit. In Figure 2.5, the gene expression in unstimulated CD4+

CD25- cells in the sargramostim-treated PD patients was compared to the placebo- treated PD patients at each visit. At baseline (visits 1-3), some genes that were more than 2-fold up- or downregulated. These changes were relatively small, indicating little differences in gene expression between T cells isolated from PD patients at baseline.

Starting by 2 weeks (visit 4) after treatment initiation and increasing to visit 5 noticeable increases were evident in expression of non-associated genes such as GATA4,

TNFRSF9, PERP, and KIF2C. Expression of several non-associated genes such as

HOXA3, RUNX1, STAT4, and STAT1 also was decreased. As in Figure 2.3, expression of proinflammatory genes such as IL21, IL12B, IL1R1, and IL17A was increased.

Downregulated expression of proinflammatory genes included RORA and IL18RAP. As hypothesized, sargramostim treatment increased the expression of anti-inflammatory genes such as CCL7, TNFRSF11, IL13, IL1RL1, IL4, LRRC32, PPARG, and PTGDR2, but downregulated some genes such as STAT6, CCL5, and ID2. 56

Figure 2.5 Gene expression comparing unstimulated CD4+ CD25- cells from placebo- treated compared to sargramostim-treated PD patients 57

Figure 2.5 Gene expression comparing unstimulated CD4+ CD25- cells from placebo-treated compared to sargramostim-treated PD patients

RNA was extracted from the isolated CD4+ CD25-, and copied to cDNA prior to PCR using an array for human CD4+ T cell differentiation. Fold changes of expression from each gene was compared for the PD patients compared to healthy controls using the

ΔΔCt method. Heat maps were generated for all genes more than 2-fold up- or downregulated for the unstimulated CD4+ CD25 cells. Red shaded cells are upregulated genes and green shaded cells are downregulated genes. The color shade denotes the degree of upregulation. Pro-inflammatory genes are those related to Th1 and Th17 cells. Anti-inflammatory genes are related to Th2 and Tregs. The non- associated genes are found in all T cell subsets.

58

These data are in line with Figure 2.3 demonstrating that sargramostim increases the expression of CD4+ T cell proliferation/differentiation, proinflammatory and anti- inflammatory genes. The peak expression appears to be at visit 5 (4 weeks after initiation of treatment), although the timing is different for different genes. Gene expression appears to remain elevated at visits 6 and 7, but by visit 8 (the post- treatment visit), gene expression returned to near baseline. This demonstrates that sargramostim treatment increased the expression of genes relative to placebo patients as well as relative to baseline.

We also tested the expression of genes in the CD3/CD28 bead-stimulated CD4+

CD25- T cells from sargramostim-treated PD patients relative to placebo-treated PD patients (Figure 2.6). As in Figure 2.4, there were insufficient samples to compare visit by visit, so we tested by treatment phase. There was only one sample for the sargramostim treatment group post treatment. As in Figure 2.5, at baseline, there were some genes that were more than 2-fold increased or decreased in expression after sargramostim treatment. These were relatively small changes and indicative that PD patients display similar responses to CD3/CD28 stimulation at baseline. During the on- treatment phase, sargramostim treatment increased the expression of several non- associated genes, such as GATA4, PERP, and KIF2C. Sargramostim also downregulated several non-associated genes such as IL2, STAT4, IL2RA, STAT4, and

STAT1. Even after ceasing sargramostim treatment, KIF2C and GATA4 expression remained elevated.

Sargramostim treatment also increased the expression of proinflammatory genes such as IL17RE, IL17A, TLR6, and IL1R1, and significantly downregulated the expression of several proinflammatory genes including IFNG, TNF, IL18RAP, RORA, 59

Figure 2.6 Gene expression comparing unstimulated CD4+ CD25- T cells from placebo- treated PD patients compared to sargramostim-treated PD patients

60

Figure 2.6 Gene expression comparing unstimulated CD4+ CD25- T cells from placebo-treated compared to sargramostim-treated PD patients

RNA was extracted from the isolated CD4+ CD25- T cells and copied to cDNA, prior to

PCR using an array for human CD4+ T cell differentiation. The fold change of expression from each gene was compared for the PD patients compared to healthy controls using the ΔΔCt method. Heat maps were generated for all genes more than 2- fold up- or downregulated for the stimulated CD4+ CD25 cells. Red colored cells are upregulated genes and green colored cells are downregulated genes. The color shade denotes the degree of upregulation. Pro-inflammatory genes are those related to Th1 and Th17 cells. Anti-inflammatory genes are related to Th2 and Tregs. The non- associated genes are found in all T cells subsets.

61

and IRF1. Post-treatment, gene expression appears to return to near baseline for both the upregulated and downregulated genes. Sargramostim also increased the expression of several anti-inflammatory genes such as IL1R1, HOXA10, TNFSF11 and

PTGDR2, and downregulated several anti-inflammatory genes such as IRF4, IL4R,

CCL5, NR4A1, and ID2. Post-treatment gene expression returned to near baseline for all genes, although CEBPB was upregulated on-treatment and is decreased post- treatment.

These data demonstrate that sargramostim alters expression of many genes in

CD3/CD28-stimulated CD4+ CD25- T cells. As in Figure 2.4, the increases in gene expression are smaller in the CD3/CD28 bead-stimulated compared to the unstimulated cells. This is probably due to the CD3/CD28 beads activating, and leading to upregulation of gene expression in the placebo control cells, thereby diminishing the relative increase in gene expression due to sargramostim. Combined, Figures 2.3, 2.4,

2.5, and 2.6 demonstrate that sargramostim does increased the expression of genes relative to both the pre-treatment baseline as well as placebo treatment. This increase in expression is in genes associated with T cell proliferation and differentiation, and in proinflammatory and anti-inflammatory genes, demonstrating broad effects across T cell subsets.

Gene expression of individual placebo and sargramostim PD patients

From the above data, values were combined for all patients as another way to analyze these data by changes in gene expression for each patient. Seven patients (4 placebo controls (BED, HFJ, JHL, and MKO) and 3 sargramostim-treated patients (DBF,

LJN, and NLP)) who had data for all 8 trial visits were utilized. Gene expression of unstimulated cells at visits 4, 5, 6, 7 and 8 was determined relative to the baseline (visits

1-3) for each patient. In addition to determining gene expression for each individual, we 62

also tested if the degree of gene change was related to changes in the percentages of peripheral Tregs, which were increased in the sargramostim-treated patients

(Gendelman et al., 2017). Figure 2.7 shows the extent of gene expression as a function of CD4+ CD25+ Foxp3+ Treg percentages in peripheral blood for the 2 genesthat were most increased by sargramostim treatment (Figure 2.5) for each category including non- associated genes (Figure 2.7A and B), proinflammatory genes (Figure 2.7C and D), and anti-inflammatory genes (Figure 2.7E and F). The degree of gene change in response to sargramostim treatment varies between patients, but consistently patients LJN and

NLP showed the greatest gene change. These patients also had the highest percentage of Tregs within the CD4+ population. Increase in Tregs percentages was associated with increased expression of all genes reported. The sargramostim-treated patient DBF displayed gene expression and Treg percentage closer to the placebo controls, suggesting that this patient’s response was less compared to patients LJN and NLP.

This figure shows that while each patient responded to different degrees, the trend within the group is replicated with each patient. Notably, patients with the highest increase in gene expression also showed the highest increase in the percentage of Tregs, which provides evidence that these patients responded to sargramostim treatment. We hypothesized that the increase in Tregs would reduce the expression of proinflammatory genes; however, patients with the highest percentage of Tregs also display elevated expression of proinflammatory genes. This suggests that Tregs are not suppressing the expression of inflammatory cytokines in this cell population.

63

Figure 2.7 Gene expression related to Treg frequency in placebo- and sargramostim- treated PD patients

64

Figure 2.7 Gene expression related to Treg frequency in placebo- and sargramostim-treated PD patients

For each patient, the gene expression was compared to the baseline. The frequency of

Tregs (CD4+ CD25+ Foxp3+) within the CD4+ population was determined from the whole blood. GATA4 (A) and TNFSF9 (B) are the non-associated genes with the highest expression in sargramostim-treated patients relative to placebo control PD patients. IL21 (C) and IL12B (D) are the pro-inflammatory genes with the highest expression in sargramostim-treated patients compared to placebo controls. CCL7 (E) and TNFSF11 (F) are the two anti-inflammatory genes with the highest expression in sargramostim-treated patients relative to placebo controls. Each patient is a separate color. Placebo controls are filled circles and the sargramostim-treated patients are open circles.

65

Ingenuity Pathway Analysis of placebo compared to sargramostim PD patients

To determine how the genes whose expression was altered by sargramostim related to helper T cell subsets, Ingenuity Pathway Analysis (IPA) was used. In the unstimulated T cells, comparing sargramostim-treated patients relative to placebo control showed genes altered by sargramostim associated with all helper T cell subsets

(Figure 2.8A). In addition, increases in GM-CSF (which in the figure is indicated by the gene name CSF2) is associated directly or indirectly to the expression of several genes which were increased, including IL-2R, TNF-α, IL-4, IFNγ, IL-15, IL-13 and Foxp3.

Mapping the same pathways using stimulated T cell expression data to compare sargramostim-treated patients to placebo-treated patients, showed downregulation of many genes, including genes were increased in the unstimulated cells (IL2R, TNFα,

IFNγ, and IL13) (Figure 2.8B). \We also identified that GM-CSF (CSF2) was directly or indirectly associated with the expression of several altered genes including STAT1, TNF-

α, IFNγ, IL-13, Foxp3 and IL-2R.

DISCUSSION

In this chapter, gene changes in the CD4+ CD25- T cells in PD patients treated with sargramostim were described. First, the percentages of different populations of T cells over time were determined to ensure that the distribution of analyzed T cells did not change over time. While some significant differences were evident at individual visits in cells isolated between placebo-treated and sargramostim-treated T cells, overall the distribution of analyzed cells did not change over time or between treatment groups.

Interestingly, that sargramostim increased the percentage of effector memory helper T 66

Figure 2.8 Ingenuity pathway analysis of genes altered by sargramostim in unstimulated and stimulated CD4+ T cells

67

Figure 2.8 Ingenuity Pathway Analysis of genes altered by sargramostim in unstimulated and stimulated helper T cells

Ingenuity Pathway Analysis was used to map the gene changes from sargramostim- treated patients relative to placebo controls for the (A) unstimulated cells and (B) stimulated cells. The T cell differentiation pathways are showed. Genes that are upregulated are red, downregulated genes are green. In this analysis, the genes which are directly or indirectly associated with GM-CSF (CSF2) are indicated by purple lines.

68

cells and fewer naïve helper T cells, possibly indicating an activation of helper T cells, though this may be an artifact of isolation. Sargramostim increases many genes associated with proliferation and differentiation of CD4+ T cells, which is consistent with the increase in total CD4+ T cells in the whole blood of sargramostim-treated patients compared to placebo controls (unpublished data). It should be noted that CD25+ T cells were positively selected against for these analyses. Given that ~2% of the analyzed cells are Teffs (CD4+ CD127+ CD25+) and ~2% are Tregs (CD4+ CD127- CD25+), most cells are CD25-. However, by excluding CD25+ T cells, this analysis may be missing relevant gene changes in the effector T cell population, which may be the cells expressing genes that effect PD. However, since CD4+ CD25- cells are the responder T cells suppressed by Tregs in the in vitro proliferation suppression assay (Gendelman et al., 2017), these are one population cells that Tregs target to suppress in vivo.

Additional studies would be needed to determine which helper T cell populations and functions are most important for PD pathogenesis and to what degree Tregs suppress these cells and functions.

As expected, there was an increase in the expression of anti-inflammatory genes. Some of these are genes for cytokines released from Th2 cells, such as IL4,

IL13, and IL5 (Zhu and Paul, 2008). Treg-related genes were also increased such as

LRRC32, the gene for GARP, a transcription factor which is important for Treg development (Tran et al., 2009). IKZF2 is a gene within the IKAROS family of transcription factors, including Helios, which is expressed in Tregs (Bhairavabhotla et al.,

2016). Thymic-induced (or natural) Tregs contain increased IKZF2 expression compared to induced Tregs (Thornton et al., 2010). Since Helios is preferentially expressed in natural Tregs, this suggests that the increase in the percentage of Tregs 69

after sargramostim treatment may be due to the expansion of natural Tregs and not inducing Tregs in the periphery.

In addition to genes associated with Th2 and Tregs, sargramostim also increases expression of anti-inflammatory genes such as PPARG, the gene for the nuclear receptor PPARγ. Agonists to PPARγ such as pioglitazone were successful at protcting neurons in preclinical studies (Laloux et al., 2012), but unsuccessful at reducing PD symptoms (Investigators., 2015). Sargramostim increases ICOS expression, a co- stimulatory molecule on T cells which promotes the release of Th2-related cytokines

(Simpson et al., 2010). The expression of the chemokine CCL7 is responsible for the recruitment of a variety of immune cells, especially monocytes (Cheng et al., 2014).

However, CCL7 expression appears to promote a Th2 immune response (Katzman and

Fowell, 2008). Expression of TNFSF11 (the gene for RANKL) promotes the survival of

DCs and Tregs (Wong et al., 1997; Loser et al., 2006). As such, the increase in

TNFSF11 may be promoting the increase by sargramostim-treated PD patients.

Interestingly, even though Tregs represent about 1% of the population of cells in this analysis, genes associated with Tregs and immunosuppression are noticeably increased. This suggests that the expression of these genes is very high in the Treg population, or these genes are expressed in cells that are beginning to upregulate CD25 which are not phenotypically Tregs. FOXP3 expression was tested in this analysis and was generally not increased noticeably, further supporting the interpretation that few cells in this analysis are Tregs or expressed genes associated with terminal Treg differentiation. Combined, these data show that sargramostim increases the expression of anti-inflammatory genes that may be playing a role in the increased Treg numbers and/or functions. 70

Several genes were upregulated in sargramostim-treated patients relative to baseline and relative to the placebo control involved in helper T cell proliferation.

Notable genes associated with proliferation include IL2 and IL2RA, which are the major cytokine and receptor for T cell proliferation (Olejniczak and Kasprzak, 2008). One of the genes with the highest gene expression is GATA4, which is not highly transcribed in

T cells (Caramori et al., 2001). GATA4 does regulate the transcription of IL-5

(Yamagata et al., 1995; Yamagata et al., 1997), which was also upregulated by sargramostim, suggesting that the increase in GATA4 expression promotes increased

IL5 expression. TNFRSF9 (the gene for CD137 (4-1BB), a cell surface protein which is upregulated upon T cell activation) is also upregulated by sargramostim treatment (Vinay and Kwon, 1998; Myers and Vella, 2005). Both HOXA3 and HOXA10 are genes for transcription factors involved in the proper development of lymphocytes and other immune cells (Thorsteinsdottir et al., 1997; Su and Manley, 2000; Su and Manley, 2002).

Interestingly, PERP, a gene involved in p53-mediated apoptosis (Ihrie et al., 2003) is also increased by sargramostim treatment. This suggests that T cell activation and possibly apoptosis are increased.

To our surprise, sargramostim treatment increased the expression of proinflammatory genes. We found that there were genes involved in both Th1 and Th17 phenotypes that were upregulated. Th1-related genes which were increased include

IL12B, IL18 and receptors all of which promote and are released from Th1 CD4+ T cells

(Zhu and Paul, 2008). However, sargramostim also decreased expression of several factors also involved in Th1 differentiation such as TBX21 (gene for t-bet, the master

Th1 transcription factor) and EOMES (another Th1 transcription factor) (Zhu and Paul,

2008). Sargramostim also increased several genes associated with Th17 cells including

IL21, IL17A, IL17RE, and RORC (Zhu and Paul, 2008). Genes for Toll-like receptors 71

(TLR) 4 and 6 were upregulated. Notably, α-synuclein binds to TLR4 and activates astrocytes and microglia (Rannikko et al., 2015) and TLR4 may promote α-synuclein clearance (Stefanova et al., 2011). Clearly this increase in proinflammatory gene expression is not indicative of Treg-mediated suppression or other anti-inflammatory mediators. Alternatively, it suggests that sargramostim not only increases Tregs and anti-inflammatory genes; but rather induces the expression of many genes expressed by several different T cell subsets.

From the above data, the gene changes reported were obtained by combining all data from all patients within a treatment group. Additionally, we tested if these combined gene changes are indicative of gene changes in each patient. While there were differences among patients, trends from the combined data are present within each patient. From this analysis, in the patients for which a complete data set exists, sargramostim treatment increased percentage of Tregs in whole blood, but Teffs were not changed (Gendelman et al., 2017). This trial did not specifically assess gene expression by Th1, Th2, and Th17 cells, so it is unclear to what extent the percentages of these helper T cell subsets are affected by sargramostim is unclear. To test the relationship between gene expression in the CD4+ CD25- T cells and increased percentages of Tregs in whole blood, we graphed the expression of genes with the highest upregulation compared to Treg percentage. Patients LJN and NLP displayed the highest level of gene expression and the highest percentage of Tregs. In both patients, lower Treg percentage and lower gene expression at visit 8 followed treatment cessation. Placebo-treated patients BED and MKO tended to display the lowest Treg response and lower gene expression. The differences in gene expression and percentage of Tregs suggest that the responses are different between patients. These data suggest that the effects of Tregs are consistent with the increase in gene 72

expression, including the proinflammatory cytokines. Because the binding of GM-CSF to its receptor can activate several signal cascades (Shearer, 2003; Hercus et al., 2009), it appears that this activation increases the expression of many genes of different functions. Further work will be needed to determine which of these pathways are important for increasing the expression of these genes and increase Treg levels.

One of the surprising results is that the CD3/CD28-stimulated T cells from sargramostim-treated patients display decreased expression of many genes from all 3 groups compared to baseline or placebo control patients. It is unclear why this is the case, but several possibilities exist. It could be that sargramostim treatment is activating gene expression in T cells, and CD3/CD28 stimulation cannot increase expression further. As a result, the difference in gene expression is decreased compared to unstimulated cells. It may also be the case that sargramostim treatment may be altering the kinetics of CD3/CD28 stimulation. Since we tested gene expression at a single time point, it is possible that gene expression peaked earlier or later, so the relative change in expression here may not be indicative of altered kinetics. A final possibility is that sargramostim or the existing Tregs population dampens the ability of the T cells to be activated. The result was that after sargramostim treatment, T cells did not increase the expression of all genes compared to T cells from placebo-treated PD patients. The isolated CD4+ T cells from both the placebo and sargramostim-treated PD patients display similar Treg numbers that are relatively small, so it would not be thought that these Tregs are influencing the ability of the other helper T cells to activate after stimulation with CD3/CD28 beads. However, since sargramostim-treated PD patients display more functional Tregs compared to placebo control (Gendelman et al., 2017), there may be more Treg suppression when stimulated with CD3/CD28 beads. It is also possible that sargramostim has some suppressive effect apart from activated Tregs, but 73

this has not been described. The bottom line is that more work is needed to test how sargramostim is affecting T cells and T cell activation both in culture and in vivo.

In conclusion, sargramostim treatment increased the expression of proinflammatory, anti-inflammatory and non-associated genes in unstimulated CD4+

CD25- T cells. Interestingly, sargramostim treatment decreases the expression of these genes after CD3/CD28 stimulation. Increases in gene expression were associated with an increased percentage of Tregs, suggesting that the patient-to-patient differences are due to different responses to sargramostim treatment. These results suggest that sargramostim treatment does not suppress proinflammatory gene expression in CD4+

CD25- T cells, which would be expected from the increase in anti-inflammatory genes and the percentages of Tregs.

74

CHAPTER THREE

GM-CSF-GENERATED BONE MARROW DERIVED DCs INDUCE

REGULATORY T CELLS AND ARE NEUROPROTECTIVE IN MPTP

INTOXICATED NICE

ABSTRACT

As shown in Chapter 2, GM-CSF increases the percentage and the function of

Tregs in Parkinson’s disease and the MPTP model. However, the mechanism by which

GM-CSF increases the percentage and function of Tregs is unclear. Models of autoimmunity demonstrated that GM-CSF induces a tolerogenic state in DCs which promotes Tregs leading to suppression of autoimmunity and inflammation. Herein, I tested if GM-CSF promotes and maintains a tolerogenic state in bone marrow-derived dendritic cells (BMDCs) stimulated with nitrated α-synuclein (N-α-Syn) as determined by flow cytometry, gene expression, cytokine release, and ability to induce Tregs. In addition, the ability of tolerogenic DCs to protect dopaminergic neurons, decrease neuroinflammation, and increase the percentage of Tregs in the spleen, was tested. I found that GM-CSF was unable to maintain a tolerogenic state in DCs after stimulation, but there was an alteration of how the BMDCs respond to stimulation. The adoptive transfer of BMDCs did protect tyrosine hydroxylase-positive neurons, decreased neuroinflammation, and increased the percentage of Tregs in the 1-methyl-4-phenyl-

1,2,3,6-tetrahydropyridine (MPTP) model. These results suggest that tolerogenic DCs can be protective, in part by increasing the percentage of Tregs. BMDC supernatants were unable to directly protect MES23.5 cells, but did decrease the expression of 75

proinflammatory mediators from cultured microglia, suggesting tolerogenic DCs may possess functions in vivo in addition to inducing Tregs.

INTRODUCTION

As stated previously, Parkinson’s disease (PD) patients exhibit a decreased percentage of Tregs in peripheral blood (Chen et al., 2015) and function (Hutter-

Saunders et al., 2012). Administering GM-CSF increased the percentage of Tregs in the

MPTP model which is associated with neuroprotection and decreased neuroinflammation (Kosloski et al., 2013) and sargramostim (human recombinant GM-

CSF) increased the percentage of Tregs in PD patients and improved Treg function while improving motor symptoms (Gendelman et al., 2017). Interestingly, T cells are not thought to express the receptor for GM-CSF, CSF2R, based on 2 lines of evidence.

Treating T cells with GM-CSF induces proliferation to a decreased extent compared to

IL-2 and could not support the survival of T cells (Santoli et al., 1988). Also, by flow cytometry, CD3+ T cells lack surface expression of the alpha chain of the CSF2R

(Rosas et al., 2007). However, in a single report, the GM-CSF receptor is expressed on the surface of CD4+ CD25+ T cells and GM-CSF promotes the proliferation of Tregs

(CD4+ CD25+ Foxp3+) (Kared et al., 2008). These data suggest that GM-CSF is unable to induce the formation of Tregs from the non-Treg T cell population, but may promote the proliferation of existing Tregs.

The ability of GM-CSF to induce Tregs and suppress inflammation is curious given that the main role of GM-CSF is to mobilize myeloid cells, including macrophages/monocytes, neutrophils and other granulocytes from the bone marrow to the periphery (Ushach and Zlotnik, 2016). As a result, GM-CSF has been thought of as 76

a proinflammatory cytokine. GM-CSF and sargramostim have been used in clinical trials as an adjuvant for the hepatitis B vaccine (Overton et al., 2010) and several cancers including renal cancer, melanoma, prostate cancer, lung cancer, and pancreatic cancer

(Parmiani et al., 2007; Waller, 2007). In addition, in the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis, GM-CSF may (King and Thomas,

2007) or may not (Pierson and Goverman, 2017) be required for the development of paralysis in the model and the GM-CSF-secreting CD4+ T cells induce EAE (Codarri et al., 2011). Because GM-CSF is not thought to exert its effects directly on CD4+ cells and because GM-CSF has proinflammatory effects by expanding granulocytes and monocytes, it is unclear how the percentage of Tregs is being increased.

Despite the known functions of GM-CSF in myeloid cells, a growing body of research has demonstrated GM-CSF can exert beneficial effects in models of autoimmunity and inflammation. For example, administering GM-CSF suppresses the immune response to thyroglobulin in the model of autoimmune thyroiditis (Gangi et al.,

2005), skeletal muscle nicotinic acetylcholine receptor in the mouse model of myasthenia gravis (Sheng et al., 2008), and lymphocyte infiltration into the pancreas in the NOD model of type 1 diabetes (Cheatem et al., 2009). In addition, in models of inflammatory disease such as colitis (Sainathan et al., 2008) or traumatic brain injury

(Kelso et al., 2015), treatment with GM-CSF also decreases inflammation to reduce the symptoms of disease. As with the MPTP model (Kosloski et al., 2013), in many of these models GM-CSF increases the percentage of Tregs (Sheng et al., 2008; Cheatem et al.,

2009; Ganesh et al., 2009). From these data, the prevailing thought is that GM-CSF increases Tregs by acting on DCs inducing a tolerogenic state (Gaudreau et al., 2007;

Bhattacharya et al., 2011). The tolerogenic DCs express OX40L and Jagged-1 (Jag-1) which bind to OX40 and Notch-3 respectively on T cells and are important for the 77

increase in the percentage of Tregs in culture (Gopisetty et al., 2013; Haddad et al.,

2016). Tolerogenic DCs promote the proliferation of Tregs and induce naïve CD4+ T cells to become Tregs (Bhattacharya et al., 2011). The resulting Tregs induced by tolerogenic DCs may be antigen specific or antigen independent. In either case, increases in Tregs reduces the inflammatory immune response and reduces the symptoms of disease.

Like immature DCs, tolerogenic DCs display low expression of co-stimulatory molecules like MHC II, CD40, CD80, CD86, among others, and low expression of cytokines (Lutz and Schuler, 2002; Steinman et al., 2003; Rutella et al., 2006; Liu and

Cao, 2015). What cytokines are expressed are commonly anti-inflammatory, such as IL-

10. Tolerogenic DCs also display increased expression IDO, an enzyme which metabolizes with tryptophan to kynurenine, and the expression of IDO and increased release of kynurenine are associated with induction of Tregs (Rutella et al., 2006).

Unlike immature DCs, tolerogenic dendritic cells should maintain this low expression even in the presence of a maturation stimulation (Mahnke et al., 2002). Maintenance of a tolerogenic state will make DCs less able to activate naïve CD4+ cells to effector helper T cells by reducing all 3 T cell maturation signals, but maintains the ability to induce Tregs (Lutz and Schuler, 2002; Mahnke et al., 2002).

In the study described in this chapter, I hypothesized that GM-CSF promotes and maintains a tolerogenic state in DCs which is the mechanism by which GM-CSF is neuroprotective. We tested this in 2 DC lines (DC2.4 and DC3.2) and in BMDCs. We also sought to determine if adoptive transfer of tolerogenic DCs are able to protect dopaminergic neurons, reduce neuroinflammation, and increase the percentage of Tregs in the spleen after MPTP, similar to GM-CSF administration. I also tested if BMDC supernatant protects the MES23.5 dopaminergic neuron cell line and decreases the 78

expression of proinflammatory mediators from activated BV2 microglia cell lines. Lastly,

I tested if GM-CSF increases the number or tolerogenic state of splenic DCs, the DCs likely to be inducing Tregs in the spleen.

METHODS

Animals

Male 6-18-week-old C57BL6/J mice (Jackson labs, Bar Harbor, ME) were used in all experiments described below. All procedures were performed in accordance with the National Institutes of Health guidelines and all procedures were approved by the

Institutional Animal Care and Use Committee of the University of Nebraska Medical

Center.

DC2.4 and DC3.2 cell culture

DC2.4 and DC3.2 cells were generously provided by the laboratory of Dr.

Kenneth Rock (University of Massachusetts Medical Center, Worcester, MA). Both cell lines were cultured in HCM-10% media (RPMI 1640 + L-glutamine (Gibco), 10% FBS

(Atlanta Biologicals, Flowery Branch, GA), 1x MEM nonessential amino acids (Hyclone),

100 U penicillin, 100 g/ml Streptomycin (Gibco), 10 mM HEPES (Gibco), 2 mM L- glutamine (Gibco), and 55 nM 2-mercaptoethanol (Gibco)). Both cell lines were passed by drawing off the whole volume of old media into a 15 ml conical tube (BD Falcon,

Corning, NY) and pipetting 3 ml of 0.25% trypsin containing EDTA (Hyclone) per T-75 flask. Flasks were incubated at 37°C for 2 min to remove adhered cells from the bottom.

The flask was tapped to dislodge cells and the whole volume was pipetted into the 15 ml conical tube to dilute the trypsin and cells were concentrated by centrifugation at 400 xg 79

for 5 min. The supernatant was discarded and the cell pellet was resuspended in 1 ml

HCM-10% media, and an aliquot was diluted 1:10 in trypan blue (Sigma) for cell counting. A volume of cell suspension corresponding to ~2x106 cells was pipetted into a

T-75 flask (BD Falcon) in 10 ml HCM-10% media. The flasks were placed in an incubator at 37°C, 5% CO2 incubator. Cells were passed whenever they reached confluence, which was about every 3 days.

BMDC differentiation

BMDCs were differentiated from bone marrow according to standard protocols

(Lutz et al., 2000; Bhattacharya et al., 2011). The methods for differentiating and stimulating BMDCs are outlined in Figure 3.1. Briefly, male 6-18-week-old C57BL6/J mice (Jackson Labs) were sacrificed by CO2 asphyxiation and cervical dislocation and both femurs were removed and washed twice in ice cold 10 ml 1x Hanks Buffered Salt

Solution (HBSS, Gibco). Each femur was flushed with 5 ml 1x HBSS and the cells were collected into a petri dish. The cell suspension was drawn up through a 20 Ga. needle into a 50 ml conical tube (BD Falcon) and cells were concentrated by centrifugation at

200 xg for 10 min. The supernatant was discarded and the pellet was resuspended in 1 ml ACK lysis buffer (Gibco) per mouse. The red blood cells were lysed by incubation at

37°C for 2 min. Lysis was stopped by adding 1x HBSS and the cells were passed through a 70 μM cell strainer and the flow through was collected in a 50 ml conical tube.

Cells were concentrated by centrifugation at 200 xg for 10 min. The supernatant was discarded and the cell pellet was resuspended in R10 media (RPMI 1640 + L-glutamine

(Gibco), 10% heat-inactivated FBS (Atlanta Biologicals), 100 U penicillin, 100 g/ml streptomycin (Gibco), 10 mM HEPES (Gibco), 2 mM L-glutamine (Gibco), and 55 nM 2- 80

Figure 3.1 Method for BMDC generation

81

mercaptoethanol (Gibco)). A cell suspension was diluted 1:10 in trypan blue for counting. Cells were brought to a concentration so that each well of a 6-well plate (BD

Falcon) contained 4x106 cells and 20 ng/ml mouse recombinant GM-CSF (Peprotech,

Rocky Hill, NJ) in 4 ml of R10 media. For the experiment testing optimal conditions for

BMDC maturation, wells contained 10 ng/ml GM-CSF and some wells contained 10 ng/ml IL-4 (Peprotech, Rocky Hill, NJ). BMDCs were incubated for 4 days in the 37°C,

5% CO2 incubator.

Four and 6 days after isolating BMDCs the media was changed as follows. The

6-well plate was removed from the incubator and the volume of media was removed and pipetted into 50 ml conical tubes and cells were concentrated by centrifugation at 200 xg for 10 min. Two ml R10 media was added to each well. The supernatant was discarded and the cell pellet was resuspended in R10 media supplemented with 40 ng/ml mouse recombinant GM-CSF so the total volume was brought to 4 ml, and the plate was placed back in the 37°C, 5% CO2 incubator.

On the eighth day after isolating bone marrow, the BMDCs had differentiated and could be treated as follows. The 6-well plates were removed from the incubator and the whole volume was pipetted from each well into a 50 ml conical tube and 1 ml of R10 media was added to each well. Cells were concentrated by centrifugation at 200 xg for

10 min. The supernatant was discarded and the cell pellet was resuspended in R10 media supplemented with 40 ng/ml mouse recombinant GM-CSF or R10 media alone depending on experiment. One ml of the cell suspension was added to each well and the 6-well plate was placed back in the 37°C, 5% CO2 incubator for 2 days.

On day 10 after starting BMDC culture, stimulation was done as follows. BMDCs were stimulated with 10, 30 or 100 μg/ml of nitrated recombinant mouse α-Syn (N-α-

Syn), which was made as described previously (Benner et al., 2008). BMDCs were also 82

stimulated with 0.1, 10, 30 and 100 μg/ml E. coli O55:B5 LPS (Sigma). To test gene expression, BMDCs were stimulated for 6 hr, a time frame in which there should be maximal expression of proinflammatory cytokine genes (Crabtree and Durand, 1986;

Crabtree, 1989). For all other experiments, BMDCs for stimulated for 24 hr. At that time, BMDCs were removed for flow cytometry or to produce lysate for Western blot or the conditioned media was removed for cytokine detection, nitrite release detection, and kynurenine detection.

Griess assay

The Griess assay was performed 2 different ways according to the manufacturer’s protocol (Promega Fisher Waltham, MA). The assays for DC2.4 and

DC3.2 cells in Figure 3.2 were performed using the Promega kit as follows. Standards were diluted in HCM-10% media and were serially diluted to create a standard curve.

Wells of a flat-bottom 96-well plate (BD Falcon) contained 50 μl of supernatant added.

All wells contained 50 μl sulfanilamide added and the plate was incubated wrapped in foil at RT for 5 min. All wells contain 50 μl NED solution added and the plate was incubated at RT for 5 min wrapped in foil. The plate was read at 540 nm.

The Griess assays in Figure 3.5, 3.8, and 3.23 were performed using the

ThermoFisher (Waltham, MA) kit as follows. Nitrite standard was diluted in R10 or BV2 media and serially diluted in R10 or BV2 media to generate a standard curve. In addition, the 96-well plate contained 150 μl of BMDCs or BV2 cells supernatant added.

To detect nitrite, 130μl MQ water, and 20 μl of the mixed Griess assay reagents added to all wells. The 96-well plate was incubated at RT for 30 min wrapped in foil before the absorbance was read at 540 nm. 83

For both assays, the absorbance was background corrected by subtracting the average reading from wells containing fresh media, not that from cells, from all other wells. Based on the standard curve the concentration of nitrite was calculated from the absorbance.

AlamarBlue cell viability assay

For DC2.4 and DC3.2 cells, the alamarBlue cell viability was performed by plating cells at 1x105 cells per well into a U-bottom 96-well plate with 20 ng/ml mouse recombinant GM-CSF added to the appropriate wells for 2 days. To the appropriate wells, 10 μg/ml E. coli O55:B5 LPS (Sigma) was added 24 hr before reading to stimulate cells. After 24 hr of treatment, 15 μl alamarBlue solution (ThermoFisher) was added to all wells and after 4 hours of incubation the absorbance was read at 570 and 600 nm.

For BMDCs, cells were treated in 6-well plates with R10 media alone or with R10 media supplemented with 20 ng/ml mouse recombinant GM-CSF for 2 days prior to stimulation with 10 μg/ml E. coli O55:B5 LPS (Sigma) or 30 μg/ml recombinant N-α-Syn for 24 hr as described above. At this time, BMDCs were removed from the 6-well plate using a cell scraper and 1x105 BMDCs was plated into each well of a U-bottom 96-well plate and 15 μl alamarBlue was added to each well. After 4 hrs, the absorbance was read at 570 and 600 nm.

For both methods, the percentage of alamarBlue dye which was reduced (which can only happen in viable cells with functioning mitochondria) was calculated according to the manufacturer’s directions.

Flow cytometry

DC2.4, DC3.2 and BMDCs were stained for flow cytometric analysis as follows.

After removing cells from the dish by either trypsin-EDTA or cell scraper for DC2.4 and 84

DC3.2 cells or BMDCs respectively, cells were counted and ~5x105 cells were added to each 5 ml flow stain tube (BD Falcon) for staining. The Fc receptors were blocked in a solution of 10 μg/ml rat gamma globulin FSB (1x DPBS, (Gibco) supplemented with

0.5% BSA and 0.1% sodium azide). After blocking for 30-60 min on ice, anti-CD11c- alexaFluor 488, anti-CD11b-PECy7, anti-Jagged-1-PE, anti-OX40L-APC, anti-MHC II- alexaFluor 700, anti-CD86-eFluor450, anti-CD39-PerCP-eFluor 710 (eBioscience, Santa

Clara, CA) and anti-CD73-APC-vio 770 (Miltenyi Biotec) antibodies stained cells for 30 min at 4°C. Cells were washed 2 times in 2 ml FSB and cells were concentrated by centrifugation at 400 xg for 5 min. Cells were fixed in a solution of 1% formaldehyde in

1x DPBS for 15 min at RT. Cells were concentrated by centrifugation at 400 xg for 5 min and the pellet was resuspended in FSB for analysis using a BD LSR II flow cytometer

(Franklin Lakes, NJ) at the University of Nebraska Medical Center Flow Cytometry

Research Facility. Single cells were gated from the total event population and the percentage of total cells or CD11c+ cells expressing each marker was determined. The mean fluorescent intensity (MFI) of each marker in the CD11c+ cells was also determined.

For testing the percentage of Tregs in total splenocytes, non-adhered cells from

CD4+/BMDC co-culture, or cells enriched after Miltenyi CD4+ CD25+ kit, ~5x105 cells were blocked in a solution of 10 μg/ml rat gamma globulin in FSB for 30-60 min. Cells were stained with anti-CD4-PECy7 and anti-CD25-PE for 30 min at 4°C. Cells were washed twice in 2 ml FSB and concentrated by centrifugation at 400 xg for 5 min. Cells were then fixed and permeablized for 1 hr at 4°C using the buffer from the

Foxp3/transcription factor staining buffer set (eBioscience). Cells were washed in 2 ml

1x permeablization buffer and concentrated by centrifugation at 400 xg for 5 min. Cells were stained with anti-Foxp3-APC for 30 min at 4°C. Cells were washed twice by the 85

addition of 2 ml of 1x permeabization buffer followed by concentration by centrifugation at 400 xg for 5 min. Cells were resuspended in FSB prior to analysis using the BD LSR

II at the University of Nebraska Medical Center Flow Cytometry Research Facility.

Within the single cell population, the percentages of CD4+ T cells was determined. The percentage of CD4+ T cells double positive for CD25+ Foxp3+ was reported as the percentage of Tregs within the CD4+ T cell population.

Luminex array

The release of cytokines and chemokines was determined from BMDCs cultured for 2 days in R10 media alone or R10 media supplemented with 20 ng/ml GM-CSF prior to no stimulation or 1 day stimulation with 30 μg/ml N-α-Syn. The cell culture supernatant was clarified at 10,000 xg for 5 min and the resulting supernatant was transferred to a 1.5 ml snap-cap tube and stored at -80°C until analysis could be performed. Concentration of cytokines and chemokines was determined with Luminex xMAP mouse cytokine and chemokines magnetic bead panel (Milipore, Billerica, MA) according to the manufacturer’s protocol. Briefly, 25 μl of supernatant was added to the provided 96-well plate. Standards and controls were prepared and diluted according to the protocol and 25 μl was added to the appropriate well. Assay buffer was added to supernatant samples and R10 media was added to all controls and standards. A master mix containing antibody-coated magnetic beads against IFNγ, IL-1α, IL-1β, IL-2, IL-4, IL-

5, IL-6, IL-7, IL-9, IL-10, IL-12p40, IL-12p70, IL-13, Lix, IL-15, IL-17, IP-10, MIP-2, MIG,

RANTES, and TNF-α was added to all wells and the 96-well plate was incubated at 4°C overnight. Magnetic beads were concentrated and washed on a magnet. Detection antibodies were added to each well and the samples were incubated at RT for 60 min.

Strepavidin-PE was added to all wells and the 96-well plate was incubated at RT for 30 min. The beads were concentrated and washed in a magnet as above and resupended 86

in sheath solution. All wells were analyzed using the Milipore Magpix system and

Luminex Xponent 4.2 software. A standard curve was generated for each of the cytokines and chemokines and based on this standard curve the concentration of each cytokine and chemokine was calculated. For this experiment, for each of the 4 treatment groups, supernatants from 7 different wells containing BMDCs were used. Supernatant from each of the 7 wells of supernatant were run in duplicate or triplicate.

RNA isolation, cDNA conversion and PCR array

To test the expression of proinflammatory genes, BMDCs were cultured in R10 media alone or R10 media supplemented with 20 ng/ml GM-CSF for 2 days prior to stimulation of cells with 30 μg/ml N-α-Syn, or BMDCs were left unstimulated with for 6 hr. BMDCs were removed from the 6-well dish by a cell scraper and cells were washed two times in 1x DPBS (Gibco). There were 6 separate wells for BMDCs for each of four treatment groups. Cells were transferred to a 1.5 ml snap-cap tube and cells were frozen at -80°C until RNA isolation. The RNeasy mini kit (Qiagen) was used to isolate total RNA from all samples as follows. The cell pellet was resuspended in RLT buffer supplemented with 2-mercaptoethanol and homogenized by trituration through a 27 Ga. needle 10 times. The lysate was clarified by concentration at 20,800 xg for 3 min.

To test the expression of proinflammatory genes in the midbrain, BMDCs were transferred into mice via i.v. injection at 14 and 7 days prior to MPTP intoxication. As controls, mice were injected with DPBS and intoxicated with MPTP. Two days after

MPTP intoxication, mice were sacrificed by CO2 asphyxiation and cervical dislocation.

Brains were removed, hemisected and the midbrain was placed in a 1.5 ml tube containing 1 ml RNAlater (ThermoFisher). Brains were incubated at 4°C overnight before being weighed and frozen at -80°C. To isolate RNA, the midbrains were thawed, homogenized in 350 µl β-mercaptoethanol-supplemented RLT buffer (Qiagen, Valencia, 87

CA) for every 30 mg tissue. Tissues were sequentially homogenized by trituration through 18, 20 and 27 Ga needles. The tissue homogenate was clarified by concentration at 20,800 xg for 3 min.

The clarified supernatants from both cell pellets and from tissue were transferred to a new 1.5 ml tube containing 70% ethanol in water. This mixture was pipetted into a filter and the nucleic acids were concentrated on the filter by centrifugation at 10,000 xg for 20 sec. The flow through was discarded. The filter was washed in RW1 at 10,000 xg for 20 sec. DNase I from the RNase-free DNase kit (Qiagen) was diluted in RDD buffer, pipetted on the membrane, and incubated at RT for 15 min. The filter was washed in

RW1 then 2 times in RPE buffer by centrifugation at 10,000 xg for 20 sec. The filter was eluted by the addition of 50 μl RNase-free water and centrifugation at 10,000 xg for 1 min. The flow through was added back to the filter and was eluted a second time at

10,000 xg for 1min to increase RNA yield. The concentration of RNA, as well as the

260/280 and 260/230 ratios were determined by UV spectrophotometry using a Thermo

ND-100 Nanodrop spectrophotometer.

Copying of mRNA to cDNA was achieved using the RevertAID first strand cDNA synthesis kit (ThermoFisher) according to the manufacturer’s protocol. Briefly, 500 ng of

RNA was brought to 11 μl by the addition of nuclease-free water and 1 μl oligo dT primers was added to each sample and heated at 65°C for 5 min. After breaking the secondary structure of RNA, a master mix containing 5x reaction buffer, Ribolock Rnase inhibitor, 10 mM dNTPs, and RevertAID M-MuLV reverse transcriptase was added to all samples and incubated at 42°C for 1 hr. After the production of cDNA from the mRNA template, the reaction was terminated by 5 min incubation at 70°C. All samples were brought to 100 μl by the addition of nuclease-free water and samples were frozen at -

20°C until PCR could be performed. 88

Each cDNA sample was added to RNase-free and DNase-free water (Invitrogen),

2x RT2 SYBR green master mix (Qiagen), and 25 μl of this solution was added to each well of a mouse inflammatory response and autoimmunity array (Qiagen), and PCR was performed using an Eppendorf Realplex 2S Master Cycler (Eppendorf, Hamburg,

Germany). The PCR program used for this array which was as follows: hot start at 95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 1 min. After 40 cycles, a melting curve was performed, Ct values were determined, and fold changes determinedusinng the ΔΔCt method using RT2 Profiler PCR array data analysis version

3.5. Changes were normalized relative to media-cultured, unstimulated BMDCs or from the PBS control midbrains.

IDO expression by Western blot

To determine the relative expression of IDO relative to β-actin, Western blots were run as follows. BMDCs were pretreated for 2 days in R10 media alone or R10 media supplemented with 20 ng/ml GM-CSF and either unstimulated or 30 μg/ml N-α-

Syn for 1 day. These BMDCs were removed from the dish by cell scraper, washed in 1x

DPBS and concentrated into a 1.5ml snap-cap tube which was frozen at -80°C until analysis could be performed. The cell pellets were thawed and lysed in 250 μl RIPA buffer (ThermoFisher) for 5 min on ice. After mixing by vortexing, the lysate was clarified by centrifugation at 14,000 xg for 15 min at 4°C. The supernatant was transferred to a new 1.5ml tube.

Protein concentration of the clarified lysate was determined by BCA assay

(Pierce) as described by the manufacturer. Briefly, BSA standards (ThermoFisher) were pipetted to a flat-bottom 96-well plate to make a standard curve and 25 μl of each sample was added to the same 96-well plate. The BCA working solution was prepared and 200 μl added to each well and the 96-well plate was incubated at 37°C for 30 min. 89

The plate was allowed to equilibrate to RT for 10 min and read at 562 nm. The absorbance reading was background corrected by subtracting the absorbance of the control not containing protein from all other wells. Based on the standard curve, the protein concentration for each sample was calculated.

To perform Western blots, a volume of lysate corresponding to 10 μg of total protein was brought to 30 μl MQ water and 30 μl 2x sample buffer (Bio-Rad, Hercules,

CA) supplemented with 2-mercaptoethanol. Proteins were denatured by heating at 95°C for 5 min. The whole volume was loaded into a pre-cast ExpressPlus PAGE 4-20% gel

(Genscript, Piscataway, NJ) and gels electrophoresed for 85 min until the dye front was near, but not at the foot of the gel. Proteins were transferred to PVDF membranes (Bio-

Rad) using the Bio-Rad wet-transfer apparatus in 1x Bio-Rad transfer buffer for 90 min.

The membrane was blocked in 5% dried milk in PBS supplemented with 0.1% Tween 20

(PBST) at RT for 60 min. Membranes were incubated with a primary antibody solution containing mouse anti-mouse IDO antibody (Milipore) in blocking buffer overnight at 4°C.

Membranes were washed for 30 min in PBS changing the PBS every 5 min prior to incubation in goat-anti-mouse HRP-conjugated secondary antibody solution for 30 min at

RT. Membranes were washed in PBST for 30 min as above. The membrane was imaged with ECL femto solution (ThermoFisher) and images were taken with a chemiluminescent camera (ThermoFisher). The ECL femto solution was washed off in

PBST for 30 min. The membrane was stripped using Restore Western blot stripping buffer (ThermoFisher), washed in PBST, and blocked in 5% dried milk blocking buffer for

30 min. The membrane was incubated in rabbit anti-mouse β-actin (Santa Cruz, Dallas,

TX) overnight at 4°C, washed in PBST as above, incubated in donkey anti-rabbit HRP- conjugated secondary antibody (Santa Cruz) for 30 min at RT, washed again in PBST for 30 min prior to development with ECL Pico solution (ThermoFisher), and the imaged 90

with a chemiluminescent camera. From the images, ImageJ was used to determine the intensity of the IDO and β-actin and the IDO/ β-actin ratio was determined.

Kynurenine detection and quantification

The release of kynurenine was determined from BMDCs cultured for 2 days in

R10 media alone or R10 media supplemented with 20 ng/ml GM-CSF prior to no stimulation or stimulation with 30 μg/ml N-α-Syn for 1 hr. Cell culture supernatant was clarified at 10,000 xg for 5 min, the resulting supernatant was transferred to a 1.5 ml snap-cap tube, and stored at -80°C until analysis. For analysis, to each of the 100 μl of supernatants, 10 μl of 10 μg/ml tryptophan HCl (Sigma Aldrich) was added to each sample as an internal standard. A kynurenine standard curve was made by 2-fold serially dilutions 50 μg/ml kynurenine (Sigma Aldrich) in acetonitrile (Fisher) down to acetonitrile alone. Each standard also had 10 μl of 10 μg/ml tryptophan HCl added as an internal standard. Ice-cold acetonitrile was added to each sample and standard and mixed by vortexing for 3 min. Samples and standards were clarified by centrifugation at

14,000 RPM for 10 min, and 1 ml of supernatant was pipetted into a new 2 ml snap-cap tube, and samples and standards were concentrated by speed vacuum for 3 hr, and pellets were stored at -20°C overnight. All samples and standards were resuspended in

100 μl 0.1% formic acid in HPLC-grade water (Fisher), vortexed briefly, concentrated by centrifugation at 14,000 RPM for 15 min, and 30 μl of each supernatant from each standard/sample was pipetted into a 96-well plate for mass spectrometry analysis on a

C8 Luna PFP column attached to a Waters TQS Micro mass spectrometer.

CD4+ and CD11c+ cell isolation and BMDC co-culture

The CD4+ T cells used in co-culture were isolated from the spleens of male

C57BL/6J mice (Jackson Laboratories) as follows. Mice were sacrificed by CO2 91

asphyxiation and cervical dislocation and spleens removed, and pressed through a 70

μM cell strainer into 10 ml 1x HBSS. This cell suspension was drawn up through a 20

Ga. needle into a 50 ml conical tube. Cells were concentrated by centrifugation at 200 xg for 10 min, the supernatant was discarded, and the pellet was resuspended in 1 ml

ACK lysis buffer (Gibco) per spleen. Red blood cells were lysed by incubation at 37°C for 2 min. Cells and ACK were diluted by adding 1x HBSS and cells concentrated by centrifugation at 200 xg for 10 min. The supernatant was discarded, the cell pellet was resuspended in Miltenyi CD4+ isolation buffer (0.5% BSA and 2 mM EDTA in 1x DPBS), and an aliquot was diluted for cell counting. CD4+ T cells were negatively selected using the Miltenyi CD4+ isolation kit according to the manufacture’s protocol. Briefly, splenocytes were brought to 1x107 splenocytes per 90 μl CD4+ isolation buffer, 10 μl of

CD4+ selection antibodies were added, and incubated at 4°C for 15 min. After incubation, 30 μl CD4+ isolation buffer and 20 μl microbeads were added for every

1x107 splenocytes and the mixture was incubated at 4°C for 10 min. The labeled splenocytes were pipetted into a Miltenyi LS column, which was placed on a magnetic rack. The antibody-bound, non-CD4+ cells adhered to the column and the CD4+ cells flowed through the column into a collection tube. The column was washed 3 times in

Miltenyi CD4+ isolation buffer and cells were also collected into the same tube. The

CD4+ cells that flowed through the column were concentrated by centrifugation at 200 xg for 10 min. The pellet was resuspended in complete media (RPMI 1640 media + L- glutamine (Gibco), 10% FBS (Sigma), 1 mM sodium pyruvate (Cellgro, Pittsburgh, PA),

1x MEM non-essential amino acids (Hyclone, Logan, UT), 100 U penicillin, 100 g/ml

Streptomycin (Gibco), 10 mM HEPES (Gibco), 2 mM L-glutamine (Gibco), and 55 nM 2- mercaptoethanol (Gibco)) and an aliquot was diluted 1:10 in trypan blue for cell counting.

Cell concentration was adjusted to 2x106 cells per ml of complete media. and an aliquot of CD4+-enriched cells was stained for Tregs to determine purity. 92

BMDCs were pretreated in R10 media alone or R10 media supplemented with 20 ng/ml GM-CSF for 2 days and either left unstimulated or stimulated with 30 μg/ml N-α-

Syn for 1 day. BMDCs were scraped from the dish, washed 2 times in 1x DPBS, resuspended in complete media, and an aliquot was diluted in trypan blue for cell counting. The concentration was adjusted to 1x106 BMDCs per ml of complete media.

Co-cultures were initiated so that 1 ml of BMDCs and 1 ml of CD4+ T cell-enriched splenocytes were pipetted into each well of a 24-well plate, then incubated at 37°C, 5%

CO2 for 5 days.

For transwell experiments, BMDCs and CD4+ cells were differentiated and isolated as above, however, CD4+ T cells were brought to 10x106 cells/ml. To a 24-well pate, 1 ml BMDCs (~1x106 BMDCs) was added to each well. The 3 μm transwell inset

(Corning, Corning, NY) was placed in the well and 100 μl CD4+ cells and 100 μl complete media was added into the insert. The 24-well plate was incubated at 37°C, 5%

CO2 for 5 days.

CD11c+ cells were isolated from the spleen using the Miltenyi CD11c isolation kit according to the manufacturer’s protocol. Single cell splenocytes were prepared and adjusted to 1x108 cells/400 μl Miltenyi CD4+ isolation buffer. For every 400 µl of splenocytes, 100 μl CD11c microbeads was added and incubated for 15 min at 4°C.

Cells were washed in Miltenyi CD4+ isolation buffer, concentrated by centrifugation at

200 xg for 10 min and added onto a Miltenyi LS column placed within a magnetic field and the column washed in Miltenyi CD4+ isolation buffer. The LS columns were removed from the magnet and CD11c+ cells were flushed from the column into a new collection tube and cells were concentrated by centrifugation at 200 xg for 10 min. After cell counting in trypan blue, CD11c+ cells were adjusted to ~1x106 CD11c+ cells/ml in complete media and 1 ml of CD11c+ cells were co-cultured with 1 ml of enriched 2x106 93

CD4+ T cells/ml into wells of a 24-well plate which was incubated at 37°C in 5% CO2 incubator for 5 days.

To determine the percentage of Tregs after co-culture, non-adherent cells were removed from the culture, washed twice in 1x DPBS (Gibco), and concentrated by centrifugation at 200 xg for 10 min. The cell pellet was resuspended in FSB, an aliquot was counted in trypan blue, and 2.5-5x105 cells were stained for Treg purity.

Treg functional assay

Treg function was tested for its ability to suppress the proliferation of the responder T cells (CD4+ CD25-) as described previously (Saunders et al., 2012; Olson et al., 2015). Briefly, C57BL/6J mice were sacrificed by CO2 asphyxiation and cervical dislocation. Splenocytes were isolated as described above and non-adhered cells were removed as described above from co-culture of CD4+ cells and BMDCs. The Miltenyi

CD4+ CD25+ isolation kit was used to enrich Tregs (CD4+ CD25+) and responder T cells (CD4+ CD25-) according to the manufacturer’s protocol. Briefly, for every 1x107 cells of splenocytes or non-adherent cells was added to 40 μl CD4+ isolation buffer and

10 μl CD4+ biotinylated antibodies and the suspension was incubated at 4°C for 5 min.

Then to every 1x107 cells was added 30 μl CD4+ isolation buffer, 20 μl Miltenyi anti- biotin magnetic beads, and 10 μl anti-CD25-PE antibody and were incubated at 4°C for

10 min. The labelled cells were passed through Miltenyi LD columns attached to magnetic racks and the filtrate (CD4+ T cells) collected. Cells were concentrated by centrifugation at 200 xg for 10 min, the pellet resuspended in 90 μl CD4+ isolation buffer,

10 μl anti-PE microbeads per 1x107 cells added and cell suspensions incubated at 4°C for 15 min. The cellular filtrates served as responder T cells. The MS columns were flushed by adding 1 ml CD4+ isolation buffer and the volume forced through the column using the plunger. These cells were enriched Tregs. Aliquots of both cells were counted 94

by trypan blue. Flow cytometry was performed on ~5x105 cells to confirm purity. Treg concentrations were adjusted to ~1x106 cells/ml by the addition of complete media and responder T cells were adjusted to 2x106 cells/ml in CFSE buffer (0.1% BSA in 1x

DPBS).

The responder T cells were labelled with carboxyfluorescein succinimidyl ester

(CFSE, ThermoFisher) according to the manufacturer’s protocol. Briefly, each vial of

CFSE was reconstituted with 18 μl dimethyl sulfoxide (DMSO, Sigma). For every 1 ml of responder T cells to be labelled, 3 μl of CFSE was added to 1 ml CFSE buffer, cells and buffer combined, and incubated for 15 min at 37 °C. To quench the reaction, 30 ml ice- cold complete media was added and the cell suspension was incubated on ice for 5 min.

Cells were concentrated by centrifugation at 300 xg for 5 min, the supernatant discarded, the cell pellet resuspended in complete media, and cells concentrated by centrifugation at 300 xg for 5 min. An aliquot was diluted in trypan blue for cell counting and responder T cells were adjusted to 5x105 responder T cells/ml in complete media.

To a U-bottom 96-well plate, 100 μl of Treg suspension and 100 μl complete media was added to a well and this volume was diluted by 2-fold serial dilutions into 5 wells to yield ~50,000, 25,000, 12,500, 6,250 or 3,125 Tregs per well. To the first 4 wells in the dilution was added to 50,000 CFSE-labelled responder T cells in 100 µl to yield

Treg:responder T cell ratios of 1, 0.5, 0.25, and 0.125:1. The fifth well is a Treg-only control. CD3/CD28 Transactivator beads (ThermoFisher) were used to stimulate the proliferation of responder T cells by washing 1.25 μl transactivator beads per well

(~50,000 beads) in CFSE buffer, adjusting the concentration of 10 µl per well in complete media, and pipetting 10 µl of beads to each well of the plate. The plate was incubated at 37°C in 5% CO2 for 3 days. The plate was concentrated by centrifugation 95

at 300 xg for 5 min, fixed in 100 µl 1% paraformaldehyde in 1x DPBS, and assessed by flow cytometric analysis.

Adoptive transfer, MPTP intoxication, mouse perfusion, and immunochemistry for mac-1 or tyrosine hydroxylase

To determine if BMDCs, CD4+ CD25+, and non-adherent cells from the co- culture of BMDCs and CD4+ T cells were neuroprotective after MPTP intoxication, the following protocol was used. BMDCs, CD4+ CD25+, and non-adherent cells from co- culture were isolated, and adoptively transferred via intravenous (i.v.) tail injection ~8-10 hr post MPTP intoxication. MPTP was handled and administered according to established protocols (Jackson-Lewis and Przedborski, 2007) and animal protocols approved by National institutes of Health and the University of Nebraska Medical Center

Institutional Animal Care and Use Committee. Briefly, mice were injected 4 times intraperitoneally at 2 hour intervals with 10 ml DPBS/kg as a control or 12 or 16 mg free base MPTP (Sigma)/kg. Seven days after MPTP intoxication, mice were sacrificed to assess surviving tyrosine hydroxylase (TH) immunohistochemistry as described below.

To determine whether BMDCs are neuroprotective if administered prior to MPTP intoxication, BMDCs were differentiated in R10 media supplemented with 20 ng/ml GM-

CSF for 8 days and then in R10 media alone for 3 days. At this time, BMDCs were removed from the dish by cell scraper and adjusted to ~1.5x106 BMDCs/250 μl 1x

DPBS. This volume was administered by tail vein to 8-9-week-old male C57BL/6J mice at 1 and 2 weeks prior to MPTP intoxication with 16 mg/kg of MPTP. This adoptive transfer protocol for BMDCs was described for the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis (Prado et al., 2012). 96

Two (for Mac-1 immunohistochemistry) or 7 (for TH immunohistochemistry) days after MPTP intoxication, mice were sacrificed, perfused, and brains removed and processed for immunohistochemistry as previously described (Kosloski et al., 2013;

Olson et al., 2015). Briefly, mice were terminally anesthetized with Fatal-Plus

(pentobarbital) (Vortech, Dearborn, MI) and mice were transcardially perfused with 1x

DPBS (Cellgro, Corning, NY) until all blood was flushed. At this time, the mouse was fixed with 4% paraformaldehyde in DPBS. The brain was removed and post-fixed in 4% paraformaldehyde for 24 hr. Brains were then incubated in 30% sucrose at 4°C to remove the formaldehyde and cryopreserve tissues prior to snap-freezing brains for 15 sec in 2-methyl butane (ThemoFisher) which has been chilled on dry ice for 20 min.

Brains were embedded in OCT (Tissue Tek, Radnor, PA) and 30 μm sections of the midbrain and striatum were cut and stored in 0.1% sodium azide (Sigma) in PBS at 4°C until immunohistochemistry was performed (Kosloski et al., 2013; Olson et al., 2015).

For Mac-1 staining, 6 free floating sections of substantia nigra from each mouse were washed in 1x PBS (Cellgro), endogenous peroxidases blocked by incubation in 3% hydrogen peroxide (ThermoFisher) and 10% methanol (ThemoFisher, in PBS), and nonspecific staining blocked in 5% normal rabbit serum in PBS (Vector labs, Burlingame,

CA). Sections were incubated at 4°C with 1:500 dilution of rat anti-Mac-1 primary antibody (Bio-Rad) in 2% normal rabbit serum (in PBS) overnight, washed, and incubated in a 1:500 dilution of biotinylated rabbit anti-rat antibody in 2% normal rabbit serum in PBS (Vector labs) followed by ABC biotin-avidin peroxidase solution (Vector

Labs) prior to color generation with 3,3'-diaminobenzidine (DAB). The number of reactive Mac-1+ microglia per total area was determined by stereological analysis using

Stereo Investigator software and the optical fractionator module (MBF Bioscience,

Williston, VT) (Kosloski et al., 2013; Olson et al., 2015). 97

For TH staining, 6 free-floating sections of the striatum and 12 free-floating sections of the substantia nigra from each mouse were washed in 0.1 M Tris-buffered saline (TBS) and endogenous peroxidases were blocked with 3% hydrogen peroxide in methanol and 0.1 M TBS. Sections were blocked in 5% normal goat serum (Vector

Laboratories, in 0.1 M TBS) in 0.1 M TBS prior to staining with 1:2,000 or 1:1,000 dilution of anti-tyrosine hydroxylase (TH) antibody (EMD/Milipore, Burlington, MA) in 2% normal goat serum in 0.1 M TBS for 48 hr at 4°C. Sections were then washed in 0.1 M

TBS, stained with a goat anti-rabbit secondary antibody (Vector Laboratories) in 2% normal goat serum in 0.1 M TBS, and washed again in 0.1 M TBS prior to incubation with avidin-biotin complex kit (Vector Laboratories) and development with DAB in PBS.

Sections were washed in 0.1 M TBS and transferred to slides. Midbrain sections containing the substantia nigra were Nissl counter-stained prior to counting TH+ Nissl+ and TH- Nissl+ neurons by stereological analysis using Stereo Investigator software with the optical fractionator module (MBF Bioscience, Williston, VT). TH density of dopaminergic termini in the striatum was determined using ImageJ (Kosloski et al., 2013;

Olson et al., 2015).

Adoptive transfer of BMDCs into mice with and without MPTP for Treg percentage and function

To determine if BMDCs induce Tregs in vivo, BMDCs were differentiated in R10 media supplemented with 20 ng/ml GM-CSF for 8 days and then in R10 media alone for

3 days. At this time, BMDCs were removed from the dish by cell scraper and brought to

~1.5x106 BMDCs/250 μl 1x DPBS. This volume was administered via tail vein injection to 8-9 week-old male C57BL/6J mice (Jackson Labs) 1 and 2 weeks prior to sacrifice to test Treg induction in the absence of MPTP or 2 days after MPTP intoxication which was done 1 week after the second BMDC adoptively transfer. As a control for Treg induction 98

without MPTP, mice received 50 µg/kg mouse recombinant GM-CSF i.p. injected on each of 5 days prior to sacrifice. For both experiments, mice were sacrificed by CO2 asphyxiation and cervical dislocation. Spleen and lymph nodes (cervical, brachial, axillary, and inguinal) were removed, kept separate for each mouse, and single cell suspensions prepared for each tissue. Flow cytometry was performed to test the percentage of Tregs in ~500,000 splenocytes or lymph node cells. The remaining splenocytes were pooled by treatment group, Tregs (CD4+ CD25+), and responder T cells (CD4+ CD25-) were isolated, and Treg suppression assays performed. Blood samples were collected by submandibular bleed using 5 mm GoldenRod lancets

(ThermoFisher) and blood was collected in BD microtainer EDTA-coated tubes 1 day prior to each adoptive transfer and sacrifice. At least 50 μl blood was collected per mouse and Treg frequencies were determined by flow cytometry.

MES23.5 cell culture

MES23.5 cells were cultured in MES23.5 media (DMEM/F12 with L-glutamine and 15mM HEPES supplemented (Gibco) and supplemented with 2% heat-inactivated

FBS (Sigma), 100 U penicillin, 100 g/ml Streptomycin (Gibco), 1x N-2 supplement

(Gibco)). The cells were grown on poly-D-lysine-coated T-75 flasks (BD Falcon). Flasks or 96-well plates (BD Falcon) were coated with 10 ml or 100 µl/well in 5 µg poly-d- lysine/ml at 37°C overnight followed by removing the volume of coating solution, washing with sterile MQ water, and allowing the flask or plate to dry. Cells were removed from the flask by incubation in 1 mM EDTA (Gibco) and ~1x106 MES23.5 cells were passed to a new flask with cells passed 2 times per week.

BV2 cell culture 99

BV2 cells were cultured in BV2 media (1x DMEM with glucose, glutamine, and phenol red (Gibco) supplemented with 5% FBS, 100 U penicillin, 100 g/ml streptomycin

(Gibco), 10 mM HEPES (Gibco), 2 mM L-glutamine (Gibco), and 55 nM 2- mercaptoethanol (Gibco)). BV2 cells were passed 2 times per week by removing the

BV2 cells from T-75 flasks (BD Falcon) using 1 mM EDTA (Gibco) and ~1x106 BV2 cells were passed to a new flask for passage into new T-75 flasks.

To stimulate BV2 cells, ~1x106 BV2 cells in 0.5 ml of volume was added to each well of a 24-well plate. Each well was adjusted to 1 ml total volume by adding BV2 media, BMDC supernatant (supernatants of BMDC culture from which cells were removed by centrifugation at 400 xg for 10 min), or R10 media. LPS stimulation was achieved by adding 100 ng of LPS from E. coli O55:B5 (Sigma) to each well.

Supernatants from each well were combined by treatment group and cells concentrated by centrifugation at 400 xg for 10 min.

Celltiter-Glo Cell Viability Assay

Cell viability was determined using the CellTiter-Glo assay (Promega) as follows.

Either ~20,000 or 10,000 MES23.5 cells (for MPP+ or BV2 supernatant respectively) was added to a black-walled 96-well assay plate which was coated with poly-D-lysine.

An equal volume (50 µl) of MES23.5 media, R10 media, or BMDC supernatant was added to all wells. As controls, equal volumes of media without cells were added to other wells. After 24 hours of incubation, an equal volume (100 µl) of either MES23.5 media containing 0 – 1,000 µM MPP+ (Sigma), or BV2 media or supernatant was added.

MES23.5 cells were incubated for 24 hr. Afterwards, the media was removed from each well and 100 µl CellTiter-Glo assay reagent (reconstituted as described in the kit) was added and the 96-well plate was incubated at RT for 2 min with shaking followed by equilibration at RT for 10 min prior to reading luminescence. The raw luminescence 100

values from the no cell control wells were subtracted from all other wells. The background corrected values were divided by the control values obtained from treatment with media or BV2 media to determine the percent viability.

Cytokine bead array

Concentrations of IL-2, IL-4, IL-6, IL-10, IL-17a, TNF, and IFNγ were determined using the Th1/Th2/Th17 cytokine bead array (BD) as follows. Standards from the kit were reconstituted according to the manufacturer’s protocol and 2-fold serially diluted into 5 ml flow cytometry tubes in duplicate. In triplicate 50 µl samples of BV2 supernatant were added to 5 ml flow cytometry tubes. Capture beads from all analytes were combined and 50 µl assay detection reagent were added to 50 µl standards or samples which were incubated at RT for 2 hr. All samples and standards were washed in the assay wash buffer and beads were concentrated by centrifugation at 200 xg for 5 min. The supernatant was discarded and pellet resuspended in 300 µl wash buffer and analyzed by flow cytometry. The mean fluorescent intensity was determined for each cytokine in all samples and standards. The log10-transformed mean fluorescent intensity was plotted against log10-transformed concentration of the standards. The curve was fitted by five parameter logistical regression to determine the standard curve. Based on standard curve, the concentration of each cytokine was determined from all samples.

Testing splenic DC percentage after GM-CSF administration

Approximately 11-week-old C57BL/6J mice (Jackson Labs) were i.p. injected with 50 µg/kg mouse recombinant GM-CSF or a commensurate volume of 1x DPBS each of five days prior to sacrifice. Mice were sacrificed by CO2 asphyxiation and cervical dislocation, spleens removed, and a single cell suspensions prepared. Cells were immunostained for the surface expression of CD11c, CD11b, CD86, MHC II, Jag-1, 101

OX40L, CD39, and CD73 and percentages within the lymphocyte population determined by flow cytometric analysis using the previously-described antibodies.

Statistics

For comparisons of multiple groups with only one parameter, one-way ANOVA was performed followed by Tukey’s post-hoc test was used. For comparisons of multiple groups with 2 or more parameters a two-way ANOVA was performed followed by

Tukey’s or Newmann-Keuls post-hoc test. To determine if surface marker expression on

BMDCs correlates with Treg induction, Pearson’s correlation was calculated using Prism

GraphPad version 6. For immunohistochemistry data of neuron counts and striatal density, all values outside the 99% confidence interval were censored and significance was determined by one-way ANOVA followed by Tukey’s post-hoc test. Statistics were performed using Prism GraphPad version 6 (GraphPad Software Inc.). A value was considered significant when the multiple comparison-adjusted P value was less than

0.05.

RESULTS

Activation of DC2.4 and DC3.2 cells with LPS and N-α-Syn is mitigated by GM-CSF

One of the functional definitions of tolerogenic DCs is the lack of an inflammatory response to maturation stimulation (Mahnke et al., 2002). To test if GM-CSF is inducing a tolerogenic state, we first needed to determine culture conditions where DCs are activated by LPS (a strong, general proinflammatory stimulus), N-α-Syn (a post- translational form of α-Syn which is only found in synucleopathy patients (Duda et al.,

2000; Giasson, 2000)), or Poly I:C (a mimic of double stranded RNA). One of DCs 102

responses to proinflammatory stimulus is increasing nitric oxide (NO) production. The

DC lines DC2.4 and DC3.2 were stimulated for 24 hr with increasing (0-100 µg/ml) LPS,

N-α-Syn, or Poly I:C concentrations and the Griess assay was used to test the concentration of released nitrite (the product of the oxidation of nitric oxide after release from BMDCs). LPS, and to a lesser extent N-α-Syn, significantly increase the release of nitrite in the media of both DC lines compared to media alone (0 µg/ml) (Figure 3.2A).

Poly I:C did not increase the concentration of nitrite compared to media at any concentration. For LPS, maximal nitrite concentration was found at 10 µg/ml and for N-

α-Syn the maximal nitrite concentration was found at 30 µg/ml. These concentrations were used for all future experiments with these cell lines.

To determine if GM-CSF can diminish the concentration of released nitrite,

DC2.4 and DC3.2 cells were cultured in cells with 10 µg/ml LPS and increasing concentrations (0-200 ng/ml) of GM-CSF, and the concentration of nitrite was tested with the Griess assay. None of the tested concentrations of GM-CSF significantly decreased the LPS-induced increase in nitrite concentration in either cell line (Figure 3.2B). This was done with GM-CSF administered at the same time as LPS. If GM-CSF is inducing tolerogenic DCs, it may need to be administered before the stimulation with LPS to see an effect. DC2.4 cells were treated for 2 days with increasing concentrations of GM-

CSF (0-200 ng/ml) prior to stimulation with LPS and N-α-Syn for 24 hr. GM-CSF induces a dose-dependent decrease in nitrite after LPS and N-α-Syn stimulation (Figure

3.2C). It appears that even though 200 ng/ml GM-CSF leads to the most suppression of nitrite release, this is significant, but not dramatically different than 20 ng/ml. To determine if this decrease in nitrite concentration is due to a loss of DC viability, the 103

Figure 3.2 GM-CSF pre-treatment, but not co-treatment, decrease the LPS- and N-α-

Syn-induced increase in nitrite release in DC2.4 and DC3.2 cells with no decrease in cell viability

104

Figure 3.2 GM-CSF pre-treatment, but not co-treatment, decrease the LPS- and N-

α-Syn-induced increase in nitrite release in DC2.4 and DC3.2 cells with no decrease in cell viability

A) DC lines DC2.4 and DC3.2 were stimulated with increasing concentrations of LPS, N-

α-Syn, or Poly I:C for 24 hours and the Griess assay was used to test nitrite concentration in the supernatant. B) DC2.4 and DC3.2 cells were treated for 24 hours with 10 µg/ml LPS, increasing concentrations of GM-CSF or the combination of GM-CSF and LPS and nitrite concentration in the supernatant was determined by the Griess assay. C) DC2.4 cells were pre-treated with increasing concentrations of GM-CSF for 2 days prior to 24 hours of stimulation with 10 µg/ml LPS or 30 10 µg/ml N-α-Syn and the

NO concentration in the supernatant was determined by the Griess assay. D) Cell viability was determined after 2 days of 20 ng/ml GM-CSF pre-treatment, then 24 hours of stimulation with 10 µg/ml LPS or the combination, and the AlamarBlue cell viability assay was used to determine the percentage of the dye reduced. Significance was determined by One-way ANOVA followed by Tukey’s post-hoc test where a result is significant if p<0.05. For A and B, a-significant from media alone, b-significant from LPS control and n=4 in A and n=5 in B. For C, a-significantly different from media, b- significantly different from 10 µg/ml LPS, 1-significantly different from 0 ng/ml GM-CSF,

2-significantly different from 2 ng/ml GM-CSF, 3-significantly different from 20 ng/ml GM-

CSF and n=3. For D, a-significant from media alone, b-significant from GM-CSF control, c-LPS control and n=6.

105

alamarBlue cell viability assay was used. DC2.4 and DC3.2 cells were cultured in media or 20 ng/ml GM-CSF for two days, prior to simulation with LPS for 24 hr. For both DC lines, the neither treatment with GM-CSF or LPS stimulation separately or in combination decreased cell viability compared to media alone, unstimulated control

(Figure 3.2D).

To test the activation of DC2.4 cells by another method, we tested the surface expression of activation markers and co-stimulatory molecules. DC2.4 cells were pre- treated for 2 days with GM-CSF prior to 24 hr stimulation with LPS and N-α-Syn and flow cytometry was performed. Few DC2.4 cells (5% or less) expressed CD11c (Figure

3.3A), an integrin receptor which is a marker for myeloid lineage DCs (Ziegler-Heitbrock et al., 2010). CD86 is an activation marker for DCs which binds to CD28 on T cells, leading to T cell proliferation and activation (Lu et al., 1997). Even when cultured in media alone, about 90% of DC2.4 cells express CD86 (Figure 3.3B). Pre-treatment with

GM-CSF, LPS or N-α-Syn stimulation, or a combination can do decrease the high, basal surface expression of CD86. Less than 5% of DC2.4 cells express OX40L, a co- stimulatory molecule which preferentially promotes Th2 CD4+ cells proliferation

(Ohshima et al., 1998) and Treg survival (Griseri et al., 2010) (Figure 3.3C). Jagged-1

(Jag-1) is a Notch-1 ligand which also promotes the differentiation of Tregs (Hoyne et al.,

2000). GM-CSF, LPS and N-α- Syn, alone or in combination increases Jag-1 surface expression on DC2.4 cells (Figure 3.3D). In all conditions, GM-CSF does not promote the surface expression of co-stimulatory molecules OX40L and Jag-1. In addition,

DC2.4 cells constitutively express high CD86 in all conditions, meaning there is no upregulation following stimulation with LPS and N-α-Syn. As a result, the activation of

DC2.4 could not be determined by flow cytometry. Because CD86 surface expression is not changed on DC2.4 cells and the expression of Jag-1 and especially OX40L was low 106

Figure 3.3 Flow cytometric analysis of DC2.4 cells after GM-CSF pre-treatment and LPS and N-α-Syn stimulation

107

Figure 3.3 Flow cytometric analysis of DC2.4 cells after GM-CSF pre-treatment and

LPS and N-α-Syn stimulation

DC2.4 cells were pre-treated in 20 ng/ml GM-CSF for 2 days prior to 24 stimulation with

10 µg/ml LPS or 30 µg/ml N-α-Syn prior to staining cells for CD11c (A), CD86 (B),

OX40L (C), and Jag-1 (D). Data presented are from 1 replicate so no statistical analyses were performed.

108

and relatively unchanged, DC2.4 cells are not the best model for determining if GM-CSF is inducing a tolerogenic DCs.

BMDC activation by LPS and N-α-Syn is mitigated by GM-CSF

Due to GM-CSF being unable to induce a tolerogenic state in DC lines, we next sought to determine if GM-CSF can induce a tolerogenic state in BMDCs. First, we needed to determine optimal differentiation conditions for BMDCs by culturing BMDCs with 10 or 20 ng/ml GM-CSF with and without 10 ng/ml IL-4. After 8 days, BMDCs were cultured in media alone for 3 days, 20 ng/ml GM-CSF for 3 days or 20 ng/ml GM-CSF for

2 days prior to 1 day of stimulation with 10 μg/ml LPS and surface markers were determined by flow cytometry. In all conditions, 50% or more of the BMDCs expressed

CD11c with a generally higher percentage of CD11c+ cells found on BMDCs differentiated without IL-4 (Figure 3.4A). Unlike the DC2.4 cells, LPS stimulation increase the surface expression of CD86 in all BMDC differentiation conditions (Figure

3.4B). BMDC differentiation with IL-4 increased the percentage of unstimulated BMDCs expressing CD86. All conditions displayed fewer CD86-expressing BMDCs than DC2.4.

OX40L expression was higher when BMDCs were differentiated with IL-4 (Figure 3.4C).

When BMDCs were differentiated without IL-4, LPS-stimulation of BMDCs decreased the number of OX40L+ cells, however, BMDCs differentiated with IL-4 prior to LPS stimulation appeared to increase the number of BMDCs expressing OX40L.

Differentiating BMDCs in the presence of IL-4 appears to diminish the number of BMDCs expressing Jag-1 (Figure 3.4D). For BMDCs differentiated without IL-4, the number of

Jag-1+ BMDCs increases with GM-CSF treatment and increases further with LPS 109

Figure 3.4 Flow cytometric analysis of BMDCs after differentiation in GM-CSF or GM-

CSF and IL-4 prior to stimulation with LPS

110

Figure 3.4 Flow cytometric analysis of BMDCs after differentiation in GM-CSF or

GM-CSF and IL-4 prior to stimulation with LPS

BMDCs were differentiated from the bone marrow for 8 days with 10 or 20 ng/ml GM-

CSF with 0 or 10 ng/ml IL-4. After differentiation, BMDCs were cultured in media alone, media supplemented with 20 ng/ml GM-CSF or media supplemented with 20 ng/ml GM-

CSF for two days prior to 24 hours of stimulation with 10 µg/ml LPS. Surface expression of CD11c (A), CD86 (B), OX40L (C), and Jag-1 (D) was determined by flow cytometry.

There are 2 replicates of CD11c and 1 replicate of CD86, OX40L, and Jag-1. No statistics could be performed.

111

stimulation. The conclusion from this experiment is that differentiating BMDCs in 20 ng/ml GM-CSF without IL-4 appears to be the optimal condition for expressing all 4 of the markers tested. BMDC differentiated with this condition showed a high percentage positive for CD11c, a noticeable increase in CD86+ cells after stimulation and a noticeable increase in the percentage of Jag-1+ BMDCs with GM-CSF treatment or LPS stimulation. Because BMDCs exhibit lower basal activation state compared to the

DC2.4 cells, BMDCs are a better model for testing the tolerogenic state of DCs.

Since there were surface expression differences on BMDCs with GM-CSF culture followed by LPS and N-α-Syn stimulation compared to DC2.4 cells, we next tested if BMDCs increase the release of nitrite after stimulation with LPS and N-α-Syn stimulation. Stimulation with LPS and N-α-Syn increase the release of nitrite compared to BMDCs continued in culture with GM-CSF. The continued culture of BMDCs with

GM-CSF prior to LPS and N-α-Syn stimulation mitigates the increase in nitrite concentration (Figure 3.5A). These data show that BMDCs can release nitrite after stimulation, just as DC2.4 cells. However, it should be noted that the increase in nitrite after stimulating BMDCs with LPS and N-α-Syn stimulation is less than the increase after stimulating DC2.4 cells.

To ensure that the changes in nitrite production were not due to the loss of

BMDCs by GM-CSF culture without or with LPS or N-α-Syn stimulation, the alamarBlue cell viability assay was performed. The continued culture of BMDCs in GM-CSF prior to

LPS stimulation significantly lowered cell viability compared to media and GM-CSF controls, however the cell viability was still over 90% (Figure 3.5B). LPS stimulation alone, and N-α-Syn stimulation with or with continued GM-CSF culture, did not affect cell viability. These data demonstrate that the decrease in nitrite production is not due to a dramatic loss of BMDCs. 112

Figure 3.5 BMDC culture with GM-CSF mitigates the LPS- or N-α-Syn-induced increase in the nitrite concentration in the supernatant, without effecting cell viability

113

Figure 3.5 BMDC culture with GM-CSF mitigates the LPS- or N-α-Syn-induced increase in the nitrite concentration in the supernatant, without effecting cell viability

BMDCs were matured for 8 days with 20 ng/ml GM-CSF. BMDCs were either cultured in media alone or 20 ng/ml GM-CSF for 2 days prior to being stimulated for 24 hours with 100 ng/ml LPS or 30 µg/ml N-α-Syn. A) The Griess assay was used to measure the nitrite concentration in the supernatant. Significance was determined by One-way

ANOVA followed by Tukey’s post-hoc test, n=9. a-significantly different from media alone, b-significantly different from GM-CSF control, c-significantly different from LPS or

N-α-Syn control. B) Cell viability was also tested using the alamarBlue assay where cell viability was determined by measuring the percentage of the alamarBlue dye reduced.

Significance was determined by One-way ANOVA followed by Tukey’s post-hoc test and p<0.05. a-significantly different from media alone, b-significantly different from GM-CSF control, n=9.

114

After determining the conditions to differentiate the BMDCs, the next experiments performed were designed to test if GM-CSF promotes a tolerogenic state in BMDCs.

Tolerogenic DCs display low expression of co-stimulatory molecules, even with a maturation stimulus. To test this hypothesis, flow cytometric analysis was used to determine if GM-CSF, N-α-Syn or the combination significantly increases the surface expression of different co-stimulatory molecules. The percentage of CD11c+ cells in the single cell suspension was determined by flow cytometry. More than 75% of cells in all treatment groups are positive for the DC marker CD11c, even though there was a significant reduction of CD11c+ cells in BMDCs continued in GM-CSF culture prior to N-

α-Syn (Figure 3.6A). Flow cytometry was also used to determine the surface expression of the myeloid-origin cell marker CD11b. As with CD11c, BMDCs continued in GM-CSF culture prior to N-α-Syn stimulation display significantly reduced surface CD11b, however more than 95% of single cells are CD11b+ (Figure 3.6B). When DCs mature, they upregulate their surface expression of co-stimulatory molecules, such as MHC II and CD86. The mean fluorescent intensity (MFI) of the activation state of MHC II and

CD86 on CD11c+ BMDCs was significantly increased after N-α-Syn stimulation of media-cultured BMDCS, but this was significantly diminished by pretreatment with GM-

CSF (Figure 3.6C, D). The MFI of OX40L was not altered by the pretreatment with GM-

CSF, N-α-Syn stimulation, or a combination (Figure 3.6E). The surface expression of

Jag-1 is significantly increased with GM-CSF pretreatment which is further increased by

N-α-Syn stimulation (Figure 3.6F). The surface expression of CD11c, OX40L, and Jag-1 are similar to the results in Figure 3.4. We also tested the expression of the surface

ATPases CD39 and CD73. CD39 is the enzyme that converts extracellular ATP to AMP and CD73 converts AMP to adenosine (Deaglio et al., 2007). Neither the continuation of

GM-CSF culture, nor the stimulation with N-α-Syn increased the surface expression of 115

Figure 3.6 Surface expression of co-stimulatory molecules on CD11c+ BMDCs

116

Figure 3.6 Surface expression of co-stimulatory molecules on CD11c+ BMDCs

BMDCs were cultured in media alone or with 20 ng/ml GM-CSF for 2 days prior to stimulation with 30 µg/ml N-α-Syn for 1 day. On the X-axis, treatment groups are designated as (1) media-cultured, unstimulated BMDCs; (2) GM-CSF-cultured, unstimulated BMDCs; (3) media-cultured, N-α-Syn-stimulated BMDCs; and (4) GM-CSF- cultured, N-α-Syn-stimulated BMDCs. BMDCs were stained for CD11c, CD11b, MHC II,

CD86, OX40L. Jag-1, CD39 and CD73 for flow cytometric analysis. Panels A and B are the percentage of single cells positive for CD11c and CD11b respectively. The numbers are the mean percentage of single cells positive for each marker. The mean fluorescent intensity (MFI) was determined on CD11c+ cells for MHC II (C), CD86 (D), OX40L (E),

Jag-1 (F), CD39 (G) and CD73 (H). There are 7 replicates for each treatment group.

Significance was determined by one-way ANOVA and Tukey’s post-hoc test. a- significantly different from media BMDCs, b-significantly different from GM-CSF-treated

BMDCs c-significantly different from N-α-Syn-treated BMDCs.

117

CD39 and CD73 (Figure 3.6G, H). These data demonstrate that continued culture with

GM-CSF diminishes the N-α-Syn-induced increase of CD86 and MHC II surface expression and increases Jag-1 surface expression. In total, these data indicate that the

BMDCs tested are in a tolerogenic state due to their decreased expression of the co- stimulatory molecules (MHC II and CD86) and increased expression of anti-inflammatory

Jag-1.

Next, we wanted to test the expression of proinflammatory cytokines and other proinflammatory mediators. From all 4 BMDC treatment groups, RNA was isolated, cDNA was generated, and PCR arrays were run to test the expression of proinflammatory genes relative to the media-cultured, unstimulated BMDCs. GM-CSF culture without stimulation did cause 2-fold upregulation of IL6, CXCL9, CXCL3, CCL17,

SELE, CCL24, CCR1, CXCR2, and TNFSF14 and the downregulation of CXCL10,

CD40, TNF, TLR1, and BCL6 (Figure 3.7). N-α-Syn stimulation of media-cultured

BMDCS increased expression of several genes including CXCL10, IL6, IL23A, PTGS2,

CD40, among others. Several genes are downregulated including CCR2, FOS, CXCR4,

IL6RA among others. The continued culture in GM-CSF prior to N-α-Syn stimulation diminished the expression of several genes such as IL6, PTGS2, CD40, and LTA among others. Conversely, several genes were increased with the continued culture in GM-

CSF prior to N-α-Syn stimulation, for example, IL23A, NOS2, and IL1A. Three genes that were downregulated by N-α-Syn stimulation in media-cultured BMDCs were further downregulated with GM-CSF pretreatment; these genes are CXCR2, CXCR4 and FOS.

However, other genes downregulated by N-α-Syn stimulation of media-cultured BMDCs were not as downregulated in GM-CSF-cultured BMDCs such as TNFSF14, IL6RA, and

CCR2.

118

Figure 3.7 Gene expression of GM-CSF, N-α-Syn, or GM-CSF+N-α-Syn-treated BMDCs compared to media-cultured, unstimulated BMDCs

119

Figure 3.7 Gene expression of GM-CSF, N-α-Syn, or GM-CSF+N-α-Syn-treated

BMDCs compared to media-cultured, unstimulated BMDCs

BMDCs were differentiated for 8 days with 20 ng/ml GM-CSF and half of the BMDCs were cultured for 2 days with media alone 20 ng/ml GM-CSF. Some BMDCs from each culture condition were stimulated with 30 µg/ml N-α-Syn for 6 hr. RNA was isolated, converted to cDNA and PCR arrays of pro-inflammatory genes were performed. Gene expression was determined relative to the media-cultured, unstimulated BMDCs. n=6.

The fold change is indicated, and red shading indicates upregulation of gene expression and green shading indicates downregulation.

120

This gene expression data highlights several key points. N-α-Syn stimulation increases the expression of proinflammatory cytokines like IL6, IL23A, TNF, IL1A, IL1B,

IL18, and IFNG which may be increased or decreased by continued culture with GM-

CSF. The expression of the anti-inflammatory gene IL10 was increased with N-α-Syn stimulation of media-cultured BMDCs, and further increased with continued culture in

GM-CSF prior to N-α-Syn stimulation. The increase in anti-inflammatory and decrease in proinflammatory cytokines are in line with GM-CSF-cultured BMDCS being tolerogenic. However, some proinflammatory genes are increased, which is counter to our hypothesis. Next, we tested the expression of other markers of maturation in our

BMDCs. Several genes associated with Toll-like receptor signaling, such as TLR1,

TLR3, TLR2, CD14, MYD88, and NFKB1 are increased by N-α-Syn stimulation of media-cultured BMDCs, and continued culture in GM-CSF decreased the expression of

TLR1 and TLR3 but increased the expression of TLR2. This suggests that GM-CSF pretreatment may alter the expression of the TLRs, including TLR1, TLR2, and TLR4, which can be receptors for α-Syn (Su et al., 2008; Stefanova et al., 2011; Fellner et al.,

2013; Daniele et al., 2015). However, the expression of MYD88 and NFKB1, the downstream signaling cascade of the Toll-like receptors, and CD14, the co-receptor for

TLR4, was not affected by the continued culture of BMDCs in GM-CSF. This suggests that it is not that GM-CSF is able to diminish the ability of the BMDCs to respond to N-α-

Syn, but rather GM-CSF alters the response of the BMDCs. The continued culture of

GM-CSF diminished the N-α-Syn-induced expression of CD40. The gene expression of

CD40, a T cell co-stimulatory molecule, is similar to the surface expression of CD86, a co-stimulatory molecule with similar function. CCR7 is marker of DC maturation (Worbs and Forster, 2007), which was decreased -1.42-fold in the continued culture in GM-CSF without stimulation. However, N-α-Syn stimulation increases CCR7 expression, which is further increased by continued culture of BMDCs with GM-CSF. Lastly CEBPB, a 121

transcription factor involved in TNFα expression (Pope et al., 1994) was increased by stimulation with N-α-Syn and expression was further increased by culture with GM-CSF prior to N-α-Syn stimulation. In total, these gene expression changes demonstrate that

GM-CSF does not prevent the maturation of BMDCs or diminish the ability of BMDCs to respond to the N-α-Syn stimulation. However, GM-CSF does alter how BMDCs respond to N-α-Syn stimulation, which increases the expression of some proinflammatory and anti-inflammatory genes.

We also tested the release of nitrite from these BMDCs. Overall, there was no significant change in nitrite concentration between the BMDC treatment groups (Figure

3.8). However, there was a trend to more nitrite release with N-α-Syn stimulation compared to all other groups, but the concentration of nitrite was decreased compared to previous experiments (Figure 3.5). Interestingly, in the PCR array from Figure 3.7,

NOS2, the gene for the major producer of NO in activated BMDCs, inducible nitric oxide

(iNOS), was significantly increased with N-α-Syn stimulation and NOS2 was more highly expressed in BMDCs continued in culture with GM-CSF prior to N-α-Syn stimulation.

The expression of NOS2 does not correlate with the nitrite concentration determined by the Griess assay.

In addition to testing the expression of proinflammatory genes, release of proinflammatory cytokines and chemokines from different treatment groups of BMDCs was tested by a Luminex magnetic bead assay. N-α-Syn stimulation of media-cultured

BMDCs significantly increased the release of IL-1β, IL-10, Lix, MIG, IL-6, IL-12p70, IL-

13, IL-15, IFNγ, RANTES and TNFα compared to the media-cultured, unstimulated controls (Figure 3.9). Continued culture in GM-CSF prior to N-α-Syn stimulation increased the release of IL-1α, IL-1β, IL-2, IL-5, IL-7, IL-9, IL-10, IL-12p40, IL-17, and IP-

10. The continued culture in GM-CSF prior to N-α-Syn stimulation decreased the 122

Figure 3.8 Nitrite concentration in supernatant from BMDCs

123

Figure 3.8 Nitrite concentration in supernatant from BMDCs

BMDCs were matured for 8 days with 20 ng/ml GM-CSF. BMDCs were then cultured in media alone or 20 ng/ml GM-CSF for 2 days prior to being stimulated for 24 hours with

30 µg/ml N-α-Syn or being left unstimulated. The Griess assay was used to measure nitrite concentration in the supernatant. Significance was determined by One-way

ANOVA followed by Tukey’s post-hoc test and p<0.05. n=28 (4 technical replicates from each of 7 wells of BMDCs from each treatment group).

124

release of Lix and MIG. The continued culture in GM-CSF decreased MIP2 release without N-α-Syn stimulation; however, N-α-Syn stimulation after the continued culture of

BMDCs in GM-CSF prevented this decrease. Continued culture of BMDCs in GM-CSF prior to N-α-Syn stimulation increased the release of IL-10, an anti-inflammatory cytokine. The continued culture in GM-CSF did not significantly alter the N-α-Syn- induced release of proinflammatory cytokines IL-6, IL-12p70, IL-13, IL-15, IFNγ, MIP2,

RANTES, and TNFα. In total, these data show that the continued culture in GM-CSF pretreatment does not mitigate the release of proinflammatory cytokines or chemokines, but does increase the release of IL-10. While there are changes in the expression on the RNA level, but this does occur on the level of protein release. These data suggest that GM-CSF is more an immune modulator altering the response of BMDCs as opposed to broadly suppressing the immune response.

Lastly, we tested if the continued culture of BMDCs in GM-CSF with or without N-

α-Syn stimulation increases the expression of IDO in lysate as well as kynurenine release in the supernatant. IDO is the enzyme that converts tryptophan to kynurenine

(King and Thomas, 2007). The increase in IDO expression, kynurenine concentration, and concentration of kynurenine’s metabolites are inducers of Tregs (Munn and Mellor,

2013) and may be part of the mechanism of Treg induction. BMDCs were cultured in media alone or with GM-CSF and left unstimulated or stimulated with N-α-Syn as above.

BMDCs were lysed and Western blot was used to determine the expression of IDO relative to β-actin. Media-cultured, unstimulated BMDCs displayed the highest expression of IDO, which was significantly decreased by continued culture in GM-CSF, with or without N-α-Syn stimulation (Figure 3.10A). Mass spectrometry was used to quantify the kynurenine concentration in the media. There was no significant difference in the concentration of released kynurenine between 125

Figure 3.9 Cytokines and chemokines released from BMDCs

126

Figure 3.9 Cytokines and chemokines released from BMDCs

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to culture with media alone or 20 ng/ml GM-CSF for 2 days, then stimulation with 30 µg/ml N-α-Syn for 1 day.

On the X axis, treatment groups are designated as (1) media-cultured, unstimulated

BMDCs; (2) GM-CSF-cultured, unstimulated BMDCs; (3) media-cultured, N-α-Syn- stimulated BMDCs; (4) GM-CSF-cultured, N-α-Syn-stimulated BMDCs. The supernatant was removed and the Luminex XMAP pro-inflammatory cytokine and chemokine array.

There are three 2-3 technical replicates for each of 7 wells of BMDCs (n=19 for each group). Significance was determined by One-Way ANOVA and Tukey’s post-hoc test. a-significantly different from media BMDCs, b-significantly different from GM-CSF-tested

BMDCs c-significantly different from N-α-Syn-treated BMDCs (p>0.05).

127

treatment groups, but the media-cultured, unstimulated BMDCs appear to exhibit the highest concentration of released kynurenine (Figure 3.10B). This finding is contrary to our hypothesis that the continued culture in GM-CSF would increase expression of IDO or release of kynurenine, which are markers of tolerogenic DCs.

In conclusion, from the flow cytometric data, it appears that GM-CSF decreases the surface expression of co-stimulatory molecules, and increases the surface expression of Jag-1, which is in line with our hypothesis of GM-CSF inducing a tolerogenic state. However, GM-CSF does not mitigate the expression or release of proinflammatory cytokines, or increase the expression of IDO or kynurenine. These data do not support the hypothesis. Overall, it appears that GM-CSF is altering how BMDCs respond to stimulation with N-α-Syn, but does not maintain the tolerogenic state.

BMDCs increase Treg frequencies

In the above section, we characterized the surface markers, genes expressed, and products released by BMDCs cultured in media alone or GM-CSF prior to N-α-Syn stimulation. One of the properties of tolerogenic DCs is the ability to induce Tregs

(Maldonado and von Andrian, 2010) and the next experiment was to determine if

BMDCs cultured in media or GM-CSF with and without N-α-Syn stimulation were able to induce Tregs in culture. BMDCs were differentiated and cultured as described above and were co-cultured with CD4+ cells isolated from spleens of C57BL/6J mice. After 5 days, the percentage of Tregs (CD4+ CD25+ Foxp3+) in the non-adhered cells was determine by flow cytometric analysis. In CD4+ cells prior to co-culture, ~2% of the

CD4+ cells are Tregs (Figure 3.11A). All BMDC treatment groups significantly increased the percentage of Tregs (CD4+ CD25+ Foxp3+) compared to CD4+ cells alone. Media- cultured, unstimulated and N-α-Syn-stimulated BMDCs induced significantly more Tregs 128

Figure 3.10 Expression of IDO and kynurenine release in the supernatant

129

Figure 3.10 Expression of IDO and kynurenine release in the supernatant

BMDCs were differentiated for 8 days with 20 ng/ml GM-CSF. At this time, the BMDCs were cultured in media alone or in 20 ng/ml GM-CSF for 2 days. BMDCs were left unstimulated or stimulated with 30 µg/ml N-α-Syn for 24 hour. A) Western blot was performed on lysate and the ratio of intensity for IDO and β-actin from each of the treatment groups (n=2 media, n=6 GM-CSF, n=7 N-α-Syn, and n=7 GM-CSF + N-α-Syn) was graphed. Significance was determined by one-way ANOVA followed by Tukey’s post-hoc test and was significance was determined by p<0.05. B) HPLC and Mass spectroscopy was used to determine the concentration of kynurenine in the supernatant of BMDCs treated as above. n=7 for all groups. Significance was determined by One- way ANOVA followed by Tukey’s post-hoc test.

130

than the GM-CSF-cultured BMDCs. GM-CSF-cultured, N-α-Syn-stimulated BMDCs induced significantly fewer Tregs than all other BMDC treatment groups. CD4+ T cells cultured with CD3/CD28 transactivator beads also increased the percentage of Tregs relative to CD4+ cells prior to culture, but to a significantly decreased degree than all

BMDC treatment groups except for the GM-CSF-cultured, N-α-Syn-stimulated BMDCs

(Figure 3.11A). In a separate experiment, CD4+ cells were stained before culture (CD4+ alone), with media-cultured, unstimulated BMDCs directly or in transwell, or with CD11c+ cells positively enriched from the spleen of naïve C57BL/6J mice. Non-adhered cells were removed and the percentage of Tregs was determined by flow cytometry. While there were fewer Tregs in this experiment compared to the experiment in Figure 3.11A, media-cultured, unstimulated BMDCs still increased the percentage of Tregs ~20 fold compared to the starting CD4+ T population (Figure 3.11B). When CD4+ T cells were cultured in a transwell assay with media-cultured BMDCs, the percentage of Tregs was reduced compared to direct culture, (Figure 3.11B). Culture of CD4+ cells in a transwell assay increased Tregs ~3-fold, which was not significant compared to the starting CD4+ population, which is similar to the ~5-fold increase with CD3/CD28 transactivator beads.

When splenic CD11c+ DCs were cultured with CD4+ cells, there was not a significant increase in the percentage of Tregs compared to the starting CD4+ population (Figure

3.11B). However, the splenic CD11c+ cells increased Tregs ~3-fold, compared to the starting CD4+ population, similar to the CD3/CD28 transactivator beads and transwell culture. Combined, these data show that direct contact is needed for BMDCs to induce

Tregs, and that Tregs induction is more than activation of T cells in general since neither

CD3/CD28 activator beads nor CD11c+ DCs induce Tregs to the same percentage.

However, contrary to our hypothesis, continuing GM-CSF culture significantly decreases the ability of BMDCs to induce Tregs, both with and without N-α-Syn stimulation. 131

Figure 3.11 BMDCs induce functional Tregs in vitro

132

Figure 3.11 BMDCs induce functional Tregs in vitro

BMDCs were cultured with media or 20 ng/ml GM-CSF for 2 days prior to stimulation with 30 µg/ml N-α-Syn for 1 day. These BMDCs or CD3/CD28 transactivator beads were then co-cultured with CD4+ T cells isolated from spleens for 5 days. Non-adhered cells were removed from co-culture and the percentage of Tregs (CD25+ Foxp3+) was determined within the CD4+ T population. (A) Representative, flow cytometric plots of non-adhered cells from the starting CD4+ T cells or the co-culture of CD4+ cells with

BMDCs or CD3/CD28 beads are presented. Quantification of 3 co-cultures of BMDCs of each treatment group with CD4+ T cells. Significance was determined by one-way

ANOVA followed by Newman-Keuls post-hoc test. a-significantly different from CD4+ T cells alone, b-significantly different from media-cultured BMDCs, c-significantly different from GM-CSF-cultured, unstimulated BMDCs, d-significantly different from media- cultured, N-α-Syn-stimulated BMDCs, e- significantly different from GM-CSF-cultured, N-

α-Syn-stimulated BMDCs. (B) CD4+ T cells were cultured with media-cultured, unstimulated BMDCs directly or in transwell and CD4+ T cells were cultured with

CD11c+ cells isolated from the spleen of naïve mice. Quantification of 3 co-cultures of

BMDCs of each treatment group with CD4+ T cells. Significance was determined by one-way ANOVA followed by Tukey’s post-hoc test. a-significantly different from CD4+

T cells alone, b-significantly different from media-cultured BMDCs. (C) Non-adhered cells from media-treated BMDCs were removed after 5 days of co-culture and Tregs

(CD4+ CD25+). Tregs (CD4 CD25+) and responder T cells (CD4+ CD25-) were isolated from the spleens of naïve mice. Dilutions of Tregs were cultured with CFSE-labelled

CD4+ CD25- cells, which were stimulated with CD3/CD28 beads. Linear regression was performed to compare the slopes of the lines. a-significantly different slope (p=0.006) from spleen-derived Tregs. 133

To determine if CD4+ CD25+ Foxp3+ cells induced by BMDCs possess the proliferation suppression function, CD4+ CD25+ cells were isolated from co-culture with

BMDCs and CD4+ T cells from naïve mouse spleen to test the ability of Tregs to suppress the proliferation of CFSE-labelled CD4+ CD25- responder T cells (Figure

3.11C). The BMDC-induced Tregs suppress CD4+ CD25- proliferation to a greater extent than splenic Tregs. These results demonstrate that BMDCs can induce the formation of Tregs in the contact-dependent manner and these Tregs possess suppressive function.

Next, to determine if the surface expression of markers on BMDCs correlates with the induction of Tregs, 3 replicates of BMDCs were cultured with CD4+ cells isolated from mouse spleens. The surface expression of each of the 8 surface markers was determined by mean fluorescent intensity (MFI). Treg percentage was determined in the non-adhered cells. Interestingly, there was a positive correlation between MFI of

CD39, MHC II, and CD11c expression (Pearson’s correlation of 0.8357. 0.7573 and

0.6266 respectively, Figure 3.12). There were weaker positive correlations with Jag-1,

CD73, and CD11b (Pearson’s correlation of 0.5535. 0.5396 and 0.5310 respectively,

Figure 3.12). There were moderate negative correlations between OX40L and CD86 and Treg induction (Pearson’s correlation of -0.5504 and -0.4373 respectively, Figure

3.12). These data may suggest that the surface expression of proteins like CD39 may play a role in the development of Tregs. Interestingly, Jag-1 expression did not seem to strongly correlate with Tregs and OX40L expression negatively correlates with Treg induction. This is of note since both markers are reported to be important for Tregs induction (Gopisetty et al., 2013).

The increased percentage of the Tregs induced by BMDCs could be due to proliferation of the existing Treg population or inducing non-Tregs to become Tregs. To 134

test this, splenocytes were enriched for CD4+ CD25- (non-Tregs, non-effector T cells) and CD4+ CD25+ (Tregs) and both were separately labelled with CFSE and cultured with media-cultured, unstimulated BMDCs or alone for 5 days. Non-adhered cells were stained for CD4, CD25, and Foxp3 and flow cytometry was performed to determine the percentage of Tregs and the percentage of proliferating cells within the Treg population.

For the CD4+ CD25- cells alone, less than 1% of the CD4+ cells were Tregs, but when cultured with media-cultured, unstimulated BMDCs, the percentage of Tregs in the population increased to ~11% of the population (Figure 3.13A). Based on the CFSE labelling, ~40% of the induced Tregs are proliferating. Because the majority of Tregs did not proliferate and the Tregs that did proliferate, this demonstrates that BMDCs are inducing non-Tregs to become Tregs. The 10-fold increase in Treg percentage cannot be accounted for by the proliferation of a minority of Tregs. In the culture of CD4+

CD25+ cells alone, ~22% of the CD4+ T cells are Tregs, but when cultured with BMDCs the percentage of Tregs increased to ~50%. Of these Tregs, ~41% are proliferating

(Figure 3.13B); however, ~52% of the Tregs alone are proliferating. Since the percentage of proliferating Tregs is similar with or without culture with BMDCS, the increase in the percentage of Tregs is likely due to the induction of non-Tregs in the culture to become Tregs. Combined, these data suggest that BMDCs can induce the proliferation of Tregs, but it appears that more of the increase in the percentage of Tregs is induced from the non-Treg population.

BMDCs are neuroprotective in the MPTP model

After characterizing the effects of GM-CSF on BMDCs, we next wanted to test the ability of these cells to be protective in the MPTP model. Adoptive transfer of Tregs into MPTP-intoxicated mice leads to protection of dopaminergic neurons in the substantia nigra and the termini of those neurons in the striatum (Reynolds et al., 2007). 135

Figure 3.12 Correlation of BMDC surface markers with Treg induction

136

Figure 3.12 Correlation of BMDC surface markers with Treg induction

The mean fluorescent intensities (MFI) of CD11c, CD11b, MHC II, CD86, Jag-1, OX40L,

CD39, and CD73 on CD11c+ BMDCs were graphed with the frequency of Tregs (CD4+

CD25+ Foxp3+) after 5 days of co-culture. The black dots are co-culture of media- treated, unstimulated BMDCs with CD4+ cells, the red dots are co-culture of GM-CSF- culture, unstimulated BMDCs with CD4+ cells, the blue dots are co-culture of media- treated, 30 µg/ml N-α-Syn-stimulated BMDCs with CD4+ cells, the green dots are co- culture of GM-CSF-cultured, 30 µg/ml N-α-Syn-stimulated BMDCs with CD4+ cells.

Pearson correlation of the MFI of each marker with Treg frequency is indicated on each graph. The dotted lines represent the 95% confidence interval for the linear regression best-fit line.

137

Because Tregs are induced by DCs, we hypothesized that the adoptive transfer of tolerogenic DCs will increase the percentage of Tregs and protect dopaminergic neuron loss after MPTP. The media-cultured, unstimulated BMDCs were the most tolerogenic given the low co-stimulatory molecule expression, low cytokine production, and highest induction of Tregs, so these BMDCs were used in all in vivo experiments. Initially, the adoptive transfer of BMDCs 8 hr after intoxication with 12 or 16 mg/kg was tested for the protection of dopaminergic (tyrosine hydroxylase positive, TH+ Nissl+) neurons. We found no significant protection of TH+ Nissl+ neurons in the substantia nigra in mice receiving BMDCs compared to MPTP control mice (Figure 3.14). This result suggested that there was insufficient time for BMDCs to migrate from the site of injection into the tail vein to the secondary lymph organs to induce Tregs and protect dopaminergic neurons. However, in the 16 mg/kg MPTP mice, there was a trend toward neuroprotection with the adoptive transfer of BMDCs. This may suggest that BMDCs induce a greater effect when inflammation is greater.

In order to give BMDCs more time to induce Tregs and be neuroprotective in vivo, BMDCs were adoptively transferred 1 and 2 weeks prior to MPTP. This transfer scheme was used because it was protective in the EAE model (Prado et al., 2012).

Seven days after MPTP intoxication, at the peak of dopaminergic neuron loss

(Kohutnicka et al., 1998), mice were sacrificed and immunohistochemistry was performed for dopaminergic (TH+ Nissl+) neurons and the dopaminergic (TH+) termini in the striatum. MPTP intoxication decreased the counts of TH+ neuron counts in the substantia nigra, but this loss was mitigated by the adoptive transfer of BMDCs (Figure

3.15). There was no significant change in the number of non-dopaminergic (TH- Nissl+) neurons, suggesting that the decrease in dopaminergic neurons was not due to the downregulation of TH. MPTP also significantly decreased the density of TH staining in 138

Figure 3.13 BMDCs cause both induction and proliferation of Tregs

139

Figure 3.13 BMDCs cause both induction and proliferation of Tregs

BMDCs were generated by culturing bone marrow with 20 ng/ml GM-CSF for 8 days.

BMDCs were then cultured in media alone for 3 days. At this time, BMDCs were cultured with A) CFSE-labelled CD4+ CD25- cells or B) CFSE-labelled CD4+ CD25+ cells enriched from the spleens of naïve mice. As a control, the CD4+ CD25- and CD4+

CD25+ cells were cultured alone. After 5 days, flow cytometric analysis was performed.

The percentage of Treg (CD25+ Foxp3+ T cells) within the CD4+ T cell population is indicated in the top line. The percentage of proliferating Tregs are indicated in the bottom line. Each is a single replicate so no statistics could be performed.

140

Figure 3.14 Adoptive transfer of BMDCs before MPTP does not protect dopaminergic neurons

141

Figure 3.14 Adoptive transfer of BMDCs before MPTP does not protect dopaminergic neurons

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of pretreatment in media alone. Approximately, 1.5x106 BMDCs were transferred i.v. in

250 µl 8 hours after intoxication with 4 doses of 12 or 16 mg/kg MPTP. Seven days after

MPTP intoxication, mice were sacrificed and immunohistochemistry for tyrosine hydroxylase (TH) was performed on 30 µM sections containing the substantia nigra.

Substantia nigra sections were Nissl counter stained. Stereology was used to count dopaminergic neurons (TH+ Nissl+) in the substantia nigra. The scale bar is 200 µm. n=2 for PBS, n=4 for 12 mg/kg MPTP, n=2 12 mg/kg MPTP + BMDCs, n=5 for 16 mg/kg

MPTP and n=5 for the MPTP + BMDCs. Significance was determined by one-way

ANOVA followed by Tukey’s post hoc test. No values were determined to be statistically significant.

142

the striatum. The adoptive transfer of BMDCs did protect the TH termini in the striatum.

Combined, these data show that the adoptive transfer of BMDCs prior to MPTP administration protects dopaminergic neuron cell bodies and termini from degeneration.

To determine if BMDCs decrease neuroinflammation, reactive microglia in the substantia nigra were counted. BMDCs were adoptively transferred as before, mice were intoxicated with 16 mg/kg MPTP, and 2 days after intoxication, at the peak of neuroinflammation (Jackson-Lewis et al., 1995; Kohutnicka et al., 1998), mice were sacrificed, brains perfused and fixed in 4% paraformaldehyde and immunohistochemistry was performed on sections of the substantia nigra for Mac-1.

Based on the sterologic counting of the reactive microglia per area, MPTP significantly increased the number of reactive microglia relative to PBS (Figure 3.16). However, the adoptive transfer of BMDCs significantly decreased the number of reactive microglia compared to the MPTP control, but this decrease was not to the level of the PBS control.

To determine if the adoptive transfer of BMDCs decreases the expression of proinflammatory mediators induced by MPTP, mice were sacrificed 2 days after MPTP treatment, and RNA was isolated from hemisected midbrains to test expression using a

PCR array for the inflammatory response and autoimmunity. With gene expression normalized to the PBS control mice, MPTP intoxication increased the expression more than 2-fold of CCL3, CXCL10, CCL4, IL1R1, IL1RN, CEBPB, TLR2, and CSF1 (Figure

3.17). The expression of these genes, especially CXCL10 (the gene for IP-10), IL1R1

(IL-1 receptor 1), and CEBPB, a transcription factor upregulated in activated microglia

(Straccia et al., 2011), are indicative of the MPTP-induced neuroinflammation. MPTP more than 2-fold downregulated C3AR1, MYD88, TLR1, CCL2, CCR3. CCR1, TLR9,

CCL11, CXCR4, CCL25, CCL12, RIPK2, and TLR3. The adoptive transfer of BMDCs did not greatly decrease the expression of any genes upregulated by MPTP. However, 143

Figure 3.15 BMDCs are neuroprotective in the MPTP mouse model

144

Figure 3.15 BMDCs are neuroprotective in the MPTP mouse model

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of culture in media alone. Approximately 1.5x106 BMDCs were transferred i.v. in 250 µl 1 and 2 weeks prior to intoxication with four doses of 16 mg/kg MPTP. Seven days after MPTP intoxication, mice were sacrificed and immunohistochemistry for tyrosine hydroxylase

(TH) was performed on 30 µM sections containing the substantia nigra and striatum.

Substantia nigra sections were Nissl counter-stained. Stereology was used to count neurons (both TH+ Nissl+ and TH- Nissl+) in the substantia nigra. The scale bar is 200

µm. n=6 for PBS, n=8 for MPTP and n=7 for MPTP + BMDC group. Numbers in the

TH+ Nissl+ bars are the percentage of neurons remaining compared to PBS controls.

Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. a- significantly different from PBS, b-significantly different from MPTP. From the striatum, the scale bar is 1 mm. TH density was determined for 1.4 mm2 area for each striatum which was normalized to density for the PBS striatum. n=7 for PBS, n=7 for MPTP and n=6 for the MPTP + BMDC group. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. a-significantly different from PBS, b-significantly different from MPTP.

145

Figure 3.16 BMDCs decrease the number of reactive microglia in the MPTP model

146

Figure 3.16 BMDCs decrease the number of reactive microglia in the MPTP model

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of pretreatment in media alone. Approximately, 1.5x106 BMDCs were transferred i.v. in

250 µl one and two weeks prior to intoxication with four doses of 16 mg/kg MPTP. Two days after MPTP intoxication, mice were sacrificed and immunohistochemistry for Mac-

1+ microglia was performed on 30 µM sections containing the substantia nigra.

Stereology was used to count the number of reactive microglia per area the substantia nigra. The scale bar is 200 µm for the larger image and the inset scale bar is 20 µm. n=5 for PBS, MPTP, and MPTP + BMDC group. Significance was determined by One-

Way ANOVA followed by Tukey’s post hoc test. a-significantly different from PBS, b- significantly different from MPTP.

147 there was an almost 2-fold increase in the expression of IL1RN, the antagonist for IL-1 receptor (Gabay et al., 2010), which would diminish IL-1 signaling. Of the genes downregulated by MPTP, there was an increase the expression of CCL2, CCL12,

RIPK2, and TLR3, suggesting BMDCs are compensating for the downregulation of these genes by MPTP. In addition, the adoptive transfer of BMDCs more than 2-fold increased proinflammatory genes IL6RA, IL17A, TNF, and IL6 and the anti-inflammatory IL10.

BMDCs also increased the expression of chemokine-related genes: CXCL2, CXCR2,

CCL1, CXCL1, CXCR1, CXCL3, CXCL9, CCR4, CCR2, CCL20, CCL24, CXCL5, and

CCL8. Combined, these data demonstrate that BMDCs do not suppress the expression of all pro-inflammatory genes, but does increase the expression of anti-inflammatory genes such as IL10 and IL1RN, which is contributing to the decreased microgliosis in

Figure 3.16. In addition to these anti-inflammatory genes, BMDCs are also increase the expression of chemokines which may be altering the balance of infiltrating immune cells may which also be contributing to the decreased neuroinflammation.

To better identify which genes are changed in the midbrain by the adoptive transfer of BMDCs prior to MPTP intoxication compared to MPTP control, Ingenuity

Pathway Analysis was employed. There were 24 genes which were 2-fold or more changed which are associated with the inflammatory response. While the pattern of expression is indicative of inflammation as indicated by the increase in TNF, CCL11,

CXCL10, CXCL6, CXCR4, CCR1, CCR3, and genes related to pathogen recognition such as TLR1, TLR2, TLR3, TLR9, MYD88, and RIPK2. However, some gene changes are indicative of a decreased inflammation. These include downregulation of IL-1 receptor 1, which is the receptor of IL-1α and IL-1β that transduces signals intracellularly

(Sims et al., 1993), CEBPB, a transcription factor which induces TNF expression

148

Figure 3.17 Gene expression of MPTP and BMDCs + MPTP midbrain compared to PBS control midbrain

149

Figure 3.17 Gene expression of MPTP and BMDCs + MPTP midbrain compared to

PBS control midbrain

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of pretreatment in media alone. These BMDCs were transferred i.v. 1 and 2 weeks prior to intoxication with 4 doses of 16 mg/kg MPTP. Two days after intoxication, PBS, MPTP and BMDC + MPTP mice were sacrificed and the brain was removed and hemisected, and the midbrain was incubated in RNAlater for 24hr prior to freezing at -80°C. RNA was isolated from the midbrain, copied to cDNA, and PCR arrays of pro-inflammatory genes were run. (A) Gene expression was determined relative to the PBS control midbrains. n=3 for PBS and n=4 for MPTP and MPTP+BMDC. Fold change was determined using SA Bioscience software. (B) Ingenuity Pathway Analysis was used to determine the expression of genes associated with the inflammatory response in

BMDCs + MPTP midbrain compared to MPTP control mice. Red indicates an upregulation of genes and green represents a downregulation of gene expression with the darker color indicating the larger change in gene expression. The orange lines indicate an effect in line with increased inflammation. The yellow lines are gene changes inconsistent with inflammation.

150

(Pope et al., 1994), and chemokines CCL3L3 and CCL4, which promote inflammation

(Yang and Wang, 2015) and leukocyte recruitment into the brain (Ubogu et al., 2006).

CCL24 is a chemokine released from M2 macrophages which can recruit T cells, especially Tregs (Watanabe et al., 2002; Zamarron and Chen, 2011). In glioblastoma multiforme, CCL2 is important for Treg recruitment into the brain and may perform a similar function after MPTP intoxication (Jordan et al., 2008; Chang et al., 2016). IL1RN prevents the recruitment of IL-1 receptor associated protein to IL-1 receptor which is required for transduction of signals within the cell (Dinarello, 2009), therefore its increased expression, when combined with the downregulation of IL1R1 would diminish

IL-1α and IL-1β signaling. Combined, these data show that the adoptive transfer of

BMDCs diminish, the MPTP-induced inflammation, in line with the data from Figure 6.

BMDCs induce Tregs in vivo

To determine if the neuroprotective effects of BMDCs are due to their ability to increase Treg percentage and/or function in vivo, BMDCs were adoptively transferred into mice as described and sacrifice was performed 1 week after the second transfer of

BMDCs. As a control, 50 μg/kg GM-CSF was i.p. administered to a group of mice every day for 5 days prior sacrifice. The percentages of CD4+ T cells and Tregs were determined by flow cytometry. To test the percentages of CD4+ T cells and Tregs in the blood were determined in blood from cheek bleeds performed 1 day before each transfer or sacrifice. There was no change in the percentage of CD4+ T cells in the blood over time nor was there a change in the percentage of CD4+ cells in the spleen or in the lymph nodes at time of sacrifice in unmanipulated controls, GM-CSF-treated controls or

BMDC-transferred mice (Figure 3.18A). There was also no increase in Tregs over time in the blood, nor in the lymph nodes at time of sacrifice (Figure 3.18B). In the spleen,

GM-CSF did trend to increase the percentage of Tregs, however, the increase was not 151

significant and was not an increase to the degree previously reported (Kosloski et al.,

2013). The adoptive transfer of BMDCs decreased the percentage of Tregs in the spleen significantly compared to the GM-CSF-treated control, but the decrease was not significantly different from the unmanipulated control. Since there was a significantly decreased percentage of Tregs in the spleen, next the suppressive function of Tregs was determined. There was no significant change in the function of Tregs from control or GM-CSF-treated mice (Figure 3.18C). However, the adoptive transfer of BMDCs significantly decreased the function of Tregs relative to both the control and GM-CSF- treated Tregs. These data demonstrate that, without any inflammation, the adoptive transfer of BMDCs decreased the percentage and function of Tregs.

In models where the adoptive transfer of BMDCs ameliorate disease and increase Tregs, there was an ongoing to autoimmune response (Li et al., 2008; Mari et al., 2016). To test if BMDCs increase the percentage and/or function after MPTP intoxication, BMDCs were adoptively transferred 1 week apart and mice were MPTP intoxicated one week after the second transfer. Two days after MPTP intoxication, at the peak of neuroinflammation, mice were sacrificed and CD4+ T cell and Treg percentages and Treg function from the spleen was determined. There was no significant change in the percentage of CD4+ cells in any treatment group, but mice receiving BMDCs prior to

MPTP displayed significantly more splenic Tregs than either PBS or MPTP controls

(Figure 3.19A). By multiplying the percentage of CD4+ T cells and Tregs by the count of total splenocytes, we determined the number of CD4+ T cells and Tregs. MPTP intoxication decreased the number of CD4+ cells in the spleen compared to the PBS control. This was increased by the adoptive transfer of BMDCs. The number of Tregs in the spleen was decreased in the MPTP control mice, but was increased by the adoptive transfer of 152

Figure 3.18 Treg frequency and function after the transfer of BMDCs

153

Figure 3.18 Treg frequency and function after the transfer of BMDCs

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF followed by culture in media alone for 3 days. These BMDCs were i.v. transferred into the tail vein of C57BL/6J mice two weeks and one week prior to sacrifice. One day prior to each transfer and sacrifice, mice were cheek bleed. As a positive control, mice were i.p. administered with 50 µg/kg

GM-CSF once per day for 5 days. (A) The frequency of CD4+ T cells in the single cell population and (B) Tregs (CD4+ CD25+ Foxp3+) in the CD4+ T cell populations were determined by flow cytometry. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test a-significantly different from control and b-significantly different from GM-CSF mice. Control n=11, GM-CSF n=10, and i.v. BMDC n=12.

Pooled spleens from all three groups had Tregs (CD4+ CD25+) enriched and dilutions were cultured with CFSE-labelled CD4+ CD25- cells that were stimulated with

CD3/CD28 beads and proliferation suppression assay was used to test Treg function.

Linear regression was used to compare the slope and elevation. a-significantly different slope rom control Tregs, b -significantly different elevation compared to GM-CSF Tregs.

N=6 for all three treatment groups.

154

Figure 3.19 Treg frequency and function after adoptive transfer of BMDCs prior to MPTP

155

Figure 3.19 Treg frequency and function after adoptive transfer of BMDCs prior to

MPTP

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of pretreatment in media alone. These BMDCs were transferred i.v. 1 and 2 weeks prior to intoxication with 4 doses of 16 mg/kg MPTP. Two days after MPTP intoxication, mice were sacrificed and spleens were removed. (A) Flow cytometry was used to determine the frequency of CD4+ cells in the total splenocyte population and the frequency of

Tregs (CD25+ Foxp3+) in the CD4+ population. Based on this frequency and the splenocyte counts, the counts of CD4+ and Tregs were determined. n=5 for all 3 groups. Significance was determined by One-way ANOVA followed by Tukey’s post hoc test. a-significantly different from PBS and b-significantly different from MPTP. (B)

CD4+ CD25+ cells were enriched from the pooled splenocytes for each treatment group and dilutions of Tregs were cultured with CFSE-labelled CD4+ CD25- cells and

CD3/CD28 transactivator beads for 3 days. Linear regression was used to compare the slope and elevation. a-significantly different elevation from PBS Tregs, b-significantly different elevation compared to MPTP Tregs. N=3 for all 3 treatment groups except for

0.125:1 dilution which only had 1 replicate for each treatment group and 1:1 MPTP

Tregs with 2 replicates.

156

Tregs. These data highlight that BMDCs increase the percentage and the number of

Tregs in the spleen following MPTP intoxication. To test if BMDCs increase the function of Tregs, we isolated CD4+ CD25+ Tregs from spleens and assessed isolates for the capacity to suppress CD3/CD28 stimulated responder T cells. Interestingly, it appears that MPTP treatment increases the suppressive function of Tregs relative to the Tregs from PBS control mice (Figure 3.19B). Transfer of BMDCs diminishes the MPTP- induced increase in Treg suppressive function. More work will need to be done to verify the result of this experiment, but the data in Figure 3.19 show that BMDCs increase the percentage of Tregs in the spleen after MPTP, demonstrating the necessity of inflammation to increase the percentage of Tregs. However, these BMDC-induced

Tregs possess the same suppressive function compared to the PBS control.

BMDC-induced Treg neuroprotection

To determine if the Tregs that are generated from the co-culture of BMDCs and

CD4+ cells are neuroprotective, BMDCs differentiated for 8 days in 20 ng/ml GM-CSF followed by 3 days in media alone were co-cultured with CD4+ cells for 5 days. Either the total non-adhered population or the CD4+ CD25+ cells enriched from the non- adhered population using the Miltenyi CD4+ CD25+ isolation kit as described above were transferred into MPTP-intoxicated mice. The resulting ~400,000 CD4+ CD25+ cells or ~4,000,000 non-adherent cells were adoptively transferred into the tail vein of mice ~10 hours after intoxication with 16 mg/kg MPTP. Based on the purity of the CD4+

CD25+ cells, the ~400,000 cells correspond to ~200,000 Tregs (CD4+ CD25+ Foxp3+) and the ~4,000,000 non-adhered cells correspond to ~400,000 Tregs. The adoptive transfer of BMDCs increases the number of Tregs in the spleen by ~200,000. The adoptive transfer of CD4+ CD25+ cells or non-adhered cells did not protect TH+ Nissl+ 157

Figure 3.20 BMDC-induced Tregs are not neuroprotective

158

Figure 3.20 BMDC-induced Tregs are not neuroprotective

BMDCs were differentiated for 8 days in 20 ng/ml GM-CSF prior to 3 days of pretreatment in media alone. BMDCs were co-cultured with CD4+ cells from splenocytes. After 5 days, the non-adhered cells were removed and either the whole population or enriched the CD4+ CD25+ cells were transferred i.v. in 250 µl 10 hours after intoxication with 16 mg/kg MPTP. The flow cytometry was performed to determine purity of Tregs. For the CD4+ CD25+ enriched cells, 60.6% of the CD4+ were CD25+

Foxp3+ cells which corresponds to ~200,000 Tregs transferred. For the non-adhered population, 12% of the CD4+ population were CD25+ Foxp3+ cells which corresponds to

~400,000 Tregs transferred. Seven days after MPTP intoxication, mice were sacrificed and immunohistochemistry for tyrosine hydroxylase (TH) was performed on 30 µM sections containing the substantia nigra and striatum. Substantia nigra sections were

Nissl counter stained. Stereology was used to count neurons (both TH+ Nissl+ and TH-

Nissl+) in the substantia nigra. The scale bar is 200 µm. n=12 for PBS, n=9 for MPTP, n=7 for the MPTP + BMDC Treg groups, and n=5 for the MPTP+BMDC non-adhered cells. Numbers in the TH+ Nissl+ bars are the percentage of neurons remaining compared to PBS controls. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. a-significantly different from PBS, b-significantly different from

MPTP. From the striatum, the scale bar is 1 mm. TH density was determined for 1.4 mm2 area for each striatum which was normalized to density for the PBS striatum. n=7 for PBS, n=7 for MPTP and n=6 for the MPTP + BMDC group. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. a-significantly different from PBS.

159

neurons in the substantia nigra or TH+ termini in the striatum (Figure 3.20). These data suggest that this number of Tregs from BMDC co-culture is insufficient to protect dopaminergic neurons. Previously, the lowest number of transferred Tregs which lead to neuroprotection was 500,000 (Reynolds et al., 2007), so more Tregs may need to be transferred to promote neuroprotection. It is also possible that the Tregs from BMDC co- culture may be less neuroprotective, even if they possess increased proliferation suppression function (Figure 3.11).

BMDC supernatant does not protect MES23.5 cells, but does decrease BV2 release of nitrite, TNF and IL-6

We next wanted to determine if BMDCs are having direct effects on microglia and dopaminergic neurons apart from inducing Tregs. In our experiments, we have not determined if the BMDCs can migrate to the brain, however, DCs can migrate to the brain after EAE (Clarkson et al., 2014; Clarkson et al., 2015) and stroke (Felger et al.,

2010; Manley et al., 2013) in order to promote the T cell response. It is unclear to what degree the neuroprotective effects are due to the co-stimulatory surface expression or the released products of BMDCs. To test the role of released products from BMDCs in decreased neuroinflammation and neurodegeneration, the BMDC supernatant was tested for the ability to protect the MES23.5 dopaminergic neuron line from toxic insult.

MES23.5 cells were cultured for 24 hours with clarified conditioned media from the culture of media-cultured, unstimulated BMDCs after 11 days in culture. As controls,

MES23.5 cells were also cultured in MES23.5 media or R10 (BMDC) media. After this pre-treatment, MES23.5 cells were treated with increasing concentrations of MPP+ (0-

1,000μM) for 24 hours. At this time, the CellTiter-Glo cell assay was used to determine the production of ATP in MES23.5 cells, which is a surrogate marker for cell viability.

MPP+ was toxic starting at 10 µM in the R10 media-cultured MES23.5 cells, 100 µM for 160

Figure 3.21 BMDC supernatant does not protect MES23.5 cells viability after culture

MPP+ and BV2 supernatant

161

Figure 3.21 BMDC supernatant does not protect MES23.5 cells viability after culture MPP+ and BV2 supernatant

MES23.5 neurons were cultured for 24 hours in and equal volume of MES23.5 media, supernatant from BMDCs cultured for 8 days in 20 ng/ml GM-CSF and 3 days in R10 media, or R10 media. (A) MES23.5 neurons were treated for 24 hr in an equal volume of MES23.5 media containing 0 - 1,000 µM MPP+. Cell viability was determined by celltiter-glo assay. Significance was determined by one-way ANOVA followed by

Tukey’s post-hoc test. The following markers correspond to significance from the following treatment groups a-0, b-0.1, c-1, d-10 e-100 µM or 1-MES23.5 media. n=4.

(B) MES23.5 neurons were treated with BV2 media or supernatants from BV2 cells cultured with and without 100 ng/ml LPS in BV2 media, BMDC supernatant, or R10 media. After 24 hr of treatment, cell viability was determined by the CellTiter-Glo assay.

Significance was determined by one-way ANOVA followed by Tukey’s post-hoc test.

The following markers correspond to significance from the following treatment groups a-

BV2 media, b-BV2 supernatant, c-BV2 + BMDC supernatant, d-BV2 + R10 media, e-

BV2 media + LPS, f-BV2 + BMDC supernatant + LPS. n=10 for BV2 media and BV2 supernatant unstimulated, n=7 BV2 + BMDC supernatant, BV2 + R10 media, BV2 supernatant, BV2 + BMDC supernatant, and BV2 + R10 media stimulated with LPS.

162

the BMDC-cultured MES23.5 cells, and 1,000 μM for MES23.5 media-cultured MES23.5 cells (Figure 3.21A). BMDC supernatant was unable to protect MES23.5 cells relative to the two media controls. In fact, there was a trend toward decreased cell viability, though this was not significant. It is interesting that the MES23.5 cells cultured with R10 media or BMDC supernatant decreased cell viability at lower concentration of MPP+ than

MES23.5 cells. This suggests that something in the R10 media, independent of products released from BMDCs, facilities MPP+ toxicity. Possibly, this is due to the increased concentration of FBS or decreased concentration of N2 supplement that does not directly decrease cell viability, but fails to maintain MES23.5 cell viability with increased MPP+ toxicity.

Because only a small amount of the neurodegeneration is caused by MPP+ in the MPTP model (Benner et al., 2008; Brochard et al., 2009), BMDC supernatant was tested for the ability to protect MES23.5 cells from the supernatant from BV2 microglia cell line. BV2 cells were cultured for 1 day with BV2 media, BV2 media and BMDC supernatant, and BV2 media and R10 media with and without 100 ng/ml LPS. The conditioned media from these different treatment groups or BV2 media alone was then added to MES23.5 cells cultured with MES23.5 media, BMDC supernatant, or R10 media as above and cell viability was determined with the celltiter-glo assay. MES23.5 cells cultured with different BV2 supernatants tended to exhibit decreased cell viability

(Figure 3.21B). BV2 cells stimulated with LPS were not significantly more toxic to

MES23.5 cells than the unstimulated BV2 cells. Regardless if BMDC supernatant was cultured with BV2 cells or MES23.5 cells, there was not significant protection of

MES23.5 cells. Interestingly, the group that trended to display decreased cell viability was the supernatant from BV2 cells cultured with BMDC supernatant and LPS 163

stimulation. As with MPP+-treated MES23.5 cells, BMDC supernatant did not increase the cell viability of MES23.5 cells.

Because BMDC supernatant was unable to protect MES23.5 cells, next the ability of BMDC supernatant to decrease the release of proinflammatory mediators was tested. First, we used the Griess assay to test the concentration of nitrite in the supernatant of LPS-stimulated or unstimulated BV2 cells cultured with BV2 media,

BMDC supernatant, or R10 media. There was limited nitrite in the any unstimulated BV2 supernatant (Figure 3.22). However, there was a significant increase in nitrite concentration in all LPS-stimulated BV2 cell supernatants. This increase in nitrite concentration was significantly mitigated by culture of BV2 cells with BMDC supernatant compared to either of the media-cultured BV2 cells. Cytokine bead array (CBA) was used to test the release of IL-2, IL-4, IL-6, IL-10, IL-17, TNF, and IFNγ from the same treatment groups of BV2 cells as above. Only IL-6 and TNF were significantly elevated in LPS-stimulated BV2 cells (Figure 3.22). As with nitrite, BMDC supernatant decreased the release of both IL-6 and TNF compared to either media. These data suggest that

BMDCs release some factor that decreases the reaction of BV2 cells to activation with

LPS. However, the reduced release of IL-6, TNF, and nitrite does not protect MES23.5 cells from MPP+ or BV2 supernatant (Figure 3.21).

GM-CSF does not increase the percentage of CD11c+ dendritic cells in the spleen

To determine if administration of GM-CSF increases the percentage of splenic myeloid DCs (CD11c+ CD11b+ cells), 50 μg/kg GM-CSF was i.p. administered each day for 5 days prior to sacrifice. The spleen was removed and flow cytometry was used to determine the percentage and the mean fluorescent intensity of DC markers. There was no significant change in the percentage of CD11c+ CD11b+ myeloid DCs in the spleen

(Figure 3.23A). There was not change in the percentage of CD11c+ CD11b+ cells 164

Figure 3.22 Release of cytokines and nitrite from BV2 cells treated with BMDC supernatant and LPS

165

Figure 3.22 Release of cytokines and nitrite from BV2 cells treated with BMDC supernatant and LPS

BV2 cells were cultured in BV2 media, BMDC supernatant from BMDCs cultured for 8 days in 20 ng/ml GM-CSF and 3 days in R10 media, or R10 media with or without 100 ng/ml LPS. After 24 hr of culture the supernatant was removed and nitrite was determined by the Griess assay and the concentration of IL-6 and TNF was determined by CBA. Significance was determined by one-way ANOVA followed by Tukey’s post hoc test. The following markers correspond to significance from the following treatment groups a-BV2 media, b-BMDC supernatant, c-R10 media, d-BV2 media + LPS, e-BMDC supernatant + LPS. n=6 for all nitrite samples, n=3 for CBA.

166

Figure 3.23 Frequency of DCs in the spleen after GM-CSF administration

167

Figure 3.23 Frequency of DCs in the spleen after GM-CSF administration

Mice were either not injected (control) or were given i.p. 50 µg/kg GM-CSF daily for 5 days. Spleens were removed and flow cytometry was performed to determine the frequency of CD11c+ CD11b+ dendritic cells in the single cells of the spleen (A). We also determined the frequency of CD11c+ CD11b+ cells expressing surface MHC II+,

CD86+, Jag-1+, OX40L+, CD39+, and CD73+. Significance was determined by unpaired t test and no surface markers were determined to change significantly. (B) We also determined the MFI of CD11c, CD11b, MHC II, Jag-1, CD39, and CD73 on the surface of CD11c+ CD11b+ cells. Significance was determined for by unpaired t test significance was determined if the p <0.05. a-significantly different from control. n=5 for all groups.

168

expressing MHC II, CD86, OX40L, Jag-1, CD39, and CD73. There was also no significant change in the MFI of CD11c, CD11b, MHC II, Jag-1, and CD39 on CD11c+

CD11b+ cells ((Figure 3.23B). GM-CSF did significantly decrease the surface expression of CD73 on CD11c+ CD11b+ cells. Combined, these data demonstrate that

GM-CSF does not increase the percentage and expression of these markers. This is notable, given that culturing BMDCs with GM-CSF increasing the surface expression of

Jag-1 and trends to increase surface CD73 (Figure 3.6). While it is possible that GM-

CSF may be decreasing the expression of other markers not tested, these data suggest that GM-CSF alone is insufficient to alter splenic DCs to increase the percentage of

Tregs as reported previously (Kosloski et al., 2013). It possible that if these mice were intoxicated with MPTP it would increase the percentage of CD11c+ CD11b+ cells expressing Jag-1 or other surface markers which would promote Tregs in the spleen.

DISCUSSION

Our first hypothesis was that GM-CSF induces and maintains a tolerogenic state in DC lines. DC2.4 and DC3.2 both release more nitrite when stimulated with LPS or N-

α-Syn which is mitigated by pre-culture with GM-CSF (Figure 3.2). However, the DC2.4 cell line expresses high surface levels of the co-stimulatory molecule CD86 without stimulation which is not decreased by GM-CSF culture (Figure 3.3). This suggests that the cells lines are not in a tolerogenic state and GM-CSF cannot induce a tolerogenic state given the conditions tested. It is possible that a tolerogenic state could be induced with another inducer such as vitamin D, IL-10, or TGFβ, but GM-CSF unto itself cannot induce this state. Since the DC2.4 and DC3.2 cell lines are derived from BMDCs (Shen et al., 1997), but are more mature than the other BMDCs used in these experiments, it is 169

unclear if these results are due to the immortalization of these cells, or because of their more mature state.

BMDCs display a trend to release lower concentrations of nitrite after continuing

GM-CSF culture prior to LPS or N-α-Syn stimulation (Figures 3.5 and 3.8), similar to pretreating DC2.4 and DC3.2 cells with GM-CSF. However, BMDCs exhibit lower surface expression of CD86 (Figure 3.4 and 3.6) compared to DC2.4 cells making them a better model of tolerogenic DCs than DC2.4 cell lines. Interestingly, adding IL-4 during maturation increases the surface expression of CD86 and OX40L and decreases Jag-1 surface expression (Figure 3.4). As a result, IL-4 was omitted from all future experiments because it promotes differentiation of DCs to a more mature state compared to the BMDCs matured without IL-4, in line with prior research (Sallusto and

Lanzavecchia, 1994; Lutz et al., 2000).

In the absence of LPS or N-α-Syn stimulation, BMDCs are in a tolerogenic state as determined by low costimulatory molecules (Figure 3.6), low expression and release of proinflammatory cytokines (Figure 3.7, 3.9), increased expression of IDO, a trend to increased kynurenine (Figure 3.10), and induction of Tregs in culture (Figure

3.11). However, when cultured in media alone prior to stimulation with N-α-Syn, surface expression of co-stimulatory molecules is increased (Figure 3.6), expression and release of proinflammatory cytokines is increased (Figure 3.7, 3.9), and IDO expression and kynurenine is decreased (Figure 3.10), indicating these BMDCs are in a semi-mature state. GM-CSF culture prior to N-α-Syn stimulation does mitigate the surface expression of CD86 and MHC II (Figure 3.6) and increases Jag-1 surface expression. There was a mitigation of the expression and release of some proinflammatory cytokines and chemokines, but most were not affected or were increased by GM-CSF culture (Figure

3.7 and 3.9). IDO expression and kynurenine was not effected by GM-CSF culture prior 170

to N-α-Syn stimulation (Figure 3.10). Additionally, GM-CSF decreases the expression of

CD40 (Figure 3.9), after N-α-Syn stimulation, in line with the decrease in the CD86 surface expression (Figure 3.6). However, GM-CSF does not decrease the expression of maturation markers such as CEBPB and CCR7 or TLR signaling genes such as

CD14, MYD88, and NFKB1 (Figure 3.9). Combined, these data suggest GM-CSF- differentiation of bone marrow to BMDCs induces a tolerogenic state which is maintained with the culture in media and GM-CSF. However, upon stimulation, N-α-Syn stimulation induces a semi-mature state. GM-CSF cannot maintain the tolerogenic state, given that

BMDCs increase the expression and release of proinflammatory cytokines, increased expression of maturation markers, and decreased IDO expression and kynurenine concentration. However, GM-CSF is changing how the BMDCs respond to stimulation with altered expression and release of cytokines, including increased IL-10 expression and release as well as increased Jag-1 surface expression (Figure 3.7, 3.9).

Another characteristic of tolerogenic DCs is their ability to induce Tregs

(Maldonado and von Andrian, 2010). Culture of media-alone BMDCs with or without N-

α-Syn stimulation with CD4+ cells, leads to an increase in the percentage of Tregs

(Figure 3.11). However, this was significantly mitigated by continuing the culture with

GM-CSF, even without N-α-Syn stimulation (Figure 3.11). The resulting CD4+ CD25+ cells do possess Treg suppressive function indicating these are bonafide Tregs. The

Tregs were being induced by a mechanism other than proliferation of CD4+ cells, since

CD3/CD28 beads do not increase the percentage of Tregs to the same degree as

BMDCs. This was borne out by CFSE-labelled CD4+ CD25- and CD4+ CD25+ cells cultured with BMDCs. There is minimal proliferation of Tregs in both populations (Figure

3.13). Since a minority proportion of the Tregs proliferated, the majority of Tregs from the co-culture were induced from the non-Treg CD4+ population. In addition, the 171

induction of Tregs is DCs. This demonstrates that some molecules on the surface of

BMDCs promote Treg induction. Based on correlation analysis, several surface markers that bind to receptors on CD4+ cells do positively correlate with Treg induction, including

MHC II and Jag-1 (Figure 3.12). It is unclear to what degree these, or other co- stimulatory molecules, are responsible for inducing Tregs. While the increase in Jag-1 expression is line with other research demonstrating a role for Jag-1 inducing Tregs

(Gopisetty et al., 2013), the negative correlation with OX40L does not correspond with prior work showing OX40L is involved in Treg induction (Haddad et al., 2016). This may indicate that surface expression of OX40L is sufficient to promote Tregs, or other co- stimulatory molecules, which have not been identified, may be playing a role. It is interesting that the increase in IDO expression, kynurenine release and surface expression of CD39 and CD73 appear to increase in the BMDCs which are strong inducers of Tregs, yet do so through soluble mediators. Since Tregs are still induced in the transwells, there may be a role for these mediators, in addition to the co-stimulatory molecules.

The ability of Tregs generated from BMDC-CD4+ co-culture to be neuroprotective in the MPTP model was also tested. As of now, the mechanism by which Tregs are neuroprotective has not been determined. In Figure 3.11, the ability of

Tregs to suppress proliferation of CFSE-labelled CD4+ CD25- responder T cells is shown. To see if this increase in Treg function correlates with increased neuroprotection, both non-adhered cells and enriched CD4+ CD25+ were transferred into MPTP-intoxicated mice. In both cases, there was no significant neuroprotection

(Figure 3.20). The most probable explanation is that insufficient cell numbers were transferred to detect protection, given that previously the fewest number of Tregs transferred which demonstrated an effect was ~500,000 Tregs (Reynolds et al., 2007). 172

The transfer of a greater number of Tregs into each mouse could test this. Another possibility is that the BMDC-induced Tregs posses less neuroprotective function and even more Tregs will need to be transferred to detect a significant effect.

Our second aim was to determine if the transfer of tolerogenic DCs leads to the same effects as GM-CSF in the MPTP model. Adoptive transfer of BMDCs after MPTP did not lead to protection of dopaminergic (TH+ Nissl+) neurons in the substantia nigra

(Figure 3.14). This is thought to be due to the BMDCs having insufficient time to migrate to the secondary lymph organs and induce Tregs that release mediators and/or migrate to the substantia nigra to protect neurons. To test this, we adoptively transferred

BMDCs prior to MPTP intoxication. In this case, BMDCs protect both the (TH+ Nissl+) neurons in the substantia nigra and the TH+ termini in the striatum (Figure 3.15). Given this BMDC transfer paradigm, the number of reactive microglia is decreased in the substantia nigra (Figure 3.16), indicative of decreased neuroinflammation. Interestingly, of the genes 2-fold increased in MPTP control mice relative to PBS control mice, none were noticeably decreased by the adoptive transfer of BMDCs, suggesting BMDCs were not decreasing the expression of genes upregulated by MPTP (Figure 3.17A). There was an almost 2-fold increase in the expression of Il1RN, which may be part of the mechanism of the decreasing neuroinflammation by suppressing IL-1 receptor signaling.

However, there was an increase in the expression of several genes for which expression was decreased in the MPTP control mice relative to PBS control, including CCL2,

CCL12, RIPK2, and TLR3. Of the genes altered between the MPTP midbrain and the

BMDC + MPTP midbrain, 24 genes were associated with inflammatory response (Figure

3.17B). The adoptive transfer of BMDCs increases the expression of several chemokine-related genes including: CCR1, CCR3, CXCL10, CXCL6, CXCR4, CCL11,

CCL24, CCL25, and CCL2, the gene upregulated in the pathway. It is possible that at 173

least part of the protective mechanism of BMDCs is altering the chemokine balance changing the immune cell infiltration. This is an interesting possibility since GM-CSF treatment also increases the expression of several chemokines, like CCL2 increased here (Kosloski et al., 2013). CCL2 is an interesting chemokine since it is a ligand for

CCR4 (Yoshie and Matsushima, 2015), and CCR4 is the major chemokine receptor on

Tregs (Yi and Zhao, 2007; Chakraborty et al., 2012). It is possible that the increase in

CCL2 is increasing the migration of Tregs into the midbrain, which facilitates neuroprotection. More work will be needed to test this possibility. These data demonstrate that, like GM-CSF, BMDC transfer decreases neuroinflammation and neurodegeneration following MPTP.

GM-CSF increases the percentage of Tregs in mouse spleen without MPTP

(Kosloski et al., 2013). BMDCs were not able to increase Treg percentage or Treg function in vivo in the absence of MPTP (Figure 3.18). However, GM-CSF, used here as a positive control, was unable to increase Tregs to the same degree as published previously (Kosloski et al., 2013). Treg function was also not increased by either GM-

CSF, which was not previously reported, or by BMDC transfer. This result suggests that

GM-CSF may work on another cell type in vivo without inflammation to induce Tregs. It is possible that Tregs express the receptor for GM-CSF (Kared et al., 2008) and therefore GM-CSF may directly promote the proliferation of Tregs. Another possibility is that GM-CSF induces the proliferation of other immune cells such as myeloid-derived suppressor cells (MDSCs) which protect against and promote recovery after neurodegeneration (Saiwai et al., 2013) and induce Tregs (Huang et al., 2006; Serafini et al., 2008). When BMDCs were transferred prior to MPTP intoxication, this increased the percentage of Tregs in the spleen (Figure 3.19). This also increased the number of

Tregs in the spleen. This is in line with prior data, which showed GM-CSF promotes 174

Tregs in vivo during ongoing inflammation in various models of autoimmunity (Sheng et al., 2008; Cheatem et al., 2009; Ganesh et al., 2009). Interestingly, MPTP intoxication, but not the adoptive transfer of BMDCs, increases Treg function. This experiment was performed once and there was overall less inhibition compared to prior functional assays

(Figures 3.11, 3.18). More work would need to be done to verify these results. While most research has not focused on the suppressive function of Tregs, the results here suggest that BMDCs and GM-CSF increase Treg number, but not function. This is contrary to administering sargramostim in Parkinson’s disease does increase Treg function as well as number (Gendelman et al., 2017).

Our next experiment was to determine if BMDCs may be neuroprotective by mechanisms other than inducing Tregs. It is unclear if the mechanism of neuroprotection is mediated though cell-to-cell contact or through the release of anti- inflammatory factors by BMDCs. To test that latter possibility, we cultured the MES23.5 dopaminergic neuron cell line with supernatant from BMDCs, prior to treating cells with

MPP+ or supernatant from BV2 microglia (Figure 3.21). In both cases, BMDC supernatant did not protect the cultured MES23.5 cells from either the neurotoxin MPP+ or the neuroinflammatory mediators from BV2 microglia cell lines. Interestingly, culturing

BV2 microglia in BMDC supernatant did decrease the release of nitrite, IL-6, and TNF

(Figure 3.22). This would indicate that the factors released from BV2 cells that are toxic to MES23.5 cells are not these proinflammatory mediators. However, BMDC supernatant decreased the release of these mediators that reduce neuroinflammation in vivo, thus not leading to neuroprotection. These results suggest that BMDC supernatant may not directly protect dopaminergic neurons in vivo, but may decrease microglia activation, perhaps due to the release of kynurenine (Figure 3.10) or another factor which was not measured. 175

Lastly, I tested if GM-CSF increases the number of myeloid DCs or the tolerogenic state of DCs in the spleen. This would be consistent with myeloid DCs inducing Tregs and provide further evidence that DCs are the cells inducing Tregs.

Here, we found no increase in the percentage of CD11c+ CD11b+ cells in the spleen, nor changes in the expression of surface molecules such as Jag-1, OX40L, MHC II,

CD86, CD39, and CD73 (Figure 3.23). Based on this, GM-CSF alone does not appear to increase the percentage or tolerogenic state of splenic myeloid DCs. It is possible that other surface molecules are increased on splenic DCs that induce Tregs but were not tested here, such as PD-L1. It is also possible that other DC subsets are affected by

GM-CSF and induce Tregs. GM-CSF may act on other immune cell types, like MDSCs, to induce Tregs, or GM-CSF may act on Tregs directly. All are possibilities to be tested further.

In conclusion, GM-CSF promotes a tolerogenic state as part of the differentiation of bone marrow progenitors to immature DCs. However, GM-CSF cannot maintain the tolerogenic state after stimulation with N-α-Syn, but GM-CSF does alter how BMDCs respond to the stimulus. GM-CSF culture with BMDCs diminishes the induction of Tregs in culture. Tolerogenic BMDCs are neuroprotective, decrease neuroinflammation and increase Tregs in the spleen after MPTP intoxication. The neuroprotection may be related to inducing Tregs as wells as decreasing microglia activation, but BMDC supernatant is not directly neuroprotective, at least in cultured MES23.5 cells. GM-CSF is not increasing percentage of myeloid DCs or the expression of surface markers in vivo. This suggests that GM-CSF may act on DCs in vivo to promote Tregs during the

MPTP-induced inflammation. However, GM-CSF may be promoting Tregs via another mechanism other than through DCs. In addition, neuroprotection may be mediated by a mechanism other than Treg induction. 176

CHAPTER FOUR

CYTOKINE ENVIRONMENT IN THE VENTRAL MIDBRAIN AND

CERVICAL LYMPH NODE TWO DAYS AFTER MPTP

INTOXICATION

ABSTRACT

CD4+ T cells are required for neurodegeneration following MPTP intoxication.

The adoptive transfer of Th1 and Th17 T CD4+ cells exacerbates MPTP intoxication, but it is unknown if MPTP intoxication differentiates naive CD4+ cells to these subtypes. To test this, the expression and release of proinflammatory cytokines in the cervical lymph node and ventral midbrain was determined at peak neuroinflammation two days post

MPTP intoxication. Flow cytometry was also used to test the percentage of B cells, T cells and activated APCs in different lymph node populations after MPTP intoxication.

Based on these data, MPTP increases the MFI of MHC II on APCs in the cervical lymph nodes and increases B cells in the brachial and axillary lymph nodes. No changes in the

CD3+ T cell percentage were detected. In the cervical lymph nodes, expression of IFNγ and IL-4 was increased with increased expression of IP-10 in the ventral midbrain.

Gene expression of many proinflammatory and anti-inflammatory genes were increased in the cervical lymph node and CXCL10 in the ventral midbrain. These data found few changes in the expression of proinflammatory cytokines, which may be due to the timing or location of samples tissues. The changes seen may suggest an environment that promotes Th1 or Th2 T CD4+ cells.

177

INTRODUCTION

No rodent PD models exist that fully recapitulates all the features of PD. The

MPTP model which is commonly used to mimic neuroinflammation and dopaminergic neurodegeneration in PD. MPTP readily crosses the blood-brain barrier (BBB), taken into astrocytes, and is converted into the neurotoxic metabolite MPP+ by monoamine oxidase B (MAO-B) (Cohen et al., 1984; Heikkila et al., 1984). MPP+ is then released and specifically taken up through dopamine transporters due to the structural similarity to dopamine (Klein et al., 1985). Once in the dopaminergic neurons, MPP+ can be transported through vesicular monoamine transporter 2 (VMAT2) into intracellular vesicles for storage (Staal and Sonsalla, 2000), resulting in little dopaminergic neuron injury. As a result, VMAT2 expression is thought to play a major role in the susceptibility to MPTP (Lohr et al., 2016). MPP+ can also be transported to mitochondria and block complex I of the electron transport chain (Singer et al., 1988). In doing so, metabolic disruption and loss of ATP synthesis in dopaminergic neurons leading to neuronal death.

Why dopaminergic neurons in the substantia nigra pars compacta are specifically lost while dopaminergic neurons in neighboring brain regions are not is not clear. The difference in MPTP sensitivity may be related to VMAT2 expression, dopamine transporter expression of downstream signaling molecules like JNK and Jun both of which are downregulated in MPTP-resistant mouse strains (Boyd et al., 2007).

As dopaminergic neurons are lost in the substantia nigra, inflammation mediated by activated microglia increases (Członkowska et al., 1996; Kohutnicka et al., 1998).

After MPTP intoxication, microglia exhibit an activated M1 phenotype and release proinflammatory cytokines such as IL-1β as well as reactive oxygen and nitrogen species (Wu et al., 2002). Activated microglia are required for neurodegeneration because MyD88 knockout mice are resistant to MPTP-induced degeneration (Cote et al., 178

2011). MPP+ kills or injures a fraction of the dopaminergic neurons lost; most dopaminergic neurons are lost due to resulting neuroinflammation (Reynolds et al.,

2010; Cote et al., 2011). In addition to the activation of resident microglia, infiltrating macrophages and neutrophils play a role (Cote et al., 2011), showing that the innate immune system is important for neuroinflammation following MPTP intoxication.

In addition to the innate immune response to the MPP+, the adaptive immune response plays a role as well. SCID mice, which lack lymphocytes, are relatively resistant to MPTP-induced neurodegeneration (Benner et al., 2008). More specifically, mice that lack CD4+ T cells do not lose nigral dopaminergic neurons following MPTP intoxication, but mice lacking CD8+ cells are sensitive to MPTP intoxication (Brochard et al., 2009). Interestingly, more CD8+ T cells infiltrate into the substantia nigra than CD4+

T cells in the MPTP model and in PD patients (Kurkowska-Jastrzebska et al., 1999;

Depboylu et al., 2012) and the percentage of CD8+ cells increase and CD4+ T cells decreases after MPTP intoxication (Zhou et al., 2015). Adoptive transfer of effector

CD4+ T cells exacerbates nigal dopaminergic neuron loss after MPTP intoxication

(Reynolds et al., 2007). Exacerbation of MPTP-induced neurodegeneration was greatest by adaptive transfer of N-α-Syn-specific Th17 cells, and to a lesser extent, with

Th1 CD4+ cells, but not Th2 cells (Reynolds et al., 2010; Liu et al., 2017). On the other hand, the adoptive transfer of Tregs were neuroprotective (Reynolds et al., 2007) in the

MPTP model.

MPTP not only alters the number and distribution of innate and adaptive immune cells, release of proinflammatory cytokines and chemokines are increased. In the serum and midbrain after MPTP, there is an increase in TNF-α, IL-6, and IL-1β levels are increased (Kaku et al., 1999; Barcia et al., 2005; Shen et al., 2005; Zhao et al., 2007).

Not only do these cytokines play a role in neurodegeneration, they polarize CD4+ T cells 179

to Th1 or Th17 cells which also play a degenerative role in the MPTP model as indicated above. In addition, antigens from brain, including nitrated α-synuclein, are detected in cervical lymph nodes, the regional draining lymph nodes from the brain (Benner et al.,

2008). Given that the cervical lymph nodes also contain activated APCs, the possibility of an induced adaptive immune response toward this neoantigen exists. To better characterize the early stage of the inflammatory response to MPTP, lymph node populations and the ventral midbrain were isolated to test immune cell numbers by flow cytometric analysis and cytokine expression by PCR array and Luminex arrays. It is hypothesized that MPTP intoxication will increase proinflammatory gene expression in the lymph nodes and the midbrain. The resulting increase in proinflammatory cytokines in these regions would promote differentiation of antigen-specific CD4+ T cells toward

Th1 and Th17 subtypes and neurodegeneration and neuroinflammation.

While CD4+ T cells play a role in MPTP pathology that worsens by Th1 and Th17 cells, whether and to what extent, MPTP activates CD4+ cells in the periphery is unclear.

MPTP increases the concentration of several cytokines in the plasma, including IL-6

(Luo et al., 2004; Shen et al., 2005), TNF-α (Barcia et al., 2005), and to a lesser extent

IL-1β (Barcia et al., 2005; Shen et al., 2005). However, the concentration and expression of cytokines in the cervical lymph nodes has not been determined. In addition, the expression of IL-1β, TNF-α, CCL3, CXCL10 and CCL2 is increased

(Kalkonde et al., 2007; Zhao et al., 2007) in the ventral midbrain and striatum. However, the expression of these cytokines and chemokines is dependent on the time post MPTP intoxication and the brain region. This experiment was undertaken to test the gene expression and cytokine release in the cervical lymph nodes and ventral midbrain.

Because of the importance of the peripheral immune system, we well also tested the 180

percentage of B cells, T cells, and activated antigen presenting cells by flow cytometry in different lymph node populations.

MATERIALS and METHODS

Mice, MPTP intoxication and Isolating lymph nodes and ventral midbrain

Male, approximately 6-week-old C57Bl/6J mice were obtained from Jackson

Laboratories. All studies were conducted in accordance with National Institutes of

Health (NIH) and the Institutional Animal Care and Use Committee (IACUC) of the

University of Nebraska Medical Center (UNMC). All animals had ad libitum access to food and water and were maintained in a 12-hour light/12-hour dark cycle.

Mice were randomized into one of two groups. One was given 4 subcutaneous

(s.c) injections at two hour intervals of 18 mg (free base)/kg MPTP (Sigma-Aldrich). The other group was a control group with s.c. injection of DPBS (Gibco). MPTP was handled and administered in accordance with the published guidelines from the NIH and UNMC

IACUC (Jackson-Lewis and Przedborski, 2007).

After 2 days, 3 mice from MPTP- and DPBS-treated groups were sacrificed by

CO2 asphyxiation and cervical dislocation for flow cytometric analysis. The cervical and deep cervical lymph nodes (cLN), brachial and axillary lymph nodes (b/aLN) and inguinal lymph nodes (iLN) were removed and kept separate for each mouse and were placed in a 70 µM filter. The remaining 3 mice from each treatment group were sacrificed by CO2 asphyxiation and cervical dislocation for RNA and protein extraction. The cLN were removed and placed in a 1.5ml snap-cap tube containing RNAlater (ThermoFisher). The brain was removed, the frontal cortex and cerebellum was dissected and the remaining 181

ventral midbrain (VMB) was placed in a 1.5ml snap-cap tube containing RNAlater.

These samples were kept separate and were kept on ice until tissue homogenization.

The PARIS kit (Ambion, Waltham, MA) was used to homogenize cLN and VMB tissue and isolate protein and RNA. Tissue was removed from RNAlater, blotted on a

Kimwipe, placed in a 1.5 ml tube containing 600 µl cold disruption buffer. Each tissue from each mouse was homogenized separately. Approximately 300 µl of homogenized tissue was pipetted into two 1.5ml tubes, one for protein isolation and the other for RNA isolation as described below.

Flow cytometry

The different lymph node populations were pressed through a 70 µM filter and the cells were concentrated by centrifugation at 300 xg for 10 min. The pellet was resuspended in 1 ml 1x Hanks Buffered saline solution (HBSS) (Gibco) for cell counting.

A volume containing ~500,000 cells was pipetted into 5 ml flow cytometry tubes and cells were concentrated by centrifugation at 300 xg for 10 min. Cells were blocked for

20 min on ice in 10 µg/ml rat gamma globulin in FSB (0.5% bovine serum albumin

(BSA), 0.1% sodium azide in 1x DPBS (Gibco)) to block Fc receptors. Cells were stained with anti-CD11b-PECy7, anti-MHC II-AlexaFluor 700, anti-CD3-PE and anti-

CD19-FITC (eBioscience) for 20 min at 4°C. Cells were washed in FSB and cells were concentrated by centrifugation at 400 xg for 5 min 2 times. Cells were fixed in FACS Fix

(1% formaldehyde in 1x DPBS (Gibco)) at RT for 15 min. Cells were concentrated by centrifugation at 400 xg for 5 min. Cells were resuspended in FSB for analysis using a

LSR II (BD) in the Flow Cytometry Research Facility at UNMC.

Gates were drawn so that ~2% of the isotype control sample were positive for each of the markers. The percentage of the total population positive for CD3 (T cells), 182

CD19 (B cells), and CD11b+MHC II+ (activated APCs) were determined. The MFI was also determined for each marker.

Protein isolation and luminex array

Proteins were isolated from the tissue homogenates as follows. The homogenate was clarified by centrifugation at 10,621 xg in a Beckman (Brea, CA)

1470R centrifuge for 10 min. The supernatant was transferred to a 1.5 ml tube and was stored at -20°C until future analysis.

Total protein concentration was determined in each sample using the bicinchoninic acid assay(BCA) total protein assay (Pierce, Waltham, MA) performed by adding 200 µl of the BCA working solution (50:1 solution A:solution B) to 25 µl of an undiluted or 1:10 dilutions of VMB homogenate or 1:2 dilution of cLN homogenate. The plate was incubated at 37°C for 30 min, allowed to equilibrate to RT for 10 min prior to reading at 562 nm. Based on the normalized absorbance readings and a BSA standard curve, the protein concentration was determined.

The Milipore (Billerica, MA) Milliplex xMAP mouse cytokine and chemokine magnetic bead kit was used to determine the concentration of cytokines and cytokines from each sample as follows. The 96-well plate was washed in wash buffer prior to adding 25 µl of standards, quality control samples and triplicates of each tissue (~250 µg cLN, ~650 µg VMB total protein). To wells with standards and quality controls, 25 µl of

PARIS disruption buffer was added. The samples had 25 µl assay buffer added.

Antibody solution containing antibodies against IFNγ, IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-6,

IL-7, IL-9, IL-10, IL-12p40, IL-12p70, IL-13, Lix, IL-15, IL-17, IP-10, MIP-2, MIG,

RANTES, and TNF-α was added to all wells and the plate was incubated at 4°C with shaking overnight. The beads were concentrated on a magnet and washed in wash 183

buffer 2 times. To each well was added 25 µl of detection antibodies and incubated at

RT for 1 hr. To each well was added 25 µl streptavidin-PE was added to each well and the plate was incubated at RT for 30 min. The beads were then concentrated on a magnet and washed in wash buffer twice before being for resuspended in 150 µl sheath fluid and data was collected using a Millipore Magpix system with Luminex Xponent 4.2 software (Luminex corporation Austin, TX). The software calculated the concentration of each of the cytokines and chemokines in each sample. The mean cytokine and chemokine concentrations were determined for 3 replicates from the 3 mice for each tissue for the DPBS mice. Mean concentrations from MPTP-treated mice were normalized to those of DPBS-treated mice.

RNA isolation, cDNA conversion, PCR arrays and gene expression analysis

RNA was isolated from the cLN and VMB homogenate using the PARIS kit. An equal volume of 2x PARIS lysis/binding buffer (ThermoFisher) (~300 µl) was added to

~300 µl of homogenate and 300 µl 200 proof ethanol and the solution was mixed. This volume was pipetted into a spin column and was passed through at 20,817 xg

(Eppendorf 5417R, Hamburg, Germany) for 1 min. The eluate was discarded and 700 µl

PARIS wash solution 1, was passed through the filter at 20,817 xg for 1 min and the eluate was discarded. For each RNA sample, 500 µl PARIS wash solution 2/3 was passed through the filter at 20,817 xg for 1 min and the eluate was discarded. After drying the filter by centrifugation at 20,817 xg for 1 min, 40 µl of PARIS elution buffer heated to 95°C was added to each column and eluted from the column at 20,817 xg for

1 min. This was repeated with an additional 10 µl PARIS elution buffer passed through at 20,817 xg for 1 min. RNA concentration and A260/A280 was determined by UV specteophotometry (ND-1000, NanoDrop Technologies Inc. Wilmington, DE). Samples were stored at -20°C until cDNA conversion could be performed. 184

To generate cDNA from the RNA template, the RevertAid first strand cDNA synthesis kit (ThermoFisher) was used as follows. A volume of the RNA to yield 80 ng from each of the samples was added to each of five 0.2 ml tubes with enough water to bring the volume to 11 µl, and 1 µl oligo dT primers was added to each sample and incubated at 65°C for 5 min. Eight microliters of master mix (containing 5x reaction buffer, Ribolock Rnase inhibitor, 10 mM dNTPs, and revertAID RT) was added and incubated at 42°C for 60 min for single strand synthesis. The reaction was terminated by incubation at 70°C for 5 min and aliquots were combined and frozen at -20°C until the

PCR arrays were run.

To run the array, the whole volume of 2x RT2 SYBR green master mix (Qiagen)

(1,350 µl), 1,248 µl Rnase-free water, and ~400 ng cDNA from a single sample was mixed with 25 µl was added to each well of a Mouse Inflammatory Cytokines and

Receptors array (Qiagen). The array was performed on an Eppendorf Realplex2 mastercycler gradient S using a PCR program with a 10 min hot start at 95°C, 40 cycles of 15 sec at 95°C, and 60°C for 1 min using a 26% ramp rate. A melting curve was run at the end. Ct values were determined for each well from each sample. Fold change was calculated using the ΔΔCt method. Housekeeping genes were averaged for normalization included for cLN samples are ACTB, GAPDH, GUSB, and HSP90AB1 and for VMB samples ACTB, B2M, GAPDH, GUSB, and HSP90AB1 were averaged for normalization. Fold regulation was -1/fold change.

Statistics

Significance of fold changes for cell percentages or MFIs in MPTP-intoxicated mice compared to DPBS control mice were determined by one-way ANOVA followed by

Sidak’s post hoc test. The fold change between the cytokine in the cLN and VMB was assessed by two-way ANOVA followed by Sidak’s post hoc test comparing the DPBS 185

and MPTP for each tissue for each cytokine. All statistics and graphs were generated by using GraphPad Prism version 6.

RESULTS

Flow cytometry

The purpose of this experiment was to determine the effect of MPTP on the distribution of T cells (CD3+), B cells (CD19+), and activated APCs (APC, CD11b+ MHC

II+) at peak neuroinflammation, once mice were administered 18 mg/kg MPTP or DPBS in 4 s.c. injections, each administered every 2 hours. After 2 days, mice were sacrificed and the cervical lymph nodes (cLN), brachial/axillary (b/aLN), and inguinal lymph nodes

(iLN) were isolated and flow cytometry was performed to determine the percentages of cell types and then normalized to the DPBS controls. In all 3 lymph node populations, there was a trend toward increased activated APCs after MPTP intoxication, but this did not reach statistical significance (Figure 4.1A). As CD11b+ cells upregulate MHC II surface expression when they become activated (Ponomarev et al., 2005), we assessed

MFI on the surface of MHC II on CD11b+ cells. MHC II MFI for MPTP-intoxicated mice was increased compared to DPBS controls, but only reached statistical significance in the cLN (Figure 4.1B). This result is congruent with previous research showing antigens from the brain drain into the cLN population after MPTP intoxication (Benner et al.,

2008). We also found that percentages of CD19+ B cells were significantly greater in b/aLN in the MPTP-intoxicated mice (Figure 4.1C). B cells appeared to be elevated in the iLN of MPTP mice as well, but did not reach significance. Interestingly, MPTP intoxication did not seem to affect CD19+ cell percentages in cLN. Lastly, no significant changes in the percentages of CD3+ T cells, though a trend toward decreased 186

Figure 4.1 Flow cytometric profile in lymph nodes after MPTP intoxication

187

Figure 4.1 Flow cytometric profile in lymph nodes after MPTP intoxication

Cervial (cLN), brachial and axillary (b/aLN) and inguinal lymph nodes (iLN) from DPBS control or MPTP-intoxicated mice were pressed through a cell strainer and the resulting single cells were stained with anti-CD11b, anti-MHC II, anti-CD19 and anti-CD3 antibodies for flow cytometric analysis. The frequency of cell type was divided by the average of the DPBS for each lymph node population to normalize. Black bars are the

DPBS controls and the red bars are from MPTP mice. (A) Fold change in activated antigen presenting cells (CD11b+ MHC II+) in MPTP-intoxicated mice compared to control for all lymph nodes. (B) Fold change in the MHC II mean fluorescent intensity on

CD11b+ cells was compared for MPTP-intoxicated LN compared to DPBS LN. (C) Fold change in B cells (CD19+) in MPTP-intoxicated mice compared to control for all lymph nodes. (D) Fold change in T cells (CD3+) in MPTP-intoxicated mice compared to control for all lymph nodes. Mean ±SEM for n=3 mice per group and compared by one-way

ANOVA and Sidak’s post-hoc test whereby a-p ≤ 0.05 compared to DPBS-treated controls. n.s. = not significant.

188

percentages were found in all lymph nodes (Figure 4.1D) which may be due to MPTP- induced cell death (Chi et al., 1992; Zhou et al., 2015). These data demonstrate that

MPTP-intoxication induces changes to the peripheral immune compartment including increased activated APCs and B cells.

Cytokine and chemokine expression in the cLN and VMB after MPTP

To determine the cytokine environment at the peak of neuroinflammation post

MPTP, ventral midbrain (VMB) and cLN were homogenized and proteins extracted 2 days after MPTP intoxication. The extracted proteins were tested using a Millipore xMAP magnetic bead array for inflammatory cytokines and chemokines. MPTP- intoxicated mice show slight, though not significant, elevated levels of IFNγ, IL1-β, IL-4,

IL-5, IL-7, IL-9, IL-10, IL-12p40, IL-12p70, IL-13, Lix, IL-15, IL-17, IP-10, MIP-2,

RANTES, and TNF-α, but only IFNγ and IL-4 reached statistical significance in cLN

(Figure 4.2A). In the ventral midbrain (VMB), only IP-10 was significantly elevated following MPTP (Figure 4.2B). These were unexpected results since MPTP is known to induce proinflammatory cytokines such as IL-6 and TNF-α (Luo et al., 2004; Barcia et al.,

2005; Shen et al., 2005; Kalkonde et al., 2007; Zhao et al., 2007).

To test if MPTP increases the expression of proinflammatory cytokine and chemokine genes, RNA was extracted from VMB and cLN homogenate, RNA was converted to cDNA, and real time PCR was performed using an array for proinflammatory cytokines and chemokines and their receptors. Genes increased or decreased by more than 2-fold in the cLN (Figure 4.3A) or VMB (Figure 4.3B) are listed.

The only 2 genes with more than 10-fold increase in expression in the cLN were IL2RG and BMP2. Expression of proinflammatory cytokines such as IL17A, TNF, IL1A, and

IL17B were increased, as were the anti-inflammatory gene expression is increased 189

Figure 4.2 Changes in cytokines and chemokines in the cervical lymph nodes (cLN) and ventral midbrain (VMB) after MPTP intoxication

190

Figure 4.2 Changes in cytokines and chemokines in the cervical lymph nodes

(cLN) and ventral midbrain (VMB) after MPTP intoxication

Homogenates of cLN and VMB were made and proteins extracted using the PARIS kit for Milliplex xMAP analysis for inflammatory cytokines and chemokines. (A) The average from each cLN sample was divided by the average of the DPBS control to determine the fold increase for each cytokine or chemokine. (B) The average from each

VMB sample was divided by the average of the DPBS control to determine the fold increase for each cytokine or chemokine. Means ± SEM for n=3 per treatment group were determined by two-way ANOVA and Sidak’s post hoc test whereby a- p ≤ 0.05 compared to DPBS-treated controls.

191

including IL33, IL5 and IL4, and the chemokines CCL1, CCL12, CCL3, CCL24, CXCL9,

CCL17, CCL11, and CCL5. Interestingly, despite significant increase of IFNγ protein in the cLN, there is a more than 2-fold downregulation of IFNG expression. Several cytokine and chemokine receptors were downregulated, including CCR4, IL5RA, IL6RA,

CXCR5, IL10RB, IL1R1, IL10RA, CXCR3, CCR6, IL2RB, CCR3, CCR5, CCR1, CCR10, and CCR2. Cytokine genes such as IL21, IL16, IL11, IL27, CSF1, CSF3, IL13, and IL15 were also downregulated. To visualize the connections between changed genes in cLN after MPTP intoxication, Ingenuity Pathway Analysis was performed. Figure 4.4 is a pathway of the cytokines and chemokines altered in this experiment. As noted above, there is an increase and decrease in the expression of pro- and anti-inflammatory cytokines. For example, Th2 helper T cell-related cytokines IL5, IL4 were increased, but

IL13 was decreased. Th17 helper T cell-related IL17A was increased, while IL21 is decreased. The surface marker FASLG was increased, but the co-stimulatory molecule

CD40LG was decreased. These data show that MPTP induces many gene expression changes that potentially could affect many different aspects of the interplay of innate and the adaptive immunity toward the induction of responses to nitrated and misfolded proteins such as N-α-Syn.

The only gene with more than 2-fold changed expression in the VMB after MPTP intoxication was CXCL10, which is the gene for IP-10 (Figure 4.3B), also significantly upregulated in Figure 4.2B. The data also point to a paucity of genes in the midbrain that were not affected by 2 days after MPTP intoxication, the time of peak inflammation in this model.

192

Figure 4.3 Changes in gene expression following MPTP intoxication in the cLN and VMB 193

Figure 4.3 Changes in gene expression following MPTP intoxication in the cLN and VMB

Homogenates from cLN and VMN had RNA extracted using the PARIS kit prior to cDNA conversion, and PCR using an array for pro-inflammatory cytokines, chemokines and receptors. All changes reported are greater than 2-fold in the cLN (A) or in the VMB (B) sampled 2 days following MPTP intoxication and relative to DPBS controls. Red cells are more than 2-fold upregulated expression, while green cells are more than 2-fold decreases in gene expression. There are 3 samples for each group of mice treated with

MPTP or DPBS.

194

Figure 4.4 Map of dysregulated genes in the cLN following MPTP intoxication

195

Figure 4.4 Map of dysregulated genes in the cLN following MPTP intoxication

Ingenuity Pathway Analysis was used to map the connections of all genes that were up- or downregulated by at least 2-fold in the cLN assessed 2 days after MPTP intoxication.

Cells are labelled in black with a thicker boarder. Cytokines in the pathway that were not disregulated are outlined in gray. Genes upregulated have a red outline and the darker fill color signifies higher upregulation. Downregulated genes are depicted with green outlines and the darker the fill color, the greater the downregulation.

196

DISCUSSION

Data from this chapter have determined if there were changes to the distribution of T cells, B cells, and activated APCs in different lymph node populations. Our findings showed that there was no significant change in the percentage of activated APCs in the

MPTP lymph node compared to control. There was an increase in the MFI of MHC II on

APCs, but only in the cervical lymph nodes. This suggests an increase in activated

APCs in the draining lymph nodes from the brain, which is in line with previous data

(Benner et al., 2008). Little work has been published on the percentage of myeloid cells in the periphery after MPTP intoxication, especially at this early time point. These data would suggest that MPTP does not induce the proliferation or migration of APCs into lymph nodes, but activates the existing APCs. This chapter also showed an increase in

B cells post MPTP. This is in contrast with other studies which indicated that MPTP does not increase the percentage of B cells (Chi et al., 1992). However, this prior report was testing B cell percentage in the spleen one week after intoxication. It is possible that B cell percentage returns to basal levels by this time point and/or there is no change in the percentage of B cells in the spleen. Lastly, there was no change in the percentage of CD3+ T cells in the MPTP-intoxicated mouse lymph nodes, which is also in contrast to previous data (Chi et al., 1992). Prior research demonstrated that MPTP decreased CD4+ CD3+ cells, but increased CD8+ CD3+ cells (Zhou et al., 2015). In light of this result, it is possible this experiment may not detect an overall change in

CD3+ cells, even though the ratio of CD4+ to CD8+ cells may be altered. Future work could resolve this issue by more specifically testing the percentage of different T cell types. The trend toward decreased CD3+ T cells in lymph nodes due may also be explained by increased mobilization of these cells into the periphery, and possibly the brain. 197

Some care is needed in the interpretation of these results. Each group contains only 3 mice. While this is sufficient for performing statistical analysis, there is insufficient power to find statistically significant differences. This experiment also examined only single time point. While day 2 is the peak of neuroinflammation, the initiation of the adaptive immune response may begin prior to this time point. As a result, future work should look at a time course before or after day 2 to see the rise, peak and diminution of the adaptive immune response. It should also be noted we only examined in the lymph nodes, and not the spleen, the circulating blood, or the brain. In conclusion, we do not yet possess a full picture of how many of these immune cells are present and where they are located at each time point during the course of MPTP intoxication.

We next examined the cytokine and the gene expression of cytokines in the draining lymph nodes and the ventral midbrain. In the cervical lymph node, only IFNγ and IL-4 were significantly increased on the protein level. This is of interest because the increase in IFNγ has been reported previously in the serum and spleen (Huang et al.,

2014; Zhou et al., 2015). However, IL-4 was not increased in previous studies (Huang et al., 2014). While we did not test the release of cytokines in other lymph node populations, it is possible that the increase in IL-4 may be responsible for the increase in

B cells because IL-4 is produced by Th2 CD4+ T cells which support the proliferation of

B cells. In the VMB, only the chemokine IP-10 was significantly increased. It is unclear why this experiment did not find the increase of other proinflammatory cytokines found in other studies. One possibility is the low number of mice. Another possibility is that the wrong brain region was tested at the wrong time. TNF-α and IL-1β were found to be elevated in the striatum, but not the substantia nigra at 1 day post MPTP, but this increase was resolved by day 3 post MPTP (Kalkonde et al., 2007). A prior experiment tested changes in the genes encoding TNF-α, IL-1β, IL-1α, IL-6, MMP-2, MMP-9 at 198

different time points post MPTP in the ventral midbrain and striatum (Hébert et al.,

2003). This experiment reported different time courses for each gene and for each tissue. This suggests it is important to evaluate over time, and in different tissues, because the peak expression is different for each gene in each tisuue. Future experiments should examine both the ventral midbrain and the striatum at earlier and later time points. If this is done, the MPTP-induced increase in the expression of the proinflammatory cytokines would be better characterized.

We also tested mRNA expression of proinflammatory cytokines and chemokines.

In the cervical lymph node, the expression of IL4 was also increased on the mRNA level to coincide with the protein expression. It is interesting that IFNG gene expression was decreased while the concentration of IFNγ was increased. It is unclear why this is the case, but this could indicate the beginning of a phenotypic shift from the initial proinflammatory response to a compensatory anti-inflammatory response. In the ventral midbrain, there was also only a 2-fold increase in CXCL10, and expression of the gene for IP-10 was increased. Again, it is unclear why there was not more dysregulation of genes in the ventral midbrain, but as above, isolation of RNA in both the striatum and the

VMB at earlier time points would be expected to detect increases in gene expression.

In conclusion, this chapter contains data that demonstrates changes to the peripheral immune system and to the RNA and protein expression of proinflammatory cytokines and chemokines at the peak of neuroinflammation. Because of this single time point, it is possible, that there would be more changes at an earlier time point post

MPTP intoxication.

199

CHAPTER 5

DISCUSSION AND FUTURE DIRECTIONS

DISCUSSION

GM-CSF as a pro- and anti-inflammatory cytokine

Previously, GM-CSF has been thought of as a proinflammatory cytokine. This is due to GM-CSF promoting differentiation, proliferation, and survival of myeloid origin cells such as DCs, macrophages and granulocytes. The use of GM-CSF as an adjuvant improves the immune response to vaccines (Jones et al., 1996; Ryan et al., 2000;

Cruciani et al., 2007; Parmiani et al., 2007; Spitler et al., 2009; Garcia et al., 2014) and in the experimental autoimmune encephalitis (EAE) model of multiple sclerosis (MS),

GM-CSF-secreting cells are responsible for the progression of symptoms (Codarri et al.,

2011). Genetic knockout of GM-CSF prevented EAE in one model (King et al., 2009), and collagen-induced arthritis (Campbell et al., 1998) further suggesting that GM-CSF is required for disease in some models of autoimmunity. However, GM-CSF can also promote a tolerogenic state in DCs (Xu et al., 2008; Ganesh et al., 2009; Bhattacharya et al., 2011), where the immune response to antigens is decreased. The divergent effects of GM-CSF are reported in the results of Chapter 2 where sargramostim treatment increased the expression of both pro- and anti-inflammatory genes in the

CD4+ CD25- cells of PD patients. To date, no molecular mechanism has been identified to determine when GM-CSF boosts the immune response and when GM-CSF diminishes the immune response. However, there are several possible explanations.

One possibility is that the concentration, duration of treatment, or the route of 200

administration of GM-CSF determines the effect. To reconstitute the immune system post-chemotherapy, sargramostim is administered i.v. at a dose of 250 µg/m2 per day for 2 weeks (Nemunaitis et al., 1991; Rowe et al., 1995; Waller, 2007). When sargramostim has been tested as an adjuvant, the dose used is variable but range from

20 to 500 µg, usually delivered s.c. (Cruciani et al., 2007; Parmiani et al., 2007). In

Crohn’s disease (Korzenik et al., 2005; Valentine et al., 2009) and in PD (Gendelman et al., 2017) sargramostim was administered s.c. at 6 µg/kg. Because the dose of sargramostim used as an adjuvant may be lower than that used to reduce inflammation, there may be different effects depending on the concentration. If this possibility is correct, it would suggest that a higher dose of sargramostim is needed to lead to decreased inflammation and induce Tregs, which would also suggest a different, potentially lower affinity pathway compared to the pathways which differentiate and proliferate myeloid cells. Another possibility is the length of administration which tends to be shorter when sargramostim is used to repopulate the immune system and to act as an adjuvant than when sargramostim is used to reduce symptoms of Crohn’s disease or

PD. This would suggest that the longer sargramostim is administered, the more it reduces inflammation. Perhaps this is a compensatory mechanism to ensure that the increase in immune cells does not lead to autoimmune response. Lastly, post- chemotherapy, sargramostim is administered i.v., but as an adjuvant or to reduce inflammation, sargramostim is administered s.c. Perhaps the route by which sargramostim is administered may also effect the pharmacokinetics, pharmacodynamics, or cells affected. This may also change the action and effect of sargramostim. Combined, the dose, duration of treatment, or route of administration may be responsible for the divergent effects of sargramostim. 201

GM-CSF may also have effects not related to dosing. For example, there may be a single GM-CSF-induced signaling cascade in all immune cells, but there may be distinct subpopulations or activation states of immune cells which favor either immunogenic or tolerogenic responses. Other factors may control the relative abundance and distribution of these immunogenic or tolerogenic subpopulations. GM-

CSF will then expand all myeloid cells, with the resulting immune response either being immunogenic or tolerogenic due to the relative ratio of the immune cell subtypes and/or activation states. As technology improves, we are gaining more of an appreciation that genetic subpopulations exist within cells with the same surface markers. This was highlighted by a recent single cell RNA-Seq study of blood myeloid cells (Villani et al.,

2017).

Another possibility is that GM-CSF binding its receptor leads to different intracellular pathways that are activated under different circumstances. After binding its receptor, tyrosine kinases, especially JAK2, phosphorylate intracellular tyrosines in the alpha and beta chains of the GM-CSF receptor. STATs, especially STAT5, bind to the phosphorylated tyrosines which can activate MAP kinases, such as ERK1/2, p38, and

JNK, and PI3K pathways (de Groot et al., 1998). Activation of the MAPK pathway is more associated with proliferation and differentiation of cells while the PI3K pathway is more associated with the pro-survival function of GM-CSF. There may be a similar effect between immunogenic and tolerogenic functions where different signaling cascades induce different effects. To date, this possibility has not been investigated. It would also be interesting to determine if different tyrosine kinases, phosphorylation of different tyrosines, and/or different STATs binding to different tyrosines promote different cascades that promote the expression of different cytokines. This could be tested by 202

activating DCs and using inhibitors and/or overexpression of different signaling pathways to test the resulting expression of cytokines.

A final possibility is that other environmental factors bias the state of DCs which bias the effect of GM-CSF binding its receptor. Other factors may be biasing the distribution of different immune subpopulations or may be activating different pathways within target cells. If transcription factors are already increased from other signaling cascades, GM-CSF signals may be biased for those pathways that may promote different immune responses. More work is needed to differentiate these possibilities and the ultimate effect of GM-CSF treatment on the immune system. This is important work if GM-CSF will be used for therapy to ensure that the proper immune response is achieved for the desired outcome.

GM-CSF induces tolerogenic DCs that induce Tregs

The data presented in Chapters 2 and 3 support the model that GM-CSF modulates the immune response in DCs. These GM-CSF-treated DCs in turn bias the activation state of CD4+ cells toward anti-inflammatory Tregs which can then be neuroprotective in PD and the MPTP model of PD. The induction of Tregs is mediated by the surface expression of anti-inflammatory co-stimulatory molecules and not by soluble mediators.

In Chapter 2, sargramostim treatment significantly altered the expression of a wide variety of genes in CD4+ CD25- cells in PD patients. Because the alpha chain of the GM-CSF receptor is not detected on CD3+ cells (Santoli et al., 1988; Rosas et al.,

2007), it is unclear how sargramostim (human recombinant GM-CSF) is inducing the dysregulation of these genes. It has yet to be determined if CD4+ CD25- cells express the beta chain of the GM-CSF receptor, but if they do then it is possible that the 203

exogenous GM-CSF is at high enough concentration to bind to this receptor and directly alter gene expression. Another possibility is that GM-CSF is binding to its receptor on myeloid cells (i.e. macrophages, monocytes and dendritic cells, neutrophils, and eosinophils), activating these cells to release cytokines, chemokines, and/or other mediators which are acting on CD4+ CD25- cells. These possibilities are not mutually exclusive. We also do not know whether the dysregulation of genes is occurring in the same or different CD4+ CD25- populations. It is possible that sargramostim-treated patients contain different CD4+ CD25- populations in which some display increased proinflammatory genes, and another population with increased anti-inflammatory genes.

Conversely, sargramostim may be increasing the expression of pro- and anti- inflammatory cytokines in all CD4+ CD25- cells. The first possibility would suggest that the global administration of sargramostim is affecting diverse cell types in different ways.

The second possibility would suggest that while sargramostim has a uniform effect on

CD4+ CD25-, the dysregulation is due to broad activation of CD4+ CD25- cells.

In Chapter 3, in the absence of stimulation, BMDCs are in a tolerogenic state.

While GM-CSF did not maintain this state after stimulation as determined by the increase in the surface co-stimulatory molecules, increase and proinflammatory cytokines, and decrease in kynurenine, GM-CSF alters how BMDCs respond to N-α-Syn stimulation. Extending GM-CSF culture increased Jag-1 surface expression, the expression and release of IL-10, and decreased the surface expression of MHC II and

CD86. However, GM-CSF generally did not decrease the expression and release of proinflammatory cytokines and chemokines. It is unclear how GM-CSF culture leads to these seemingly divergent effects of increasing anti-inflammatory and proinflammatory co-stimulatory molecules and cytokines. The binding of GM-CSF to its receptor, JAK2 activates STAT5 which can activate several downstream signaling cascades including 204

MAPK and PI3K (de Groot et al., 1998). Tolerogenic DCs display a unique pattern of gene expression separate from mature, immunogenic DCs. Expectedly, tolerogenic

DCs exhibit decreased expression of proinflammatory cytokines and increased expression of anti-inflammatory cytokines (Torres-Aguilar et al., 2010; Lee et al., 2016).

In addition, tolerogenic DCs may increase expression of IDO, an enzyme involved in tryptophan metabolism and associated with the induction of Tregs (Steinman et al.,

2003; Maldonado and von Andrian, 2010; Li and Shi, 2015; Raker et al., 2015).

Tolerogenic DCs also increase the surface expression of PD-1, CD95L, ILT3 and ILT4, all of which are involved in the suppression of immune cell activation or induction of

Tregs, or apoptosis (Sim et al., 2016). Also, CLIP, the invariant MHC II chain, was increased on the surface of tolerogenic DCs which suggest a decrease in the presentation of antigens which could explain why tolerogenic DCs are weaker inducers of helper T cell proliferation (Torres-Aguilar et al., 2010). Tolerogenic DCs also show decreased glycolysis (also detected in activated mature DCs) and increased oxidative phosphorylation (Sim et al., 2016). This switch in energy production may be mediated by a suppression of AKT signaling in tolerogenic DCs. Tolerogenic DCs increase expression of genes involved in lipid metabolism, apoptosis, cell membrane homeostasis, and the inflammatory response, including an increase in NFKB1 (Lee et al., 2016). Despite the plethora of information about gene signatures of tolerogenic DCs, how GM-CSF is inducing and maintaining a tolerogenic state remains unclear.

In these experiments, we identified two anti-inflammatory mediators that are upregulated by continued culture in GM-CSF: Jag-1 and IL-10. We tested the surface expression of Jag-1, and not the total gene and protein expression, so whether GM-CSF increases gene expression and translation of Jag-1 or merely mobilizes the Jag-1 to the surface is unknown in this model. Cultured monocytes do increase jagged1 expression 205

after culture with GM-CSF (Nomaguchi et al., 2001), so the possibility remains in our system that total Jag-1 is increased after GM-CSF culture. How binding of GM-CSF to its receptor is upregulating the surface expression of Jag-1 remains to be determined.

TNFα treatment increases jagged1 expression (Okada et al., 2016). In this system, and the only source of TNFα would be from the BMDCs, but the continued culture of BMDCs in GM-CSF decreased the expression of Tnf (Figure 3.7) and did not change the release of TNFα (Figure 3.9), suggesting increases in Jag-1 surface expression are not related to TNFα. Macrophage colony stimulating factor (M-CSF) regulates Jag-1 expression through NF-κB (Brach et al., 1991), but the expression of Nfkb1 was not altered by continued GM-CSF (Figure 3.7). These results suggest that GM-CSF has another mechanism of increasing jagged1 through a yet undescribed pathway.

IL-10 is an anti-inflammatory cytokine important for the anti-inflammatory functions of Tregs and Th2 helper T cells. More is known about the regulation of IL-10 expression, though the regulatory effects on IL-10 by GM-CSF is unclear. Several transcription factors have been identified that regulate IL-10 gene expression, including

Sp1, Sp3, STAT3, ERK1/2, JUNK, p38, NF-κB, c/EBPα, and AP-1/c-Jun (Tone et al.,

2000; Ziegler-Heitbrock et al., 2003; Wang et al., 2005; Saraiva and O'Garra, 2010).

Whether any are increased by the binding of GM-CSF to its receptor is unknown.

Combined, the lack of knowledge about the molecular events by which GM-CSF induces and maintains a tolerogenic state provides potential future directions. If a tolerogenic state can be induced more directly and maintained without the off-target effects of GM-

CSF, could provide a putative novel therapy for PD and autoimmunity. Turning off the tolerogenic state of DCs could also serve as a novel therapy for anti-tumor immune responses.

Impact of GM-CSF on other cell types 206

GM-CSF has broad effects throughout the body. As a result, peripheral administration of GM-CSF may afford beneficial effects in PD and in models apart from the increase in the percentage and function of Tregs. As stated in Chapter 1, GM-CSF has broad effects throughout the immune system including macrophages, monocytes,

DCs, neutrophils and eosinophils. GM-CSF also induces the proliferation of myeloid- derived suppressor cells (MDSCs). These cells are either monocyte or granulocyte lineage and induced by proinflammatory stimuli including infections (Gabrilovich and

Nagaraj, 2009; Parker et al., 2015). Activated MDSCs increase the expression of reactive oxygen and nitrogen species (ROS/RNS), TGFβ, and PD-L1 as well as induce

Treg differentiation all of which suppress T cell immune responses. GM-CSF administration expands MDSCs in vitro and in vivo (Filipazzi et al., 2007; Dolcetti et al.,

2010; Rosborough et al., 2012). The adoptive transfer of MDSCs reduces the expression of perforin and IFNγ in CD3+ cells in cancer (Filipazzi et al., 2007). MDSCs are also protective in the EAE model of MS, and MDSCs from MS patients suppress

CD4+ T cell proliferation (Ioannou et al., 2012).

The role of MDSCs in PD and the MPTP model of PD has not been tested.

Possibly in PD, MDSCs could be decreased in percentage and/or function, like Tregs. If this is the case, the decrease may contribute to increased neuroinflammation of PD patients. During inflammation, MDSCs may act similar to tolerogenic DCs to induce

Tregs or directly suppress T cell or myeloid effector cells. This potential dual role for

MDSCs merits further investigation, in PD, PD models, and other autoimmune disorders.

Another potential immune cell that may be affected by GM-CSF are B cells.

These lymphocytes have unique roles for producing immunoglobulins and presenting antigens. In PD, frequency of circulating B cells are deceased compared to controls

(Bas et al., 2001; Niwa et al., 2012; Stevens et al., 2012; Horvath and Ritz, 2015). To 207

date, the percentage of different B cell subsets or plasma cells remain undefined.

However, since the production of antibodies reactive to α-Syn is elevated (Papachroni et al., 2007), antigen-specific B cells may play a significant role in the pathology of PD is unclear.

B cells express the receptor for GM-CSF, and GM-CSF promotes the survival of

B cells (Harris et al., 2000). While the effects of GM-CSF on B cells of PD patients have not yet been determined, several potential beneficial effects exist. For example, GM-

CSF may promote the survival of B cells that produce immunoglobulins to clear misfolded antigens and protect in PD. Another possibility is that GM-CSF may modulate the presentation of antigens by MHC II to helper T cells and modulate the response to be more protective in PD. Since naïve B cells can induce Tregs in vitro (Reichardt et al.,

2007), GM-CSF may possibly have similar effects on the B cells as on myeloid cells by increasing the expression of anti-inflammatory mediators and/or decreasing the expression of co-stimulatory molecules; the combination of which would promote the induction of Tregs. A final possibility is that GM-CSF increases numbers and function of regulatory B cells (Bregs), a B cell subset that can suppress immune cells through IL-10- dependent and independent mechanisms (Ray and Dittel, 2017). While a role for Bregs in PD has yet to be identified, these regulatory cells, like Tregs and MDSCs, could play a role in suppressing the inflammation and may comprise a component of the mechanism for protection provided by GM-CSF.

Apart from its effects on the immune system, GM-CSF can have direct effects on the nervous system. GM-CSF receptor is detected on neurons (Kruger et al., 2007) and neural progenitor stem cells (Kim et al., 2004; Kruger et al., 2007). For neural progenitor stem cells, administration of GM-CSF decreases differentiation of these stem cells into astrocytes or neurons and protects them against apoptosis (Kim et al., 2004). GM-CSF 208

treatment promotes axon regeneration in culture by inducing production of brain-derived neurotropic factor (BDNF) (Bouhy et al., 2006). In models of stroke, penumbral neurons around the infarct increase the expression of the GM-CSF-receptor and administration of

GM-CSF induces the expression of the pro-survival factor Bcl (Schabitz et al., 2008). In fact, GM-CSF treatment decreases post-stroke infarct size. GM-CSF also protects the

PC12 neuron cell line from MPP+-mediated toxicity by the restoration of the Bcl-1/Bax ratios, which promotes survival (Kim et al., 2009). As a result, the protection of TH- positive neurons in the substantia nigra by intraperitoneal administration of GM-CSF

(Kim et al., 2009; Kosloski et al., 2013) may not be completely mediated by Tregs, but rather by direct effects on neurons. Given that GM-CSF crosses the blood-brain barrier

(Schabitz et al., 2008), it may indeed elicit direct effects in the CNS.

In addition to neurons, the glia including astrocytes, oligodendrocytes and microglia also express GM-CSF receptors (Baldwin et al., 1993; Sawada et al., 1993;

Lee et al., 1994; Guillemin et al., 1996). The precise in vivo role that GM-CSF plays in the brain remains unclear, but in culture, astrocytes, oligodendrocytes and microglia proliferate in response to GM-CSF treatment (Baldwin et al., 1993; Lee et al., 1994;

Guillemin et al., 1996). GM-CSF is increased in the cerebrospinal fluid of stroke patients

(Tarkowski et al., 1997), and cultured astrocytes produce GM-CSF after stimulation with viral infection, or stimulation with TNF-α or LPS (Wesselingh et al., 1990; Frei et al.,

1992); all of which suggests a role for GM-CSF in response to inflammatory stimuli in the brain. Seemingly the expression of GM-CSF appears to be protective by promoting survival. In fact, GM-CSF plays a role in the normal development of the brain, as evidenced by decreased locomotor activity and spatial memory of GM-CSF-knockout mice, which may be related to the decreased length of dendrites (Krieger et al., 2012). 209

In conclusion, GM-CSF has direct, positive effects on the brain, however, to what extent that these effects are independent of Tregs remains unclear.

A final, related possibility is that GM-CSF is acts on cells in the periphery and lead to release of more products that can enter the brain and affect PD. While BMDC supernatant was unable to decrease the MPP+- or BV2-supernatatnt-mediated killing of

MES23.5 cells (Figure 2.21), BMDC supernatant did decrease the release of pro- inflammatory cytokines from stimulated BV2 cells (Figure 2.22). The possibility has not been ruled out that supernatants from DCs could go to the brain and either decrease the activation state of activated microglia or protect dopaminergic neurons.

FUTURE DIRECTIONS

While the work done in Chapter 2 was performed on a set of patient samples gathered for a specific study, this research could be continued with several potential studies. First, this project examined gene expression in non-Treg, non-effetcor helper T cells (CD4+ CD25-). This cell type was chosen because the identity of the T cells relevant to disease progression or protection in PD were unknown. However, by excluding activated (CD4+ CD127+ CD25+) effector T cells, it is possible that we excluded terminally differentiated T cells and addressed the effects of sargramostim on

T cells that could differentiate into Tregs of effector T cells. Future work could be undertaken to examine the expression of proinflammatory cytokines in these cells. In our current study, the expression of 84 genes were included in the array, thus continuation of this work using RNA-Seq or other technologies could examine the total transcriptome of the isolated population or in single cells. The advent of single cell gene expression allows the grouping of T cells into subpopulations by gene expression, and 210

would allow for measure of unique helper T cell subsets that comprise the repertoire in

PD patients and healthy controls, or allow tracking the distribution of these subsets over time. Lastly, no work has identified a population of T cells that are reactive to post- translationally modified α-Syn. For instance, if APCs (be they monocytes, monocyte- derived macrophages or DCs or B cells) were co-cultured in the presence of modified α-

Syn antigens and T cells from PD or control patients, a proliferating helper T cell population in response to the modified antigens could be identified. These further studies could vastly improve our understanding of T cell gene expression and reactivity to self-antigens such as α-Syn.

The focus of Chapter 2 is on T cells and their function in PD. However, other immune cells (for example neutrophils, or CD8+ cells) may be facilitating the neuroinflammation and neurodegeneration in PD. While some work in the clinical trial evaluated numbers and percentages of these cells in the periphery, no functional assessment of these cell types were performed. Because eosinophil numbers and percentages in the blood were increased after sargramostim treatment (data not shown), whether the function of those eosinophils are affected would be of particular interest.

This may not play a role in the development of PD or the clinical benefit in sargramostim, but these cells may be playing a role in the adverse effects of sargramostim treatment.

As a result, understanding the effects of sargramostim on these cells may lead to a better understand of possible adverse effects and procedures to mitigate those effects if sargramostim is to be included as a PD therapeutic.

This clinical trial did not measure numbers or the percentages of MDSCs. To date, no work has measured the number or function of MDSCs in PD. However, like

Tregs, these cells may be found in decreased number and function and contribute to the unharnessed inflammation in PD patients. Given that GM-CSF can expand MDSCs, the 211

observed improvement in motor function may be related to decreased inflammation related, at least in part, to these cells. These studies would highlight the function and to what extent MDSCs play a role in PD and whether increasing their number and function would improve PD symptoms

As described in Chapter 3, we tested the hypothesis that GM-CSF induces a tolerogenic state in DCs, which induce Tregs and/or are neuroprotective in the MPTP model. This research could be continued with several different research projects. First, while I examined all the cells from BMDC cultures (both adherent and non-adherent

CD11c+ and CD11c- cells), other published experiments with BMDCs isolated CD11c+ cells from only the non-adherent population. Further experiments could further test the tolerogenic markers and functions of the non-adherent CD11c+ cells compared to the total population. Also, flow cytometric analysis could be used to further characterize other DC markers and other markers for co-stimulatory molecules or functional markers that better describe the function of these cells. Recently, DCs have been characterized, not by surface markers, but by ontology. Within the bone marrow, 2 progenitors give rise to CD11c+ cells, the common DC precursor and the common monocyte progenitor

(Helft et al., 2015). While resulting DCs are all CD11c+ MHC II+, but the ontology determines whether the resulting cells are DCs and macrophages. Future experiments could better determine to what extent the BMDCs used here are macrophages or DCs.

More work is needed to determine the mechanism(s) of BMDC-mediated induction of Tregs. First, we should examine the surface expression of other co- stimulatory molecules that induce Tregs, such as PD-L1 should be examined after continued culture with GM-CSF. Next, blockade of OX40L, Jag-1, PD-L1, and others with antibodies or antagonists should be evaluated to determine whether induction of

Tregs during the co-culture of BMDCs with CD4+ T cells could be inhibited. We should 212

also begin to test the expression of these molecules on BMDCs generated in culture with inhibitors or agonists for different components of GM-CSF signaling cascade to determine what is important pathways for the induction of surface markers. Tregs from

BMDC co-cultures should also be better characterized with other Treg markers like

CD39, CD73, Lag-3, ICOS, PD-1, CTLA-4, Helios, GITR, as well as gene expression for cytokines such as IL-35, TGFβ and IL-10. Determining increased expression in this population may indicate which Treg functions are needed for suppression. This could be tested by knocking out or blocking some specific factors to determine whether Tregs are still suppressive in the proliferation suppression assay. Lastly, Tregs derived from co- culture should be further tested for their ability to protect dopaminergic neurons after

MPTP intoxication. We could use the induced Tregs to test which suppressive function(s) are important for neuroprotection, however, the efficiency of in vitro induction would need to be improved to provide sufficient numbers of Tregs for these studies.

In vivo, more experiments are needed to determine to what extent GM-CSF induces tolerogenic DCs or other regulatory cells, such as MDSCs. This could be achieved by determining the peripheral percentages of these cell types in peripheral blood after injection of GM-CSF. In vivo blocking of co-stimulatory molecules identified above can determine the role of Treg-inducing myeloid cell types. This is important to determine whether injection of GM-CSF working only through the tolerogenic DCs or other cell types. Lastly, to test the neuroprotective capability, MDSCs can be adoptively transferred to MPTP intoxicated mice and surviving neurons assessed. MDSCs can be isolated from the spleens of mice and transferred after MPTP, as was previously described with Tregs. Evidence of decreased neuroinflammation and neuroprotection would indicate that other suppressor cell types protect in the model, while no changes may suggest a uniqueness about the suppressive function of Tregs. Both of which are 213

novel findings and would lead to a better understanding of the immune system in the

MPTP model and PD.

The research in Chapter 4 can be continued by several possible future experiments. In this experiment, we tested the expression of proinflammatory cytokines at the peak of neuroinflammation. However, more than likely these cytokine genes peak in the expression before the peak of neuroinflammation as defined by reactive microglia.

This would make sense if proinflammatory cytokines are needed for the activation of microglia. Also, we tested gene expression and cytokine release at the same time. In the future, the temporal kinetics of gene and protein expression should be better delineated, particularly at earlier time points. We also only tested gene expression in the midbrain and the cervical lymph nodes. Future experiments could test gene expression and cytokine release in other tissues as well as at different time points. While not typically assessed, the kinetics of gene expression and cytokine release in the striatum would be of particular interest as well as in other lymph node populations, the peripheral blood, and the spleen. We could also further test the percentage of additional immune cells such as B cells, CD4+ and CD8+, and macrophages, and DCs in the cervical, brachial, axillary, and inguinal lymph nodes as well as in the spleen and peripheral blood. This would better complete the cellular signature of immune cells that react over time after MPTP intoxication. In addition, future experiments could test the percentage and function of other regulatory immune cells post-MPTP. While Tregs would be the obvious target population, MDSCs, B regulatory cells, and CD8+ regulatory T cells would be interesting populations to assess over time. The results of these studies would provide us a complete understanding in the MPTP model of immune repertoire of cells and cytokines that could be brought to bear as putative therapeutic modalities for PD.

214

BIBLIOGRAPHY

Aarsland D, Zaccai J, Brayne C (2005) A systematic review of prevalence studies of

dementia in Parkinson's disease. Mov Disord 20:1255-1263.

Abe T, Isobe C, Murata T, Sato C, Tohgi H (2003) Alteration of 8-hydroxyguanosine

concentrations in the cerebrospinal fluid and serum from patients with

Parkinson's disease. Neuroscience Letters 336:105-108.

Agid YA (1991) Parkinson's disease: pathophysiology. Lancet 337:1321-1324.

Ahmed I, Tamouza R, Delord M, Krishnamoorthy R, Tzourio C, Mulot C, Nacfer M,

Lambert JC, Beaune P, Laurent-Puig P, Loriot MA, Charron D, Elbaz A (2012)

Association between Parkinson's disease and the HLA-DRB1 locus. Mov Disord

27:1104-1110.

Ahn TB, Kim SY, Kim JY, Park SS, Lee DS, Min HJ, Kim YK, Kim SE, Kim JM, Kim HJ,

Cho J, Jeon BS (2008) alpha-Synuclein gene duplication is present in sporadic

Parkinson disease. Neurology 70:43-49.

Akashi K, Traver D, Miyamoto T, Weissman IL (2000) A clonogenic common myeloid

progenitor that gives rise to all myeloid lineages. Nature 404:193-197.

Alam M, Schmidt WJ (2004) L-DOPA reverses the hypokinetic behaviour and rigidity in

rotenone-treated rats. Behav Brain Res 153:439-446.

Alam ZI, Daniel SE, Lees AJ, Marsden DC, Jenner P, Halliwell B (1997a) A generalised

increase in protein carbonyls in the brain in Parkinson's but not incidental Lewy

body disease. J Neurochem 69:1326-1329.

Alam ZI, Jenner A, Daniel SE, Lees AJ, Cairns N, Marsden CD, Jenner P, Halliwell B

(1997b) Oxidative DNA damage in the parkinsonian brain: an apparent selective

increase in 8-hydroxyguanine levels in substantia nigra. J Neurochem 69:1196-

1203. 215

Anderson JP, Walker DE, Goldstein JM, de Laat R, Banducci K, Caccavello RJ, Barbour

R, Huang J, Kling K, Lee M, Diep L, Keim PS, Shen X, Chataway T,

Schlossmacher MG, Seubert P, Schenk D, Sinha S, Gai WP, Chilcote TJ (2006)

Phosphorylation of Ser-129 is the dominant pathological modification of alpha-

synuclein in familial and sporadic Lewy body disease. J Biol Chem 281:29739-

29752.

Atac Ucar C, Gokce Cokal B, Unal Artik HA, Inan LE, Yoldas TK (2016) Comparison of

neutrophil-lymphocyte ratio (NLR) in Parkinson's disease subtypes. Neurol Sci.

Azad N, Rojanasakul Y, Vallyathan V (2008) Inflammation and lung cancer: roles of

reactive oxygen/nitrogen species. J Toxicol Environ Health B Crit Rev 11:1-15.

Baba Y, Kuroiwa A, Uitti RJ, Wszolek ZK, Yamada T (2005) Alterations of T-lymphocyte

populations in Parkinson disease. Parkinsonism Relat Disord 11:493-498.

Baldwin GC, Benveniste EN, Chung GY, Gasson JC, Golde DW (1993) Identification

and characterization of a high-affinity granulocyte-macrophage colony-stimulating

factor receptor on primary rat oligodendrocytes. Blood 82:3279-3282.

Banchereau J, Steinman RM (1998) Dendritic cells and the control of immunity. Nature

392:245-252.

Barcia C, de Pablos V, Bautista-Hernandez V, Sanchez-Bahillo A, Bernal I, Fernandez-

Villalba E, Martin J, Banon R, Fernandez-Barreiro A, Herrero MT (2005)

Increased plasma levels of TNF-alpha but not of IL1-beta in MPTP-treated

monkeys one year after the MPTP administration. Parkinsonism Relat Disord

11:435-439.

Bartels T, Choi JG, Selkoe DJ (2011) alpha-Synuclein occurs physiologically as a

helically folded tetramer that resists aggregation. Nature 477:107-110. 216

Bas J, Calopa M, Mestre M, Molleví DG, Cutillas B, Ambrosio S, Buendia E (2001)

Lymphocyte populations in Parkinson's disease and in rat models of

parkinsonism. J Neuroimmunol 113:146-152.

Bellou V, Belbasis L, Tzoulaki I, Evangelou E, Ioannidis JP (2016) Environmental risk

factors and Parkinson's disease: An umbrella review of meta-analyses.

Parkinsonism Relat Disord 23:1-9.

Bendor JT, Logan TP, Edwards RH (2013) The function of alpha-synuclein. Neuron

79:1044-1066.

Benkler M, Agmon-Levin N, Hassin-Baer S, Cohen OS, Ortega-Hernandez OD, Levy A,

Moscavitch SD, Szyper-Kravitz M, Damianovich M, Blank M, Chapman J,

Shoenfeld Y (2012) Immunology, autoimmunity, and autoantibodies in

Parkinson's disease. Clin Rev Allergy Immunol 42:164-171.

Benner EJ, Banerjee R, Reynolds AD, Sherman S, Pisarev VM, Tsiperson V, Nemachek

C, Ciborowski P, Przedborski S, Mosley RL, Gendelman HE (2008) Nitrated

alpha-synuclein immunity accelerates degeneration of nigral dopaminergic

neurons. PLoS One 3:e1376.

Berard M, Tough DF (2002) Qualitative differences between naïve and memory T cells.

Immunology 106:127-138.

Betarbet R, Sherer TB, MacKenzie G, Garcia-Osuna M, Panov AV, Greenamyre JT

(2000) Chronic systemic pesticide exposure reproduces features of Parkinson's

disease. Nat Neurosci 3:1301-1306.

Beveridge RA et al. (2009) A Comparison of Efficacy of Sargramostim (Yeast-Derived

RhuGM-CSF) and Filgrastim (Bacteria-Derived RhuG-CSF) in the Therapeutic

Setting of Chemotherapy-Induced Myelosuppression. Cancer Investigation

16:366-373. 217

Beyer M, Gimsa U, Eyüpoglu IY, Hailer NP, Nitsch R (2000) Phagocytosis of neuronal or

glial debris by microglial cells: upregulation of MHC class II expression and

multinuclear giant cell formation in vitro. Glia 31:262-266.

Bhairavabhotla R, Kim YC, Glass DD, Escobar TM, Patel MC, Zahr R, Nguyen CK,

Kilaru GK, Muljo SA, Shevach EM (2016) Transcriptome profiling of human

FoxP3+ regulatory T cells. Hum Immunol 77:201-213.

Bhattacharya P, Gopisetty A, Ganesh BB, Sheng JR, Prabhakar BS (2011) GM-CSF-

induced, bone-marrow-derived dendritic cells can expand natural Tregs and

induce adaptive Tregs by different mechanisms. J Leukoc Biol 89:235-249.

Bluestone JA, Abbas AK (2003) Natural versus adaptive regulatory T cells. Nat Rev

Immunol 3:253-257.

Blum-Degen D, Müller T, Kuhn W, Gerlach M, Przuntek H, Riederer P (1995) Interleukin-

1 beta and interleukin-6 are elevated in the cerebrospinal fluid of Alzheimer's and

de novo Parkinson's disease patients. Neurosci Lett 202:17-20.

Bogdan C, Röllinghoff M, Diefenbach A (2000) Reactive oxygen and reactive nitrogen

intermediates in innate and specific immunity. Curr Opin Immunol 12:64-76.

Borish LC, Steinke JW (2003) 2. Cytokines and chemokines. Journal of Allergy and

Clinical Immunology 111:S460-S475.

Bouhy D, Malgrange B, Multon S, Poirrier AL, Scholtes F, Schoenen J, Franzen R

(2006) Delayed GM-CSF treatment stimulates axonal regeneration and functional

recovery in paraplegic rats via an increased BDNF expression by endogenous

macrophages. FASEB J 20:1239-1241.

Bour H, Peyron E, Gaucherand M, Garrigue JL, Desvignes C, Kaiserlian D, Revillard JP,

Nicolas JF (1995) Major histocompatibility complex class I-restricted CD8+ T

cells and class II-restricted CD4+ T cells, respectively, mediate and regulate

contact sensitivity to dinitrofluorobenzene. Eur J Immunol 25:3006-3010. 218

Bove J, Perier C (2012) Neurotoxin-based models of Parkinson's disease. Neuroscience

211:51-76.

Boyd JD, Jang H, Shepherd KR, Faherty C, Slack S, Jiaom Y, Smeyne RJ (2007)

Response to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) differs in

mouse strains and reveals a divergence in JNK signaling and COX-2 induction

prior to loss of neurons in the substantia nigra pars compacta. Brain Res

1175:107-116.

Braak H, Rub U, Gai WP, Del Tredici K (2003) Idiopathic Parkinson's disease: possible

routes by which vulnerable neuronal types may be subject to neuroinvasion by

an unknown pathogen. J Neural Transm (Vienna) 110:517-536.

Braak H, Ghebremedhin E, Rub U, Bratzke H, Del Tredici K (2004) Stages in the

development of Parkinson's disease-related pathology. Cell Tissue Res 318:121-

134.

Brach MA, Henschler R, Mertelsmann RH, Herrmann F (1991) Regulation of M-CSF

expression by M-CSF: role of protein kinase C and transcription factor NF kappa

B. Pathobiology 59:284-288.

Brochard V, Combadiere B, Prigent A, Laouar Y, Perrin A, Beray-Berthat V, Bonduelle

O, Alvarez-Fischer D, Callebert J, Launay JM, Duyckaerts C, Flavell RA, Hirsch

EC, Hunot S (2009) Infiltration of CD4+ lymphocytes into the brain contributes to

neurodegeneration in a mouse model of Parkinson disease. J Clin Invest

119:182-192.

Brooks AI, Chadwick CA, Gelbard HA, Cory-Slechta DA, Federoff HJ (1999) Paraquat

elicited neurobehavioral syndrome caused by dopaminergic neuron loss. Brain

Res 823:1-10.

Bus JS, Aust SD, Gibson JE (1976) Paraquat toxicity: proposed mechanism of action

involving lipid peroxidation. Environ Health Perspect 16:139-146. 219

Cadman ET, Lawrence RA (2010) Granulocytes: effector cells or immunomodulators in

the immune response to helminth infection? Parasite Immunol 32:1-19.

Calne DB (1993) Treatment of Parkinson's disease. N Engl J Med 329:1021-1027.

Campbell IK, Rich MJ, Bischof RJ, Dunn AR, Grail D, Hamilton JA (1998) Protection

from collagen-induced arthritis in granulocyte-macrophage colony-stimulating

factor-deficient mice. J Immunol 161:3639-3644.

Cannon JR, Tapias V, Na HM, Honick AS, Drolet RE, Greenamyre JT (2009) A highly

reproducible rotenone model of Parkinson's disease. Neurobiology of Disease

34:279-290.

Caramori G, Lim S, Ito K, Tomita K, Oates T, Jazrawi E, Chung KF, Barnes PJ, Adcock

IM (2001) Expression of GATA family of transcription factors in T-cells,

monocytes and bronchial biopsies. Eur Respir J 18:466-473.

Carding SR, Egan PJ (2002) Gammadelta T cells: functional plasticity and

heterogeneity. Nat Rev Immunol 2:336-345.

Carey RJ (1992) Factors in amphetamine-induced contralateral rotation in the unilateral

6-OHDA lesion rat model during the first-week postoperative: implications for

neuropathology and neural grafting. Brain Res 570:11-20.

Chakraborty R, Rooney C, Dotti G, Savoldo B (2012) Changes in chemokine receptor

expression of regulatory T cells after ex vivo culture. J Immunother 35:329-336.

Chan WY, Kohsaka S, Rezaie P (2007) The origin and cell lineage of microglia: new

concepts. Brain Res Rev 53:344-354.

Chang AL, Miska J, Wainwright DA, Dey M, Rivetta CV, Yu D, Kanojia D, Pituch KC,

Qiao J, Pytel P, Han Y, Wu M, Zhang L, Horbinski CM, Ahmed AU, Lesniak MS

(2016) CCL2 Produced by the Glioma Microenvironment Is Essential for the

Recruitment of Regulatory T Cells and Myeloid-Derived Suppressor Cells.

Cancer Res 76:5671-5682. 220

Chang D, Nalls MA, Hallgrimsdottir IB, Hunkapiller J, van der Brug M, Cai F,

International Parkinson's Disease Genomics C, andMe Research T, Kerchner

GA, Ayalon G, Bingol B, Sheng M, Hinds D, Behrens TW, Singleton AB,

Bhangale TR, Graham RR (2017) A meta-analysis of genome-wide association

studies identifies 17 new Parkinson's disease risk loci. Nat Genet 49:1511-1516.

Cheatem D, Ganesh BB, Gangi E, Vasu C, Prabhakar BS (2009) Modulation of dendritic

cells using granulocyte-macrophage colony-stimulating factor (GM-CSF) delays

type 1 diabetes by enhancing CD4+CD25+ regulatory T cell function. Clin

Immunol 131:260-270.

Chen Y, Qi B, Xu W, Ma B, Li L, Chen Q, Qian W, Liu X, Qu H (2015) Clinical correlation

of peripheral CD4+cell subsets, their imbalance and Parkinson's disease. Mol

Med Rep 12:6105-6111.

Cheng JW, Sadeghi Z, Levine AD, Penn MS, von Recum HA, Caplan AI, Hijaz A (2014)

The role of CXCL12 and CCL7 chemokines in immune regulation, embryonic

development, and tissue regeneration. Cytokine 69:277-283.

Cherry JD, Olschowka JA, O'Banion MK (2014) Neuroinflammation and M2 microglia:

the good, the bad, and the inflamed. J Neuroinflammation 11:98.

Chi DS, Gong L, Daigneault EA, Kostrzewa RM (1992) Effects of MPTP and vitamin E

treatments on immune function in mice. Int J Immunopharmacol 14:739-746.

Chiba K, Trevor A, Castagnoli NJ (1984) Metabolism of the neurotoxic tertiary amine,

MPTP, by brain monoamine oxidase. Biochem Biophys Res Commun 120:574-

578.

Chung ES, Kim H, Lee G, Park S, Kim H, Bae H (2012) Neuro-protective effects of bee

venom by suppression of neuroinflammatory responses in a mouse model of

Parkinson's disease: role of regulatory T cells. Brain Behav Immun 26:1322-

1330. 221

Chung ES, Lee G, Lee C, Ye M, Chung HS, Kim H, Bae SJ, Hwang DS, Bae H (2015)

Bee Venom Phospholipase A2, a Novel Foxp3+ Regulatory T Cell Inducer,

Protects Dopaminergic Neurons by Modulating Neuroinflammatory Responses in

a Mouse Model of Parkinson's Disease. J Immunol 195:4853-4860.

Ciaramella A, Salani F, Bizzoni F, Pontieri FE, Stefani A, Pierantozzi M, Assogna F,

Caltagirone C, Spalletta G, Bossu P (2013) Blood dendritic cell frequency

declines in idiopathic Parkinson's disease and is associated with motor symptom

severity. PLoS One 8:e65352.

Cicchetti F, Brownell AL, Williams K, Chen YI, Livni E, Isacson O (2002)

Neuroinflammation of the nigrostriatal pathway during progressive 6-OHDA

dopamine degeneration in rats monitored by immunohistochemistry and PET

imaging. Eur J Neurosci 15:991-998.

Clarkson BD, Walker A, Harris M, Rayasam A, Sandor M, Fabry Z (2014) Mapping the

accumulation of co-infiltrating CNS dendritic cells and encephalitogenic T cells

during EAE. J Neuroimmunol 277:39-49.

Clarkson BD, Walker A, Harris MG, Rayasam A, Sandor M, Fabry Z (2015) CCR2-

dependent dendritic cell accumulation in the central nervous system during early

effector experimental autoimmune encephalomyelitis is essential for effector T

cell restimulation in situ and disease progression. J Immunol 194:531-541.

Codarri L, Gyulveszi G, Tosevski V, Hesske L, Fontana A, Magnenat L, Suter T, Becher

B (2011) RORgammat drives production of the cytokine GM-CSF in helper T

cells, which is essential for the effector phase of autoimmune neuroinflammation.

Nat Immunol 12:560-567.

Cohen G, Pasik P, Cohen B, Leist A, Mytilineou C, Yahr MD (1984) Pargyline and

deprenyl prevent the neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine (MPTP) in monkeys. Eur J Pharmacol 106:209-210. 222

Cote M, Drouin-Ouellet J, Cicchetti F, Soulet D (2011) The critical role of the MyD88-

dependent pathway in non-CNS MPTP-mediated toxicity. Brain Behav Immun

25:1143-1152.

Crabtree GR (1989) Contingent genetic regulatory events in T lymphocyte activation.

Science 243:355-361.

Crabtree GR, Durand D (1986) Control of the early activation genes of T lymphocytes.

Bioessays 5:220-222.

Cruciani M, Mengoli C, Serpelloni G, Mazzi R, Bosco O, Malena M (2007) Granulocyte

macrophage colony-stimulating factor as an adjuvant for hepatitis B vaccination:

a meta-analysis. Vaccine 25:709-718.

Cubukcu HC, Yurtdas M, Durak ZE, Aytac B, Gunes HN, Cokal BG, Yoldas TK, Durak I

(2016) Oxidative and nitrosative stress in serum of patients with Parkinson's

disease. Neurol Sci 37:1793-1798.

Członkowska A, Kohutnicka M, Kurkowska-Jastrzebska I, Członkowski A (1996)

Microglial reaction in MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine)

induced Parkinson's disease mice model. Neurodegeneration 5:137-143.

Daniele SG, Béraud D, Davenport C, Cheng K, Yin H, Maguire-Zeiss KA (2015)

Activation of MyD88-dependent TLR1/2 signaling by misfolded α-synuclein, a

protein linked to neurodegenerative disorders. Sci Signal 8:ra45.

Dauer W, Przedborski S (2003) Parkinson's disease: mechanisms and models. Neuron

39:889-909.

Davidson WS, Jonas A, Clayton DF, George JM (1998) Stabilization of alpha-synuclein

secondary structure upon binding to synthetic membranes. J Biol Chem

273:9443-9449. 223

Day BJ, Patel M, Calavetta L, Chang LY, Stamler JS (1999) A mechanism of paraquat

toxicity involving nitric oxide synthase. Proc Natl Acad Sci U S A 96:12760-

12765. de Groot RP, Coffer PJ, Koenderman L (1998) Regulation of proliferation, differentiation

and survival by the IL-3/IL-5/GM-CSF receptor family. Cell Signal 10:619-628. de Lau LML, Breteler MMB (2006) Epidemiology of Parkinson's disease. The Lancet

Neurology 5:525-535.

Deaglio S, Dwyer KM, Gao W, Friedman D, Usheva A, Erat A, Chen JF, Enjyoji K,

Linden J, Oukka M, Kuchroo VK, Strom TB, Robson SC (2007) Adenosine

generation catalyzed by CD39 and CD73 expressed on regulatory T cells

mediates immune suppression. J Exp Med 204:1257-1265.

Delgado M, Ganea D (2003) Neuroprotective effect of vasoactive intestinal peptide (VIP)

in a mouse model of Parkinson's disease by blocking microglial activation.

FASEB J 17:944-946.

Depboylu C, Stricker S, Ghobril JP, Oertel WH, Priller J, Hoglinger GU (2012) Brain-

resident microglia predominate over infiltrating myeloid cells in activation,

phagocytosis and interaction with T-lymphocytes in the MPTP mouse model of

Parkinson disease. Exp Neurol 238:183-191.

Dhall R, Kreitzman DL (2016) Advances in levodopa therapy for Parkinson disease:

Review of RYTARY (carbidopa and levodopa) clinical efficacy and safety.

Neurology 86:S13-24.

Dinarello CA (2009) Immunological and inflammatory functions of the interleukin-1

family. Annu Rev Immunol 27:519-550.

Dolcetti L, Peranzoni E, Ugel S, Marigo I, Fernandez Gomez A, Mesa C, Geilich M,

Winkels G, Traggiai E, Casati A, Grassi F, Bronte V (2010) Hierarchy of 224

immunosuppressive strength among myeloid-derived suppressor cell subsets is

determined by GM-CSF. Eur J Immunol 40:22-35.

Doo AR, Kim ST, Kim SN, Moon W, Yin CS, Chae Y, Park HK, Lee H, Park HJ (2010)

Neuroprotective effects of bee venom pharmaceutical acupuncture in acute 1-

methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced mouse model of Parkinson's

disease. Neurol Res 32 Suppl 1:88-91.

Dorr RT (1993) Clinical properties of yeast-derived versus Escherichia coli-derived

granulocyte-macrophage colony-stimulating factor. Clin Ther 15:19-29.

Du Y, Ma Z, Lin S, Dodel RC, Gao F, Bales KR, Triarhou LC, Chernet E, Perry KW,

Nelson DL, Luecke S, Phebus LA, Bymaster FP, Paul SM (2001) Minocycline

prevents nigrostriatal dopaminergic neurodegeneration in the MPTP model of

Parkinson's disease. Proc Natl Acad Sci U S A 98:14669-14674.

Duda JE, Giasson BI, Chen Q, Gur TL, Hurtig HI, Stern MB, Gollomp SM, Ischiropoulos

H, Lee VM, Trojanowski JQ (2000) Widespread nitration of pathological

inclusions in neurodegenerative synucleinopathies. Am J Pathol 157:1439-1445.

Dumitriu A, Golji J, Labadorf AT, Gao B, Beach TG, Myers RH, Longo KA, Latourelle JC

(2016) Integrative analyses of proteomics and RNA transcriptomics implicate

mitochondrial processes, protein folding pathways and GWAS loci in Parkinson

disease. BMC Med Genomics 9:5.

Dursun E, Gezen-Ak D, Hanagasi H, Bilgic B, Lohmann E, Ertan S, Atasoy IL, Alaylioglu

M, Araz OS, Onal B, Gunduz A, Apaydin H, Kiziltan G, Ulutin T, Gurvit H,

Yilmazer S (2015) The interleukin 1 alpha, interleukin 1 beta, interleukin 6 and

alpha-2-macroglobulin serum levels in patients with early or late onset

Alzheimer's disease, mild cognitive impairment or Parkinson's disease. J

Neuroimmunol 283:50-57. 225

Duty S, Jenner P (2011) Animal models of Parkinson's disease: a source of novel

treatments and clues to the cause of the disease. Br J Pharmacol 164:1357-

1391.

Emborg ME (2007) Nonhuman primate models of Parkinson's disease. ILAR J 48:339-

355.

Felger JC, Abe T, Kaunzner UW, Gottfried-Blackmore A, Gal-Toth J, McEwen BS,

Iadecola C, Bulloch K (2010) Brain dendritic cells in ischemic stroke: time course,

activation state, and origin. Brain Behav Immun 24:724-737.

Fellner L, Irschick R, Schanda K, Reindl M, Klimaschewski L, Poewe W, Wenning GK,

Stefanova N (2013) Toll-like receptor 4 is required for alpha-synuclein dependent

activation of microglia and astroglia. Glia 61:349-360.

Ferger B, Leng A, Mura A, Hengerer B, Feldon J (2004) Genetic ablation of tumor

necrosis factor-alpha (TNF-alpha) and pharmacological inhibition of TNF-

synthesis attenuates MPTP toxicity in mouse striatum. J Neurochem 89:822-833.

Filipazzi P, Valenti R, Huber V, Pilla L, Canese P, Iero M, Castelli C, Mariani L, Parmiani

G, Rivoltini L (2007) Identification of a new subset of myeloid suppressor cells in

peripheral blood of melanoma patients with modulation by a granulocyte-

macrophage colony-stimulation factor-based antitumor vaccine. J Clin Oncol

25:2546-2553.

Fiszer D, Rozwadowska N, Rychlewski L, Kosicki W, Kurpisz M (2007) Identification of

IL-18RAP mRNA truncated splice variants in human testis and the other human

tissues. Cytokine 39:178-183.

Fiszer U, Mix E, Fredrikson S, Kostulas V, Olsson T, Link H (1994) gamma delta+ T cells

are increased in patients with Parkinson's disease. J Neurol Sci 121:39-45.

Fornai F, Lenzi P, Gesi M, Ferrucci M, Lazzeri G, Natale G, Ruggieri S, Paparelli A

(2003) Recent knowledge on molecular components of Lewy bodies discloses 226

future therapeutic strategies in Parkinson's disease. Curr Drug Targets CNS

Neurol Disord 2:149-152.

Fornai F, Schluter OM, Lenzi P, Gesi M, Ruffoli R, Ferrucci M, Lazzeri G, Busceti CL,

Pontarelli F, Battaglia G, Pellegrini A, Nicoletti F, Ruggieri S, Paparelli A, Sudhof

TC (2005) Parkinson-like syndrome induced by continuous MPTP infusion:

convergent roles of the ubiquitin-proteasome system and alpha-synuclein. Proc

Natl Acad Sci U S A 102:3413-3418.

Frei K, Nohava K, Malipiero UV, Schwerdel C, Fontana A (1992) Production of

macrophage colony-stimulating factor by astrocytes and brain macrophages. J

Neuroimmunol 40:189-195.

Fujiwara H, Hasegawa M, Dohmae N, Kawashima A, Masliah E, Goldberg MS, Shen J,

Takio K, Iwatsubo T (2002) alpha-Synuclein is phosphorylated in

synucleinopathy lesions. Nat Cell Biol 4:160-164.

Funk N, Wieghofer P, Grimm S, Schaefer R, Buhring HJ, Gasser T, Biskup S (2013)

Characterization of peripheral hematopoietic stem cells and monocytes in

Parkinson's disease. Mov Disord 28:392-395.

Gabay C, Lamacchia C, Palmer G (2010) IL-1 pathways in inflammation and human

diseases. Nat Rev Rheumatol 6:232-241.

Gabrilovich DI, Nagaraj S (2009) Myeloid-derived suppressor cells as regulators of the

immune system. Nat Rev Immunol 9:162-174.

Galy A, Travis M, Cen D, Chen B (1995) Human T, B, natural killer, and dendritic cells

arise from a common bone marrow progenitor cell subset. Immunity 3:459-473.

Games D, Valera E, Spencer B, Rockenstein E, Mante M, Adame A, Patrick C, Ubhi K,

Nuber S, Sacayon P, Zago W, Seubert P, Barbour R, Schenk D, Masliah E

(2014) Reducing C-terminal-truncated alpha-synuclein by immunotherapy 227

attenuates neurodegeneration and propagation in Parkinson's disease-like

models. J Neurosci 34:9441-9454.

Ganesh BB, Cheatem DM, Sheng JR, Vasu C, Prabhakar BS (2009) GM-CSF-induced

CD11c+CD8a--dendritic cells facilitate Foxp3+ and IL-10+ regulatory T cell

expansion resulting in suppression of autoimmune thyroiditis. Int Immunol

21:269-282.

Gangi E, Vasu C, Cheatem D, Prabhakar BS (2005) IL-10-Producing CD4+CD25+

Regulatory T Cells Play a Critical Role in Granulocyte-Macrophage Colony-

Stimulating Factor-Induced Suppression of Experimental Autoimmune

Thyroiditis. The Journal of Immunology 174:7006-7013.

Gao X, Chen H, Schwarzschild MA, Ascherio A (2011) Use of ibuprofen and risk of

Parkinson disease. Neurology 76:863-869.

Garcia JA, Elson P, Tyler A, Triozzi P, Dreicer R (2014) Sargramostim (GM-CSF) and

lenalidomide in castration-resistant prostate cancer (CRPC): results from a phase

I-II clinical trial. Urol Oncol 32:33 e11-37.

Garver DL, Cedarbaum J, Maas JW (1975) Blood-brain barrier to 6-hydroxydopamine:

uptake by heart and brain. Life Sci 17:1081-1084.

Gatto EM, Carreras MC, Pargament GA, Riobo NA, Reides C, Repetto M, Fernandez

Pardal MM, Llesuy S, Poderoso JJ (1996) Neutrophil function, nitric oxide, and

blood oxidative stress in Parkinson's disease. Mov Disord 11:261-267.

Gaudreau S, Guindi C, Menard M, Besin G, Dupuis G, Amrani A (2007) Granulocyte-

Macrophage Colony-Stimulating Factor Prevents Diabetes Development in NOD

Mice by Inducing Tolerogenic Dendritic Cells that Sustain the Suppressive

Function of CD4+CD25+ Regulatory T Cells. The Journal of Immunology

179:3638-3647. 228

Geering B, Stoeckle C, Conus S, Simon HU (2013) Living and dying for inflammation:

neutrophils, eosinophils, basophils. Trends Immunol 34:398-409.

Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K (2010) Development of

monocytes, macrophages, and dendritic cells. Science 327:656-661.

Gendelman HE, Zhang Y, Santamaria P, Olson KE, Schutt CR, Bhatti D, Shetty BLD, Lu

Y, Estes KA, Standaert DG, Heinrichs-Graham E, Larson L, Meza JL, Follett M,

Forsberg E, Siuzdak G, Wilson TW, Peterson C, Mosley RL (2017) Evaluation of

the safety and immunomodulatory effects of sargramostim in a randomized,

double-blind phase 1 clinical Parkinson’s disease trial. npj Parkinson's Disease 3.

Gerhard A, Pavese N, Hotton G, Turkheimer F, Es M, Hammers A, Eggert K, Oertel W,

Banati RB, Brooks DJ (2006) In vivo imaging of microglial activation with

[11C](R)-PK11195 PET in idiopathic Parkinson's disease. Neurobiol Dis 21:404-

412.

Giasson BI (2000) Oxidative Damage Linked to Neurodegeneration by Selective alpha -

Synuclein Nitration in Synucleinopathy Lesions. Science 290:985-989.

Giasson BI, Duda JE, Quinn SM, Zhang B, Trojanowski JQ, Lee VM (2002) Neuronal

alpha-synucleinopathy with severe movement disorder in mice expressing A53T

human alpha-synuclein. Neuron 34:521-533.

Glinka Y, Tipton KF, Youdim MB (1996) Nature of inhibition of mitochondrial respiratory

complex I by 6-Hydroxydopamine. J Neurochem 66:2004-2010.

Glinka YY, Youdim MB (1995) Inhibition of mitochondrial complexes I and IV by 6-

hydroxydopamine. Eur J Pharmacol 292:329-332.

Goh W, Huntington ND (2017) Regulation of Murine Natural Killer Cell Development.

Front Immunol 8:130.

Goldberg MS, Pisani A, Haburcak M, Vortherms TA, Kitada T, Costa C, Tong Y, Martella

G, Tscherter A, Martins A, Bernardi G, Roth BL, Pothos EN, Calabresi P, Shen J 229

(2005) Nigrostriatal dopaminergic deficits and hypokinesia caused by inactivation

of the familial Parkinsonism-linked gene DJ-1. Neuron 45:489-496.

Goldberg MS, Fleming SM, Palacino JJ, Cepeda C, Lam HA, Bhatnagar A, Meloni EG,

Wu N, Ackerson LC, Klapstein GJ, Gajendiran M, Roth BL, Chesselet MF,

Maidment NT, Levine MS, Shen J (2003) Parkin-deficient mice exhibit

nigrostriatal deficits but not loss of dopaminergic neurons. J Biol Chem

278:43628-43635.

Goldeck D, Maetzler W, Berg D, Oettinger L, Pawelec G (2016) Altered dendritic cell

subset distribution in patients with Parkinson's disease: Impact of CMV

serostatus. J Neuroimmunol 290:60-65.

Gonzalez-Rey E, Delgado M (2007) Vasoactive intestinal peptide and regulatory T-cell

induction: a new mechanism and therapeutic potential for immune homeostasis.

Trends Mol Med 13:241-251.

Gonzalez H, Pacheco R (2014) T-cell-mediated regulation of neuroinflammation involved

in neurodegenerative diseases. J Neuroinflammation 11:201.

Gopisetty A, Bhattacharya P, Haddad C, Bruno JC, Jr., Vasu C, Miele L, Prabhakar BS

(2013) OX40L/Jagged1 cosignaling by GM-CSF-induced bone marrow-derived

dendritic cells is required for the expansion of functional regulatory T cells. J

Immunol 190:5516-5525.

Gordon S, Taylor PR (2005) Monocyte and macrophage heterogeneity. Nat Rev

Immunol 5:953-964.

Griseri T, Asquith M, Thompson C, Powrie F (2010) OX40 is required for regulatory T

cell-mediated control of colitis. J Exp Med 207:699-709.

Grozdanov V, Bliederhaeuser C, Ruf WP, Roth V, Fundel-Clemens K, Zondler L,

Brenner D, Martin-Villalba A, Hengerer B, Kassubek J, Ludolph AC, Weishaupt 230

JH, Danzer KM (2014) Inflammatory dysregulation of blood monocytes in

Parkinson's disease patients. Acta Neuropathol 128:651-663.

Guillemin G, Boussin FD, Le Grand R, Croitoru J, Coffigny H, Dormont D (1996)

Granulocyte macrophage colony stimulating factor stimulates in vitro proliferation

of astrocytes derived from simian mature brains. Glia 16:71-80.

Gupta BK (2011) Synthetic Heroin-Induced Parkinsonism. Jefferson Journal of

Psychiatry 4:71-74.

Gupta V, Garg RK, Khattri S (2016) Levels of IL-8 and TNF-alpha decrease in

Parkinson's disease. Neurol Res 38:98-102.

Haddad CS, Bhattacharya P, Alharshawi K, Marinelarena A, Kumar P, El-Sayed O,

Elshabrawy HA, Epstein AL, Prabhakar BS (2016) Age-dependent divergent

effects of OX40L treatment on the development of diabetes in NOD mice.

Autoimmunity 49:298-311.

Haddad D, Nakamura K (2015) Understanding the susceptibility of dopamine neurons to

mitochondrial stressors in Parkinson's disease. FEBS Lett 589:3702-3713.

Hansen G, Hercus TR, McClure BJ, Stomski FC, Dottore M, Powell J, Ramshaw H,

Woodcock JM, Xu Y, Guthridge M, McKinstry WJ, Lopez AF, Parker MW (2008)

The structure of the GM-CSF receptor complex reveals a distinct mode of

cytokine receptor activation. Cell 134:496-507.

Hardy J, Cookson MR, Singleton A (2003) Genes and parkinsonism. Lancet Neurol

2:221-228.

Harris MA, Tsui JK, Marion SA, Shen H, Teschke K (2012) Association of Parkinson's

disease with infections and occupational exposure to possible vectors. Mov

Disord 27:1111-1117.

Harris RJ, Pettitt AR, Schmutz C, Sherrington PD, Zuzel M, Cawley JC, Griffiths SD

(2000) Granuloctye-Macrophage Colony-Stimulating Factor as an Autocrine 231

Survival Factor for Mature Normal and Malignant B Lymphocytes. The Journal of

Immunology 164:3887-3893.

Hartmann A et al. (2016) Bee Venom for the Treatment of Parkinson Disease - A

Randomized Controlled Clinical Trial. PLoS One 11:e0158235.

Harty JT, Tvinnereim AR, White DW (2000) CD8+ T cell effector mechanisms in

resistance to infection. Annu Rev Immunol 18:275-308.

Hasegawa M, Fujiwara H, Nonaka T, Wakabayashi K, Takahashi H, Lee VM,

Trojanowski JQ, Mann D, Iwatsubo T (2002) Phosphorylated alpha-synuclein is

ubiquitinated in alpha-synucleinopathy lesions. J Biol Chem 277:49071-49076.

Hasegawa Y, Inagaki T, Sawada M, Suzumura A (2000) Impaired cytokine production by

peripheral blood mononuclear cells and monocytes/macrophages in Parkinson's

disease. Acta Neurol Scand 101:159-164.

Hébert G, Arsaut J, Dantzer R, Demotes-Mainard J (2003) Time-course of the

expression of inflammatory cytokines and matrix metalloproteinases in the

striatum and mesencephalon of mice injected with 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine, a dopaminergic neurotoxin. Neuroscience Letters 349:191-

195.

Heikkila R, Cohen G (1971) Inhibition of biogenic amine uptake by hydrogen peroxide: a

mechanism for toxic effects of 6-hydroxydopamine. Science 172:1257-1258.

Heikkila R, Cohen G (1972) Further studies on the generation of hydrogen peroxide by

6-hydroxydopamine. Potentiation by ascorbic acid. Mol Pharmacol 8:241-248.

Heikkila RE, Manzino L, Cabbat FS, Duvoisin RC (1984) Protection against the

dopaminergic neurotoxicity of 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine by

monoamine oxidase inhibitors. Nature 311:467-469. 232

Heise H, Hoyer W, Becker S, Andronesi OC, Riedel D, Baldus M (2005) Molecular-level

secondary structure, polymorphism, and dynamics of full-length alpha-synuclein

fibrils studied by solid-state NMR. Proc Natl Acad Sci U S A 102:15871-15876.

Helft J, Bottcher J, Chakravarty P, Zelenay S, Huotari J, Schraml BU, Goubau D, Reis e

Sousa C (2015) GM-CSF Mouse Bone Marrow Cultures Comprise a

Heterogeneous Population of CD11c(+)MHCII(+) Macrophages and Dendritic

Cells. Immunity 42:1197-1211.

Hely MA, Morris JG, Reid WG, Trafficante R (2005) Sydney Multicenter Study of

Parkinson's disease: non-L-dopa-responsive problems dominate at 15 years.

Mov Disord 20:190-199.

Hercus TR, Thomas D, Guthridge MA, Ekert PG, King-Scott J, Parker MW, Lopez AF

(2009) The granulocyte-macrophage colony-stimulating factor receptor: linking its

structure to cell signaling and its role in disease. Blood 114:1289-1298.

Hirsch E, Graybiel AM, Agid YA (1988) Melanized dopaminergic neurons are

differentially susceptible to degeneration in Parkinson's disease. Nature 334:345-

348.

Hirst SJ, Ferger B (2008) Systemic proteasomal inhibitor exposure enhances dopamine

turnover and decreases dopamine levels but does not affect MPTP-induced

striatal dopamine depletion in mice. Synapse 62:85-90.

Hodara R, Norris EH, Giasson BI, Mishizen-Eberz AJ, Lynch DR, Lee VM, Ischiropoulos

H (2004) Functional consequences of alpha-synuclein tyrosine nitration:

diminished binding to lipid vesicles and increased fibril formation. J Biol Chem

279:47746-47753.

Hornykiewicz O, Kish SJ (1987) Biochemical pathophysiology of Parkinson's disease.

Adv Neurol 45:19-34. 233

Horvath S, Ritz BR (2015) Increased epigenetic age and granulocyte counts in the blood

of Parkinson's disease patients. Aging 7:1130-1142.

Hoyne GF, Dallman MJ, Lamb JR (2000) T-cell regulation of peripheral tolerance and

immunity: the potential role for Notch signalling. Immunology 100:281-288.

Huang B, Pan PY, Li Q, Sato AI, Levy DE, Bromberg J, Divino CM, Chen SH (2006) Gr-

1+CD115+ immature myeloid suppressor cells mediate the development of

tumor-induced T regulatory cells and T-cell anergy in tumor-bearing host. Cancer

Res 66:1123-1131.

Huang Y, Liu Z, Wang XQ, Qiu YH, Peng YP (2014) A dysfunction of CD4+ T

lymphocytes in peripheral immune system of Parkinson's disease model mice.

Zhongguo Ying Yong Sheng Li Xue Za Zhi 30:567-576.

Hussein AM, Ross M, Vredenburgh J, Meisenberg B, Hars V, Gilbert C, Petros WP,

Coniglio D, Kurtzberg J, Rubin P (1995) Effects of granulocyte-macrophage

colony stimulating factor produced in Chinese hamster ovary cells (regramostim),

Escherichia coli (molgramostim) and yeast (sargramostim) on priming peripheral

blood progenitor cells for use with autologous bone marrow after high-dose

chemotherapy. 54 5.

Hutter-Saunders JA, Gendelman HE, Mosley RL (2012) Murine motor and behavior

functional evaluations for acute 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

(MPTP) intoxication. J Neuroimmune Pharmacol 7:279-288.

Hwang JK, Alt FW, Yeap LS (2015) Related Mechanisms of Antibody Somatic

Hypermutation and Class Switch Recombination. Microbiol Spectr 3:MDNA3-

0037-2014.

Ihrie RA, Reczek E, Horner JS, Khachatrian L, Sage J, Jacks T, Attardi LD (2003) Perp

Is a Mediator of p53-Dependent Apoptosis in Diverse Cell Types. Current Biology

13:1985-1990. 234

Imamura K, Hishikawa N, Sawada M, Nagatsu T, Yoshida M, Hashizume Y (2003)

Distribution of major histocompatibility complex class II-positive microglia and

cytokine profile of Parkinson's disease brains. Acta Neuropathol 106:518-526.

Investigators NN-P (2006) A randomized, double-blind, futility clinical trial of creatine and

minocycline in early Parkinson disease. Neurology 66:664-671.

Investigators. NETiPDN-PF-Z (2015) Pioglitazone in early Parkinson's disease: a phase

2, multicentre, double-blind, randomised trial. The Lancet Neurology 14:795-803.

Investigators. NN-P (2006) A randomized, double-blind, futility clinical trial of creatine

and minocycline in early Parkinson disease. Neurology 66:664-671.

Ioannou M, Alissafi T, Lazaridis I, Deraos G, Matsoukas J, Gravanis A, Mastorodemos

V, Plaitakis A, Sharpe A, Boumpas D, Verginis P (2012) Crucial role of

granulocytic myeloid-derived suppressor cells in the regulation of central nervous

system autoimmune disease. J Immunol 188:1136-1146.

Jackson-Lewis V, Przedborski S (2007) Protocol for the MPTP mouse model of

Parkinson's disease. Nat Protoc 2:141-151.

Jackson-Lewis V, Jakowec M, Burke RE, Przedborski S (1995) Time course and

morphology of dopaminergic neuronal death caused by the neurotoxin 1-methyl-

4-phenyl-1,2,3,6-tetrahydropyridine. Neurodegeneration 4:257-269.

Jenner P (2008) Molecular mechanisms of L-DOPA-induced dyskinesia. Nat Rev

Neurosci 9:665-677.

Jensen PH, Islam K, Kenney J, Nielsen MS, Power J, Gai WP (2000) Microtubule-

associated protein 1B is a component of cortical Lewy bodies and binds alpha-

synuclein filaments. J Biol Chem 275:21500-21507.

Jeon BS, Jackson-Lewis V, Burke RE (1995) 6-Hydroxydopamine lesion of the rat

substantia nigra: time course and morphology of cell death. Neurodegeneration

4:131-137. 235

Jetten AM (2009) Retinoid-related orphan receptors (RORs): critical roles in

development, immunity, circadian rhythm, and cellular metabolism. Nucl Recept

Signal 7:e003.

Jones HR, Robb CT, Perretti M, Rossi AG (2016) The role of neutrophils in inflammation

resolution. Semin Immunol 28:137-145.

Jones SE, Schottstaedt MW, Duncan LA, Kirby RL, Good RH, Mennel RG, George TK,

Snyder DA, Watkins DL, Denham CA, Hoyes FA, Rubin AS (1996) Randomized

double-blind prospective trial to evaluate the effects of sargramostim versus

placebo in a moderate-dose fluorouracil, doxorubicin, and cyclophosphamide

adjuvant chemotherapy program for stage II and III breast cancer. J Clin Oncol

14:2976-2983.

Jordan JT, Sun W, Hussain SF, DeAngulo G, Prabhu SS, Heimberger AB (2008)

Preferential migration of regulatory T cells mediated by glioma-secreted

chemokines can be blocked with chemotherapy. Cancer Immunol Immunother

57:123-131.

Kahle PJ, Neumann M, Ozmen L, Muller V, Jacobsen H, Schindzielorz A, Okochi M,

Leimer U, van Der Putten H, Probst A, Kremmer E, Kretzschmar HA, Haass C

(2000) Subcellular localization of wild-type and Parkinson's disease-associated

mutant alpha -synuclein in human and transgenic mouse brain. J Neurosci

20:6365-6373.

Kaku K, Shikimi T, Kamisaki Y, Shinozuka K, Ishino H, Okunishi H, Takaori S (1999)

Elevation of striatal interleukin-6 and serum corticosterone contents in MPTP-

treated mice. Clin Exp Pharmacol Physiol 26:680-683.

Kalkonde YV, Morgan WW, Sigala J, Maffi SK, Condello C, Kuziel W, Ahuja SS, Ahuja

SK (2007) Chemokines in the MPTP model of Parkinson's disease: absence of 236

CCL2 and its receptor CCR2 does not protect against striatal neurodegeneration.

Brain Res 1128:1-11.

Karampetsou M, Ardah MT, Semitekolou M, Polissidis A, Samiotaki M, Kalomoiri M,

Majbour N, Xanthou G, El-Agnaf OMA, Vekrellis K (2017) Phosphorylated

exogenous alpha-synuclein fibrils exacerbate pathology and induce neuronal

dysfunction in mice. Sci Rep 7:16533.

Kared H, Leforban B, Montandon R, Renand A, Layseca Espinosa E, Chatenoud L,

Rosenstein Y, Schneider E, Dy M, Zavala F (2008) Role of GM-CSF in tolerance

induction by mobilized hematopoietic progenitors. Blood 112:2575-2578.

Katzman SD, Fowell DJ (2008) Pathogen-imposed skewing of mouse chemokine and

cytokine expression at the infected tissue site. J Clin Invest 118:801-811.

Kelso ML, Elliott BR, Haverland NA, Mosley RL, Gendelman HE (2015) Granulocyte-

macrophage colony stimulating factor exerts protective and immunomodulatory

effects in cortical trauma. J Neuroimmunol 278:162-173.

Kempuraj D, Thangavel R, Natteru PA, Selvakumar GP, Saeed D, Zahoor H, Zaheer S,

Iyer SS, Zaheer A (2016) Neuroinflammation Induces Neurodegeneration. J

Neurol Neurosurg spine 1:pii: 1003

Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (2011) Physiology of microglia.

Physiol Rev 91:461-553.

Kim JK, Choi BH, Park HC, Park SR, Kim YS, Yoon SH, Park HS, Kim EY, Ha Y (2004)

Effects of GM-CSF on the neural progenitor cells. Neuroreport 15:2161-2165.

Kim ME, Lee JY, Lee KM, Park HR, Lee E, Lee Y, Lee JS, Lee J (2016) Neuroprotective

effect of bee venom is mediated by reduced astrocyte activation in a subchronic

MPTP-induced model of Parkinson's disease. Arch Pharm Res 39:1160-1170.

Kim NK, Choi BH, Huang X, Snyder BJ, Bukhari S, Kong TH, Park H, Park HC, Park SR,

Ha Y (2009) Granulocyte-macrophage colony-stimulating factor promotes 237

survival of dopaminergic neurons in the 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine-induced murine Parkinson's disease model. Eur J Neurosci

29:891-900.

King IL, Dickendesher TL, Segal BM (2009) Circulating Ly-6C+ myeloid precursors

migrate to the CNS and play a pathogenic role during autoimmune demyelinating

disease. Blood 113:3190-3197.

King NJ, Thomas SR (2007) Molecules in focus: indoleamine 2,3-dioxygenase. Int J

Biochem Cell Biol 39:2167-2172.

Kitada T, Pisani A, Porter DR, Yamaguchi H, Tscherter A, Martella G, Bonsi P, Zhang C,

Pothos EN, Shen J (2007) Impaired dopamine release and synaptic plasticity in

the striatum of PINK1-deficient mice. Proc Natl Acad Sci U S A 104:11441-

11446.

Klein CL, Borne RF, Stevens ED (1985) Structure of a Neurotoxin that Induces

Parkinsonism Symptoms; l-Methyl-4-phenyl-l,2,3,6-tetra-hydropyridine

Hydrochloride. Pharm Res 2:192-194.

Kohutnicka M, Lewandowska E, Kurkowska-Jastrzebska I, Członkowski A, Członkowska

A (1998) Microglial and astrocytic involvement in a murine model of Parkinson's

disease induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP).

Immunopharmacology 39:167-180.

Kolaczkowska E, Kubes P (2013) Neutrophil recruitment and function in health and

inflammation. Nat Rev Immunol 13:159-175.

Kondo M, Weissman IL, Akashi K (1997) Identification of clonogenic common lymphoid

progenitors in mouse bone marrow. Cell 91:661-672.

Korzenik JR, Dieckgraefe BK, Valentine JF, Hausman DF, Gilbert MJ, Group. SiCsDS

(2005) Sargramostim for active Crohn's disease. N Engl J Med 352:2193-2201. 238

Kosloski LM, Kosmacek EA, Olson KE, Mosley RL, Gendelman HE (2013) GM-CSF

induces neuroprotective and anti-inflammatory responses in 1-methyl-4-phenyl-

1,2,3,6-tetrahydropyridine intoxicated mice. J Neuroimmunol 265:1-10.

Kostrzewa RM, Jacobowitz DM (1974) Pharmacological actions of 6-hydroxydopamine.

Pharmacol Rev 26:199-288.

Krieger M, Both M, Kranig SA, Pitzer C, Klugmann M, Vogt G, Draguhn A, Schneider A

(2012) The hematopoietic cytokine granulocyte-macrophage colony stimulating

factor is important for cognitive functions. Sci Rep 2:697.

Kruger C, Laage R, Pitzer C, Schabitz WR, Schneider A (2007) The hematopoietic factor

GM-CSF (granulocyte-macrophage colony-stimulating factor) promotes neuronal

differentiation of adult neural stem cells in vitro. BMC Neurosci 8:88.

Kumar P, Alharshawi K, Bhattacharya P, Marinelarena A, Haddad C, Sun Z, Chiba S,

Epstein AL, Prabhakar BS (2017) Soluble OX40L and JAG1 Induce Selective

Proliferation of Functional Regulatory T-Cells Independent of canonical TCR

signaling. Sci Rep 7:39751.

Kurkowska-Jastrzebska I, Wrońska A, Kohutnicka M, Członkowski A, Członkowska A

(1999) MHC class II positive microglia and lymphocytic infiltration are present in

the substantia nigra and striatum in mouse model of Parkinson's disease. Acta

Neurobiol Exp (Wars) 59:1-8.

Laloux C, Petrault M, Lecointe C, Devos D, Bordet R (2012) Differential susceptibility to

the PPAR-gamma agonist pioglitazone in 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine and 6-hydroxydopamine rodent models of Parkinson's

disease. Pharmacol Res 65:514-522.

Lashley T, Holton JL, Gray E, Kirkham K, O'Sullivan SS, Hilbig A, Wood NW, Lees AJ,

Revesz T (2008) Cortical alpha-synuclein load is associated with amyloid-beta 239

plaque burden in a subset of Parkinson's disease patients. Acta Neuropathol

115:417-425.

Lee EG, Jung NC, Lee JH, Song JY, Ryu SY, Seo HG, Han SG, Ahn KJ, Hong KS, Choi

J, Lim DS (2016) Tolerogenic dendritic cells show gene expression profiles that

are different from those of immunogenic dendritic cells in DBA/1 mice.

Autoimmunity 49:90-101.

Lee MK, Stirling W, Xu Y, Xu X, Qui D, Mandir AS, Dawson TM, Copeland NG, Jenkins

NA, Price DL (2002) Human alpha-synuclein-harboring familial Parkinson's

disease-linked Ala-53 --> Thr mutation causes neurodegenerative disease with

alpha-synuclein aggregation in transgenic mice. Proc Natl Acad Sci U S A

99:8968-8973.

Lee SC, Liu W, Brosnan CF, Dickson DW (1994) GM-CSF promotes proliferation of

human fetal and adult microglia in primary cultures. Glia 12:309-318.

Li H, Shi B (2015) Tolerogenic dendritic cells and their applications in transplantation.

Cell Mol Immunol 12:24-30.

Li H, Zhang GX, Chen Y, Xu H, Fitzgerald DC, Zhao Z, Rostami A (2008)

CD11c+CD11b+ dendritic cells play an important role in intravenous tolerance

and the suppression of experimental autoimmune encephalomyelitis. J Immunol

181:2483-2493.

Li W, West N, Colla E, Pletnikova O, Troncoso JC, Marsh L, Dawson TM, Jakala P,

Hartmann T, Price DL, Lee MK (2005) Aggregation promoting C-terminal

truncation of alpha-synuclein is a normal cellular process and is enhanced by the

familial Parkinson's disease-linked mutations. Proc Natl Acad Sci U S A

102:2162-2167. 240

Lindqvist D, Kaufman E, Brundin L, Hall S, Surova Y, Hansson O (2012) Non-motor

symptoms in patients with Parkinson's disease - correlations with inflammatory

cytokines in serum. PLoS One 7:e47387.

Lindstrom V, Fagerqvist T, Nordstrom E, Eriksson F, Lord A, Tucker S, Andersson J,

Johannesson M, Schell H, Kahle PJ, Moller C, Gellerfors P, Bergstrom J,

Lannfelt L, Ingelsson M (2014) Immunotherapy targeting alpha-synuclein

protofibrils reduced pathology in (Thy-1)-h[A30P] alpha-synuclein mice.

Neurobiol Dis 69:134-143.

Liu J, Cao X (2015) Regulatory dendritic cells in autoimmunity: A comprehensive review.

J Autoimmun 63:1-12.

Liu YJ (2001) Dendritic cell subsets and lineages, and their functions in innate and

adaptive immunity. Cell 106:259-262.

Liu Z, Huang Y, Cao BB, Qiu YH, Peng YP (2017) Th17 Cells Induce Dopaminergic

Neuronal Death via LFA-1/ICAM-1 Interaction in a Mouse Model of Parkinson's

Disease. Mol Neurobiol 54:7762-7776.

Lohr KM, Chen M, Hoffman CA, McDaniel MJ, Stout KA, Dunn AR, Wang M, Bernstein

AI, Miller GW (2016) Vesicular Monoamine Transporter 2 (VMAT2) Level

Regulates MPTP Vulnerability and Clearance of Excess Dopamine in Mouse

Striatal Terminals. Toxicol Sci 153:79-88.

Loria F, Vargas JY, Bousset L, Syan S, Salles A, Melki R, Zurzolo C (2017) alpha-

Synuclein transfer between neurons and astrocytes indicates that astrocytes play

a role in degradation rather than in spreading. Acta Neuropathol 134:789-808.

Loser K, Mehling A, Loeser S, Apelt J, Kuhn A, Grabbe S, Schwarz T, Penninger JM,

Beissert S (2006) Epidermal RANKL controls regulatory T-cell numbers via

activation of dendritic cells. Nat Med 12:1372-1379. 241

Lu P, Wang YL, Linsley PS (1997) Regulation of self-tolerance by CD80/CD86

interactions. Curr Opin Immunol 9:858-862.

Luckheeram RV, Zhou R, Verma AD, Xia B (2012) CD4(+)T cells: differentiation and

functions. Clin Dev Immunol 2012:925135.

Luk KC, Song C, O'Brien P, Stieber A, Branch JR, Brunden KR, Trojanowski JQ, Lee

VM (2009) Exogenous alpha-synuclein fibrils seed the formation of Lewy body-

like intracellular inclusions in cultured cells. Proc Natl Acad Sci U S A 106:20051-

20056.

Luo Y, Su Y, Shen Y, Zhao L, Li K (2004) The levels of plasma IL-1beta, IL-6 of

C57BL/6J mice treated with MPTP and brain lateralization. Cell Mol Immunol

1:219-223.

Lupar E, Brack M, Garnier L, Laffont S, Rauch KS, Schachtrup K, Arnold SJ, Guery JC,

Izcue A (2015) Eomesodermin Expression in CD4+ T Cells Restricts Peripheral

Foxp3 Induction. J Immunol 195:4742-4752.

Luthman J, Fredriksson A, Sundström E, Jonsson G, Archer T (1989) Selective lesion of

central dopamine or noradrenaline neuron systems in the neonatal rat: motor

behavior and monoamine alterations at adult stage. Behav Brain Res 33:267-

277.

Lutz MB, Schuler G (2002) Immature, semi-mature and fully mature dendritic cells:

which signals induce tolerance or immunity? Trends Immunol 23:445-449.

Lutz MB, Suri RM, Niimi M, Ogilvie AL, Kukutsch NA, Rössner S, Schuler G, Austyn JM

(2000) Immature dendritic cells generated with low doses of GM-CSF in the

absence of IL-4 are maturation resistant and prolong allograft survival in vivo. Eur

J Immunol 30:1813-1822.

Ma Y, Zhan M, OuYang L, Li Y, Chen S, Wu J, Chen J, Luo C, Lei W (2014) The effects

of unilateral 6-OHDA lesion in medial forebrain bundle on the motor, cognitive 242

dysfunctions and vulnerability of different striatal interneuron types in rats. Behav

Brain Res 266:37-45.

Mahnke K, Schmitt E, Bonifaz L, Enk AH, Jonuleit H (2002) Immature, but not inactive:

the tolerogenic function of immature dendritic cells. Immunol Cell Biol 80:477-

483.

Maldonado RA, von Andrian UH (2010) How tolerogenic dendritic cells induce regulatory

T cells. Adv Immunol 108:111-165.

Mangoni ML, McDermott AM, Zasloff M (2016) Antimicrobial peptides and wound

healing: biological and therapeutic considerations. Exp Dermatol 25:167-173.

Manley NC, Caso JR, Works MG, Cutler AB, Zemlyak I, Sun G, Munhoz CD, Chang S,

Sorrells SF, Ermini FV, Decker JH, Bertrand AA, Dinkel KM, Steinberg GK,

Sapolsky RM (2013) Derivation of injury-responsive dendritic cells for acute brain

targeting and therapeutic protein delivery in the stroke-injured rat. PLoS One

8:e61789.

Manning-Bog AB, McCormack AL, Li J, Uversky VN, Fink AL, Di Monte DA (2002) The

herbicide paraquat causes up-regulation and aggregation of alpha-synuclein in

mice: paraquat and alpha-synuclein. J Biol Chem 277:1641-1644.

Mari ER, Rasouli J, Ciric B, Moore JN, Conejo-Garcia JR, Rajasagi N, Zhang GX,

Rabinovich GA, Rostami A (2016) Galectin-1 is essential for the induction of

MOG35-55 -based intravenous tolerance in experimental autoimmune

encephalomyelitis. Eur J Immunol 46:1783-1796.

Markey SP, Johannessen JN, Chiueh CC, Burns RS, Herkenham MA (1984)

Intraneuronal generation of a pyridinium metabolite may cause drug-induced

parkinsonism. Nature 311:464-467.

Martín-Fontecha A, Lanzavecchia A, Sallusto F (2009) Dendritic cell migration to

peripheral lymph nodes. Handb Exp Pharmacol 188:31-49. 243

Martinez GJ, Nurieva RI, Yang XO, Dong C (2008) Regulation and function of

proinflammatory TH17 cells. Ann N Y Acad Sci 1143:188-211.

Masliah E, Rockenstein E, Veinbergs I, Mallory M, Hashimoto M, Takeda A, Sagara Y,

Sisk A, Mucke L (2000) Dopaminergic loss and inclusion body formation in alpha-

synuclein mice: implications for neurodegenerative disorders. Science 287:1265-

1269.

Masliah E, Rockenstein E, Adame A, Alford M, Crews L, Hashimoto M, Seubert P, Lee

M, Goldstein J, Chilcote T, Games D, Schenk D (2005) Effects of alpha-synuclein

immunization in a mouse model of Parkinson's disease. Neuron 46:857-868.

Mastaglia FL, Johnsen RD, Byrnes ML, Kakulas BA (2003) Prevalence of amyloid-beta

deposition in the cerebral cortex in Parkinson's disease. Mov Disord 18:81-86.

Matsuoka Y, Vila M, Lincoln S, McCormack A, Picciano M, LaFrancois J, Yu X, Dickson

D, Langston WJ, McGowan E, Farrer M, Hardy J, Duff K, Przedborski S, Di

Monte DA (2001) Lack of nigral pathology in transgenic mice expressing human

alpha-synuclein driven by the tyrosine hydroxylase promoter. Neurobiol Dis

8:535-539.

Mayer-Scholl A, Averhoff P, Zychlinsky A (2004) How do neutrophils and pathogens

interact? Curr Opin Microbiol 7:62-66.

McCann H, Cartwright H, Halliday GM (2016) Neuropathology of alpha-synuclein

propagation and braak hypothesis. Mov Disord 31:152-160.

McCormack AL, Thiruchelvam M, Manning-Bog AB, Thiffault C, Langston JW, Cory-

Slechta DA, Di Monte DA (2002) Environmental Risk Factors and Parkinson's

Disease: Selective Degeneration of Nigral Dopaminergic Neurons Caused by the

Herbicide Paraquat. Neurobiology of Disease 10:119-127.

McGeer PL, Itagaki S, Akiyama H, McGeer EG (1988) Rate of cell death in parkinsonism

indicates active neuropathological process. Ann Neurol 24:574-576. 244

Meda L, Cassatella MA, Szendrei GI, Otvos LJ, Baron P, Villalba M, Ferrari D, Rossi F

(1995) Activation of microglial cells by beta-amyloid protein and interferon-

gamma. Nature 374:647-650.

Medzhitov R, Janeway CAJ (1997) Innate immunity: impact on the adaptive immune

response. Curr Opin Immunol 9:4-9.

Merad M, Sathe P, Helft J, Miller J, Mortha A (2013) The dendritic cell lineage: ontogeny

and function of dendritic cells and their subsets in the steady state and the

inflamed setting. Annu Rev Immunol 31:563-604.

Mestas J, Hughes CCW (2004) Of Mice and Not Men: Differences between Mouse and

Human Immunology. The Journal of Immunology 172:2731-2738.

Michell-Robinson MA, Touil H, Healy LM, Owen DR, Durafourt BA, Bar-Or A, Antel JP,

Moore CS (2015) Roles of microglia in brain development, tissue maintenance

and repair. Brain 138:1138-1159.

Migliore L, Scarpato R, Coppede F, Petrozzi L, Bonuccelli U, Rodilla V (2001)

Chromosome and oxidative damage biomarkers in lymphocytes of Parkinson's

disease patients. International Journal of Hygiene and Environmental Health

204:61-66.

Migliore L, Petrozzi L, Lucetti C, Gambaccini G, Bernardini S, Scarpato R, Trippi F,

Barale R, Frenzilli G, Rodilla V, Bonuccelli U (2002) Oxidative damage and

cytogenetic analysis in leukocytes of Parkinson's disease patients. Neurology

58:1809-1815.

Miklossy J, Doudet DD, Schwab C, Yu S, McGeer EG, McGeer PL (2006) Role of ICAM-

1 in persisting inflammation in Parkinson disease and MPTP monkeys. Exp

Neurol 197:275-283.

Minghetti L, Levi G (1998) Microglia as effector cells in brain damage and repair: focus

on prostanoids and nitric oxide. Prog Neurobiol 54:99-125. 245

Mizuno Y, Sone N, Saitoh T (1987) Effects of 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine and 1-methyl-4-phenylpyridinium ion on activities of the

enzymes in the electron transport system in mouse brain. J Neurochem 48:1787-

1793.

Mizuno Y, Hattori N, Kondo T, Nomoto M, Origasa H, Takahashi R, Yamamoto M,

Yanagisawa N (2017) A Randomized Double-Blind Placebo-Controlled Phase III

Trial of Selegiline Monotherapy for Early Parkinson Disease. Clin

Neuropharmacol 40:201-207.

Mizutani M, Pino PA, Saederup N, Charo IF, Ransohoff RM, Cardona AE (2012) The

fractalkine receptor but not CCR2 is present on microglia from embryonic

development throughout adulthood. J Immunol 188:29-36.

Mogi M, Harada M, Kondo T, Riederer P, Inagaki H, Minami M, Nagatsu T (1994)

Interleukin-1 beta, interleukin-6, epidermal growth factor and transforming growth

factor-alpha are elevated in the brain from parkinsonian patients. Neurosci Lett

180:147-150.

Mosley RL, Hutter-Saunders JA, Stone DK, Gendelman HE (2012) Inflammation and

adaptive immunity in Parkinson's disease. Cold Spring Harb Perspect Med

2:a009381.

Muller T (2015) Catechol-O-methyltransferase inhibitors in Parkinson's disease. Drugs

75:157-174.

Müller T, Blum-Degen D, Przuntek H, Kuhn W (1998) Interleukin-6 levels in

cerebrospinal fluid inversely correlate to severity of Parkinson's disease. Acta

Neurol Scand 98:142-144.

Munn DH, Mellor AL (2013) Indoleamine 2,3 dioxygenase and metabolic control of

immune responses. Trends Immunol 34:137-143. 246

Myers LM, Vella AT (2005) Interfacing T-cell effector and regulatory function through

CD137 (4-1BB) co-stimulation. Trends Immunol 26:440-446.

Nemunaitis J, Rabinowe SN, Singer JW, Bierman PJ, Vose JM, Freedman AS, Onetto

N, Gillis S, Oette D, Gold M, al. e (1991) Recombinant granulocyte-macrophage

colony-stimulating factor after autologous bone marrow transplantation for

lymphoid cancer. N Engl J Med 324:1773-1778.

Niranjan R, Nath C, Shukla R (2010) The mechanism of action of MPTP-induced

neuroinflammation and its modulation by melatonin in rat astrocytoma cells, C6.

Free Radic Res 44:1304-1316.

Niwa F, Kuriyama N, Nakagawa M, Imanishi J (2012) Effects of peripheral lymphocyte

subpopulations and the clinical correlation with Parkinson's disease. Geriatr

Gerontol Int 12:102-107.

Noda K, Kitami T, Gai WP, Chegini F, Jensen PH, Fujimura T, Murayama K, Tanaka K,

Mizuno Y, Hattori N (2005) Phosphorylated IkappaBalpha is a component of

Lewy body of Parkinson's disease. Biochem Biophys Res Commun 331:309-317.

Nomaguchi K, Suzu S, Yamada M, Hayasawa H, Motoyoshi K (2001) Expression of

Jagged1 gene in macrophages and its regulation by hematopoietic growth

factors. Exp Hematol 29:850-855.

Nonaka T, Iwatsubo T, Hasegawa M (2005) Ubiquitination of alpha-synuclein.

Biochemistry 44:361-368.

Noyce AJ, Bestwick JP, Silveira-Moriyama L, Hawkes CH, Giovannoni G, Lees AJ,

Schrag A (2012) Meta-analysis of early nonmotor features and risk factors for

Parkinson disease. Ann Neurol 72:893-901.

O'Mahony L, Akdis M, Akdis CA (2011) Regulation of the immune response and

inflammation by histamine and histamine receptors. J Allergy Clin Immunol

128:1153-1162. 247

Oertel W, Schulz JB (2016) Current and experimental treatments of Parkinson disease:

A guide for neuroscientists. J Neurochem 139 Suppl 1:325-337.

Ohshima Y, Yang LP, Uchiyama T, Tanaka Y, Baum P, Sergerie M, Hermann P,

Delespesse G (1998) OX40 costimulation enhances interleukin-4 (IL-4)

expression at priming and promotes the differentiation of naive human CD4(+) T

cells into high IL-4-producing effectors. Blood 92:3338-3345.

Okada R, Hara T, Sato T, Kojima N, Nishina Y (2016) The mechanism and control of

Jagged1 expression in Sertoli cells. Regenerative Therapy 3:75-81.

Olanow CW, Stern MB, Sethi K (2009) The scientific and clinical basis for the treatment

of Parkinson disease (2009). Neurology 72:S1-S136.

Olanow CW, Kieburtz K, Leinonen M, Elmer L, Giladi N, Hauser RA, Klepiskaya OS,

Kreitzman DL, Lew MF, Russell DS, Kadosh S, Litman P, Friedman H, Linvah N,

The PBSGF (2017) A randomized trial of a low-dose Rasagiline and Pramipexole

combination (P2B001) in early Parkinson's disease. Mov Disord.

Olejniczak K, Kasprzak A (2008) Biological properties of interleukin 2 and its role in

pathogenesis of selected diseases--a review. Med Sci Monit 14:RA179-189.

Olofsson T, Odeberg H, Olsson I (1976) Granulocyte function in chronic granulocytic

leukemia. II. Bactericidal capacity, phagocytic rate, oxygen consumption, and

granule protein composition in isolated granulocytes. Blood 48:581-593.

Olson KE, Kosloski-Bilek LM, Anderson KM, Diggs BJ, Clark BE, Gledhill JM, Jr.,

Shandler SJ, Mosley RL, Gendelman HE (2015) Selective VIP Receptor Agonists

Facilitate Immune Transformation for Dopaminergic Neuroprotection in MPTP-

Intoxicated Mice. J Neurosci 35:16463-16478.

Olsson I, Venge P (1974) Cationic proteins of human granulocytes. II. Separation of the

cationic proteins of the granules of leukemic myeloid cells. Blood 44:235-246. 248

Ookubo M, Yokoyama H, Kato H, Araki T (2009) Gender differences on MPTP (1-

methyl-4-phenyl-1,2,3,6-tetrahydropyridine) neurotoxicity in C57BL/6 mice. Mol

Cell Endocrinol 311:62-68.

Ookubo M, Yokoyama H, Takagi S, Kato H, Araki T (2008) Effects of estrogens on

striatal damage after 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)

neurotoxicity in male and female mice. Mol Cell Endocrinol 296:87-93.

Orihuela R, McPherson CA, Harry GJ (2016) Microglial M1/M2 polarization and

metabolic states. Br J Pharmacol 173:649-665.

Ou R, Yang J, Cao B, Wei Q, Chen K, Chen X, Zhao B, Wu Y, Song W, Shang H (2016)

Progression of non-motor symptoms in Parkinson's disease among different age

populations: A two-year follow-up study. J Neurol Sci 360:72-77.

Overton ET, Kang M, Peters MG, Umbleja T, Alston-Smith BL, Bastow B, Demarco-

Shaw D, Koziel MJ, Mong-Kryspin L, Sprenger HL, Yu JY, Aberg JA (2010)

Immune response to hepatitis B vaccine in HIV-infected subjects using

granulocyte-macrophage colony-stimulating factor (GM-CSF) as a vaccine

adjuvant: ACTG study 5220. Vaccine 28:5597-5604.

Paolicelli RC, Bolasco G, Pagani F, Maggi L, Scianni M, Panzanelli P, Giustetto M,

Ferreira TA, Guiducci E, Dumas L, Ragozzino D, Gross CT (2011) Synaptic

pruning by microglia is necessary for normal brain development. Science

333:1456-1458.

Papachroni KK, Ninkina N, Papapanagiotou A, Hadjigeorgiou GM, Xiromerisiou G,

Papadimitriou A, Kalofoutis A, Buchman VL (2007) Autoantibodies to alpha-

synuclein in inherited Parkinson's disease. J Neurochem 101:749-756.

Parker KH, Beury DW, Ostrand-Rosenberg S (2015) Myeloid-Derived Suppressor Cells:

Critical Cells Driving Immune Suppression in the Tumor Microenvironment. Adv

Cancer Res 128:95-139. 249

Parmiani G, Castelli C, Pilla L, Santinami M, Colombo MP, Rivoltini L (2007) Opposite

immune functions of GM-CSF administered as vaccine adjuvant in cancer

patients. Ann Oncol 18:226-232.

Peng J, Xie L, Stevenson FF, Melov S, Di Monte DA, Andersen JK (2006) Nigrostriatal

dopaminergic neurodegeneration in the weaver mouse is mediated via

neuroinflammation and alleviated by minocycline administration. J Neurosci

26:11644-11651.

Perrin RJ, Woods WS, Clayton DF, George JM (2000) Interaction of human alpha-

Synuclein and Parkinson's disease variants with phospholipids. Structural

analysis using site-directed mutagenesis. J Biol Chem 275:34393-34398.

Peters-Golden M, Canetti C, Mancuso P, Coffey MJ (2005) Leukotrienes:

Underappreciated Mediators of Innate Immune Responses. The Journal of

Immunology 174:589-594.

Petrozzi L, Lucetti C, Gambaccini G, Bernardini S, Del Dotto P, Migliore L, Scarpato R,

Bonuccelli U (2001) Cytogenetic analysis oxidative damage in lymphocytes of

Parkinson's disease patients. Neurol Sci 22:83-84.

Pierson ER, Goverman JM (2017) GM-CSF is not essential for experimental

autoimmune encephalomyelitis but promotes brain-targeted disease. JCI Insight

2:e92362.

Pike KA, Ratcliffe MJH (2002) Cell surface immunoglobulin receptors in B cell

development. Seminars in Immunology 14:351-358.

Ponomarev ED, Shriver LP, Maresz K, Dittel BN (2005) Microglial cell activation and

proliferation precedes the onset of CNS autoimmunity. J Neurosci Res 81:374-

389.

Pope RM, Leutz A, Ness SA (1994) C/EBP beta regulation of the tumor necrosis factor

alpha gene. J Clin Invest 94:1449-1455. 250

Povoleri GA, Scotta C, Nova-Lamperti EA, John S, Lombardi G, Afzali B (2013) Thymic

versus induced regulatory T cells - who regulates the regulators? Front Immunol

4:169.

Prado C, Contreras F, Gonzalez H, Diaz P, Elgueta D, Barrientos M, Herrada AA,

Lladser A, Bernales S, Pacheco R (2012) Stimulation of dopamine receptor D5

expressed on dendritic cells potentiates Th17-mediated immunity. J Immunol

188:3062-3070.

Prusiner SB, Woerman AL, Mordes DA, Watts JC, Rampersaud R, Berry DB, Patel S,

Oehler A, Lowe JK, Kravitz SN, Geschwind DH, Glidden DV, Halliday GM,

Middleton LT, Gentleman SM, Grinberg LT, Giles K (2015) Evidence for α-

synuclein prions causing multiple system atrophy in humans with parkinsonism.

Proc Natl Acad Sci USA 112:E5308-5317.

Przedborski S, Jackson-Lewis V, Yokoyama R, Shibata T, Dawson VL, Dawson TM

(1996) Role of neuronal nitric oxide in 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine (MPTP)-induced dopaminergic neurotoxicity. Proc Natl Acad

Sci U S A 93:4565-4571.

Przedborski S, Jackson-Lewis V, Naini AB, Jakowec M, Petzinger G, Miller R, Akram M

(2001) The parkinsonian toxin 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine

(MPTP): a technical review of its utility and safety. J Neurochem 76:1265-1274.

Qian L, Flood PM, Hong JS (2010) Neuroinflammation is a key player in Parkinson's

disease and a prime target for therapy. J Neural Transm (Vienna) 117:971-979.

Qureshi GA, Baig S, Bednar I, Södersten P, Forsberg G, Siden A (1995) Increased

cerebrospinal fluid concentration of nitrite in Parkinson's disease. Neuroreport

6:1642-1644.

Raker VK, Domogalla MP, Steinbrink K (2015) Tolerogenic Dendritic Cells for

Regulatory T Cell Induction in Man. Front Immunol 6:569. 251

Rannikko EH, Weber SS, Kahle PJ (2015) Exogenous alpha-synuclein induces toll-like

receptor 4 dependent inflammatory responses in astrocytes. BMC Neurosci

16:57.

Rascol O, Brooks DJ, Korczyn AD, De Deyn PP, Clarke CE, Lang AE (2000) A five-year

study of the incidence of dyskinesia in patients with early Parkinson's disease

who were treated with ropinirole or levodopa. 342 20.

Ray A, Dittel BN (2017) Mechanisms of Regulatory B cell Function in Autoimmune and

Inflammatory Diseases beyond IL-10. J Clin Med 6.

Reale M, Iarlori C, Thomas A, Gambi D, Perfetti B, Di Nicola M, Onofrj M (2009)

Peripheral cytokines profile in Parkinson's disease. Brain Behav Immun 23:55-

63.

Reichardt P, Dornbach B, Rong S, Beissert S, Gueler F, Loser K, Gunzer M (2007)

Naive B cells generate regulatory T cells in the presence of a mature

immunologic synapse. Blood 110:1519-1529.

Rentzos M, Nikolaou C, Andreadou E, Paraskevas GP, Rombos A, Zoga M, Tsoutsou A,

Boufidou F, Kapaki E, Vassilopoulos D (2007) Circulating interleukin-15 and

RANTES chemokine in Parkinson's disease. Acta Neurol Scand 116:374-379.

Rentzos M, Nikolaou C, Andreadou E, Paraskevas GP, Rombos A, Zoga M, Tsoutsou A,

Boufidou F, Kapaki E, Vassilopoulos D (2009) Circulating interleukin-10 and

interleukin-12 in Parkinson's disease. Acta Neurol Scand 119:332-337.

Reynolds AD, Banerjee R, Liu J, Gendelman HE, Mosley RL (2007) Neuroprotective

activities of CD4+CD25+ regulatory T cells in an animal model of Parkinson's

disease. J Leukoc Biol 82:1083-1094.

Reynolds AD, Stone DK, Hutter JA, Benner EJ, Mosley RL, Gendelman HE (2010)

Regulatory T cells attenuate Th17 cell-mediated nigrostriatal dopaminergic

neurodegeneration in a model of Parkinson's disease. J Immunol 184:2261-2271. 252

Ricciotti E, FitzGerald GA (2011) Prostaglandins and inflammation. Arterioscler Thromb

Vasc Biol 31:986-1000.

Rieker C, Dev KK, Lehnhoff K, Barbieri S, Ksiazek I, Kauffmann S, Danner S, Schell H,

Boden C, Ruegg MA, Kahle PJ, van der Putten H, Shimshek DR (2011)

Neuropathology in mice expressing mouse alpha-synuclein. PLoS One

6:e24834.

Rietdijk CD, Perez-Pardo P, Garssen J, van Wezel RJ, Kraneveld AD (2017) Exploring

Braak's Hypothesis of Parkinson's Disease. Front Neurol 8:37.

Robinson TE, Noordhoorn M, Chan EM, Mocsary Z, Camp DM, Whishaw IQ (1994)

Relationship between asymmetries in striatal dopamine release and the direction

of amphetamine-induced rotation during the first week following a unilateral 6-

OHDA lesion of the substantia nigra. Synapse 17:16-25.

Rodrigues RW, Gomide VC, Chadi G (2003) Striatal injection of 6-hydroxydopamine

induces retrograde degeneration and glial activation in the nigrostriatal pathway.

Acta Cirurica Brasileira 18:272-282.

Rosas M, Gordon S, Taylor PR (2007) Characterisation of the expression and function of

the GM-CSF receptor alpha-chain in mice. Eur J Immunol 37:2518-2528.

Rosborough BR, Castellaneta A, Natarajan S, Thomson AW, Turnquist HR (2012)

Histone deacetylase inhibition facilitates GM-CSF-mediated expansion of

myeloid-derived suppressor cells in vitro and in vivo. J Leukoc Biol 91:701-709.

Roth L, MacDonald JK, McDonald JW, Chande N (2012) Sargramostim (GM-CSF) for

induction of remission in Crohn's disease: a cochrane inflammatory bowel

disease and functional bowel disorders systematic review of randomized trials.

Inflamm Bowel Dis 18:1333-1339.

Rowe JM, Andersen JW, Mazza JJ, Bennett JM, Paietta E, Hayes FA, Oette D,

Cassileth PA, Stadtmauer EA, Wiernik PH (1995) A randomized placebo- 253

controlled phase III study of granulocyte-macrophage colony-stimulating factor in

adult patients (> 55 to 70 years of age) with acute myelogenous leukemia: a

study of the Eastern Cooperative Oncology Group (E1490). Blood 86:457-462.

Rutella S, Danese S, Leone G (2006) Tolerogenic dendritic cells: cytokine modulation

comes of age. Blood 108:1435-1440.

Ryan CW, Vogelzang NJ, Dumas MC, Kuzel T, Stadler WM (2000) Granulocyte-

macrophage-colony stimulating factor in combination immunotherapy for patients

with metastatic renal cell carcinoma: results of two phase II clinical trials. Cancer

88:1317-1324.

Sainathan SK, Hanna EM, Gong Q, Bishnupuri KS, Luo Q, Colonna M, White FV, Croze

E, Houchen C, Anant S, Dieckgraefe BK (2008) Granulocyte macrophage

colony-stimulating factor ameliorates DSS-induced experimental colitis. Inflamm

Bowel Dis 14:88-99.

Saiwai H, Kumamaru H, Ohkawa Y, Kubota K, Kobayakawa K, Yamada H, Yokomizo T,

Iwamoto Y, Okada S (2013) Ly6C+ Ly6G- Myeloid-derived suppressor cells play

a critical role in the resolution of acute inflammation and the subsequent tissue

repair process after spinal cord injury. J Neurochem 125:74-88.

Sallusto F, Lanzavecchia A (1994) Efficient presentation of soluble antigen by cultured

human dendritic cells is maintained by granulocyte/macrophage colony-

stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor

alpha. J Exp Med 179:1109-1118.

Samii A, Nutt JG, Ransom BR (2004) Parkinson's disease. The Lancet 363:1783-1793.

Santoli D, Clark SC, Kreider BL, Maslin PA, Rovera G (1988) Amplification of IL-2-driven

T cell proliferation by recombinant human IL-3 and granulocyte-macrophage

colony-stimulating factor. J Immunol 141:519-526. 254

Saraiva M, O'Garra A (2010) The regulation of IL-10 production by immune cells. Nat

Rev Immunol 10:170-181.

Saunders JA, Estes KA, Kosloski LM, Allen HE, Dempsey KM, Torres-Russotto DR,

Meza JL, Santamaria PM, Bertoni JM, Murman DL, Ali HH, Standaert DG,

Mosley RL, Gendelman HE (2012) CD4+ regulatory and effector/memory T cell

subsets profile motor dysfunction in Parkinson's disease. J Neuroimmune

Pharmacol 7:927-938.

Sawada M, Itoh Y, Suzumura A, Marunouchi T (1993) Expression of cytokine receptors

in cultured neuronal and glial cells. Neurosci Lett 160:131-134.

Scalzo P, Kummer A, Cardoso F, Teixeira AL (2010) Serum levels of interleukin-6 are

elevated in patients with Parkinson's disease and correlate with physical

performance. Neurosci Lett 468:56-58.

Schabitz WR, Kruger C, Pitzer C, Weber D, Laage R, Gassler N, Aronowski J, Mier W,

Kirsch F, Dittgen T, Bach A, Sommer C, Schneider A (2008) A neuroprotective

function for the hematopoietic protein granulocyte-macrophage colony

stimulating factor (GM-CSF). J Cereb Blood Flow Metab 28:29-43.

Schenk DB, Koller M, Ness DK, Griffith SG, Grundman M, Zago W, Soto J, Atiee G,

Ostrowitzki S, Kinney GG (2017) First-in-human assessment of PRX002, an anti-

alpha-synuclein monoclonal antibody, in healthy volunteers. Mov Disord 32:211-

218.

Scotcher KP, Irwin I, DeLanney LE, Langston JW, Di Monte D (1990) Effects of 1-

methyl-4-phenyl-1,2,3,6-tetrahydropyridine and 1-methyl-4-phenylpyridinium ion

on ATP levels of mouse brain synaptosomes. J Neurochem 54:1295-1301.

Serafini P, Mgebroff S, Noonan K, Borrello I (2008) Myeloid-derived suppressor cells

promote cross-tolerance in B-cell lymphoma by expanding regulatory T cells.

Cancer Res 68:5439-5449. 255

Shearer W (2003) Biology of common β receptor–signaling cytokines IL-3, IL-5, and GM-

CSF. Journal of Allergy and Clinical Immunology 112:653-665.

Shen YQ, Hebert G, Su Y, Moze E, Neveu PJ, Li KS (2005) In mice, production of

plasma IL-1 and IL-6 in response to MPTP is related to behavioral lateralization.

Brain Res 1045:31-37.

Shen Z, Reznikoff G, Dranoff G, Rock KL (1997) Cloned dendritic cells can present

exogenous antigens on both MHC class I and class II molecules. J Immunol

158:2723-2730.

Sheng JR, Muthusamy T, Prabhakar BS, Meriggioli MN (2011) GM-CSF-induced

regulatory T cells selectively inhibit anti-acetylcholine receptor-specific immune

responses in experimental myasthenia gravis. J Neuroimmunol 240-241:65-73.

Sheng JR, Li LC, Ganesh BB, Prabhakar BS, Meriggioli MN (2008) Regulatory T cells

induced by GM-CSF suppress ongoing experimental myasthenia gravis. Clin

Immunol 128:172-180.

Shimizu K, Ohtaki K, Matsubara K, Aoyama K, Uezono T, Saito O, Suno M, Ogawa K,

Hayase N, Kimura K, Shiono H (2001) Carrier-mediated processes in blood--

brain barrier penetration and neural uptake of paraquat. Brain Res 906:135-142.

Shimoji M, Zhang L, Mandir AS, Dawson VL, Dawson TM (2005) Absence of inclusion

body formation in the MPTP mouse model of Parkinson's disease. Brain Res Mol

Brain Res 134:103-108.

Shults CW (2006) Lewy bodies. Proc Natl Acad Sci U S A 103:1661-1668.

Sim WJ, Ahl PJ, Connolly JE (2016) Metabolism Is Central to Tolerogenic Dendritic Cell

Function. Mediators Inflamm 2016:2636701.

Simon DK, Simuni T, Elm J, Clark-Matott J, Graebner AK, Baker L, Dunlop SR, Emborg

M, Kamp C, Morgan JC, Ross GW, Sharma S, Ravina B, Investigators NN-P 256

(2015) Peripheral Biomarkers of Parkinson's Disease Progression and

Pioglitazone Effects. J Parkinsons Dis 5:731-736.

Simpson TR, Quezada SA, Allison JP (2010) Regulation of CD4 T cell activation and

effector function by inducible costimulator (ICOS). Curr Opin Immunol 22:326-

332.

Sims JE, Gayle MA, Slack JL, Alderson MR, Bird TA, Giri JG, Colotta F, Re F,

Mantovani A, Shaneback K (1993) Interleukin 1 signaling occurs exclusively via

the type I receptor. Proc Natl Acad Sci U S A 90:6155-6159.

Singer TP, Ramsay RR, McKeown K, Trevor A, Castagnoli NEJ (1988) Mechanism of

the neurotoxicity of 1-methyl-4-phenylpyridinium (MPP+), the toxic bioactivation

product of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP). Toxicology

49:17-23.

Singleton AB et al. (2003) alpha-Synuclein locus triplication causes Parkinson's disease.

Science 302:841.

Siracusa MC, Kim BS, Spergel JM, Artis D (2013) Basophils and allergic inflammation. J

Allergy Clin Immunol 132:789-801; quiz 788.

Spillantini MG, Crowther RA, Jakes R, Hasegawa M, Goedert M (1998) alpha-Synuclein

in filamentous inclusions of Lewy bodies from Parkinson's disease and dementia

with lewy bodies. Proc Natl Acad Sci U S A 95:6469-6473.

Spillantini MG, Schmidt ML, Lee VM, Trojanowski JQ, Jakes R, Goedert M (1997) Alpha-

synuclein in Lewy bodies. Nature 388:839-840.

Spitler LE, Weber RW, Allen RE, Meyer J, Cruickshank S, Garbe E, Lin HY, Soong SJ

(2009) Recombinant human granulocyte-macrophage colony-stimulating factor

(GM-CSF, sargramostim) administered for 3 years as adjuvant therapy of stages

II(T4), III, and IV melanoma. J Immunother 32:632-637. 257

Staal RG, Sonsalla PK (2000) Inhibition of brain vesicular monoamine transporter

(VMAT2) enhances 1-methyl-4-phenylpyridinium neurotoxicity in vivo in rat

striata. J Pharmacol Exp Ther 293:336-342.

Stefanova N, Fellner L, Reindl M, Masliah E, Poewe W, Wenning GK (2011) Toll-like

receptor 4 promotes alpha-synuclein clearance and survival of nigral

dopaminergic neurons. Am J Pathol 179:954-963.

Steinman RM (1991) The dendritic cell system and its role in immunogenicity. Annu Rev

Immunol 9:271-296.

Steinman RM, Hawiger D, Nussenzweig MC (2003) Tolerogenic dendritic cells. Annu

Rev Immunol 21:685-711.

Stevens CH, Rowe D, Morel-Kopp MC, Orr C, Russell T, Ranola M, Ward C, Halliday

GM (2012) Reduced T helper and B lymphocytes in Parkinson's disease. J

Neuroimmunol 252:95-99.

Stocchi F, Rascol O, Kieburtz K, Poewe W, Jankovic J, Tolosa E, Barone P, Lang AE,

Olanow CW (2010) Initiating levodopa/carbidopa therapy with and without

entacapone in early Parkinson disease: the STRIDE-PD study. Ann Neurol

68:18-27.

Stone KD, Prussin C, Metcalfe DD (2010) IgE, mast cells, basophils, and eosinophils. J

Allergy Clin Immunol 125:S73-80.

Straccia M, Gresa-Arribas N, Dentesano G, Ejarque-Ortiz A, Tusell JM, Serratosa J,

Sola C, Saura J (2011) Pro-inflammatory gene expression and neurotoxic effects

of activated microglia are attenuated by absence of CCAAT/enhancer binding

protein beta. J Neuroinflammation 8:156.

Su Dm, Manley NR (2000) Hoxa3 and Pax1 Transcription Factors Regulate the Ability of

Fetal Thymic Epithelial Cells to Promote Thymocyte Development. The Journal

of Immunology 164:5753-5760. 258

Su DM, Manley NR (2002) Stage-specific changes in fetal thymocyte proliferation during

the CD4-8- to CD4+8+ transition in wild type, Rag1-/-, and Hoxa3,Pax1 mutant

mice. BMC Immunol 3.

Su X, Maguire-Zeiss KA, Giuliano R, Prifti L, Venkatesh K, Federoff HJ (2008) Synuclein

activates microglia in a model of Parkinson's disease. Neurobiol Aging 29:1690-

1701.

Sugama S, Wirz SA, Barr AM, Conti B, Bartfai T, Shibasaki T (2004) Interleukin-18 null

mice show diminished microglial activation and reduced dopaminergic neuron

loss following acute 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine treatment.

Neuroscience 128:451-458.

Sundström E, Fredriksson A, Archer T (1990) Chronic neurochemical and behavioral

changes in MPTP-lesioned C57BL/6 mice: a model for Parkinson's disease.

Brain Res 528:181-188.

Surmeier DJ, Guzman JN, Sanchez-Padilla J, Schumacker PT (2011) The role of

calcium and mitochondrial oxidant stress in the loss of substantia nigra pars

compacta dopaminergic neurons in Parkinson's disease. Neuroscience 198:221-

231.

Takazoe M, Matsui T, Motoya S, Matsumoto T, Hibi T, Watanabe M (2009)

Sargramostim in patients with Crohn's disease: results of a phase 1-2 study. J

Gastroenterol 44:535-543.

Tarkowski E, Rosengren L, Blomstrand C, Wikkelsö C, Jensen C, Ekholm S, Tarkowski

A (1997) Intrathecal release of pro- and anti-inflammatory cytokines during

stroke. Clin Exp Immunol 110:492-499.

Teh HS, Kisielow P, Scott B, Kishi H, Uematsu Y, Blüthmann H, von Boehmer H (1988)

Thymic major histocompatibility complex antigens and the alpha beta T-cell

receptor determine the CD4/CD8 phenotype of T cells. Nature 335:229-233. 259

Thornton AM, Korty PE, Tran DQ, Wohlfert EA, Murray PE, Belkaid Y, Shevach EM

(2010) Expression of Helios, an Ikaros transcription factor family member,

differentiates thymic-derived from peripherally induced Foxp3+ T regulatory cells.

J Immunol 184:3433-3441.

Thorsteinsdottir U, Sauvageau G, Hough MR, Dragowska W, Lansdorp PM, Lawrence

HJ, Largman C, Humphries RK (1997) Overexpression of HOXA10 in murine

hematopoietic cells perturbs both myeloid and lymphoid differentiation and leads

to acute myeloid leukemia. 17 1.

Tillerson JL, Miller GW (2003) Grid performance test to measure behavioral impairment

in the MPTP-treated-mouse model of parkinsonism. Journal of Neuroscience

Methods 123:189-200.

Tone M, Powell MJ, Tone Y, Thompson SAJ, Waldmann H (2000) IL-10 Gene

Expression Is Controlled by the Transcription Factors Sp1 and Sp3. The Journal

of Immunology 165:286-291.

Torres-Aguilar H, Aguilar-Ruiz SR, Gonzalez-Perez G, Munguia R, Bajana S, Meraz-

Rios MA, Sanchez-Torres C (2010) Tolerogenic dendritic cells generated with

different immunosuppressive cytokines induce antigen-specific anergy and

regulatory properties in memory CD4+ T cells. J Immunol 184:1765-1775.

Tran DQ, Andersson J, Wang R, Ramsey H, Unutmaz D, Shevach EM (2009) GARP

(LRRC32) is essential for the surface expression of latent TGF-beta on platelets

and activated FOXP3+ regulatory T cells. Proc Natl Acad Sci U S A 106:13445-

13450.

Tremblay ME, Stevens B, Sierra A, Wake H, Bessis A, Nimmerjahn A (2011) The role of

microglia in the healthy brain. J Neurosci 31:16064-16069.

Tripathi SK, Lahesmaa R (2014) Transcriptional and epigenetic regulation of T-helper

lineage specification. Immunol Rev 261:62-83. 260

Ubogu EE, Cossoy MB, Ransohoff RM (2006) The expression and function of

chemokines involved in CNS inflammation. Trends Pharmacol Sci 27:48-55.

Uéda K, Fukushima H, Masliah E, Xia Y, Iwai A, Yoshimoto M, Otero DA, Kondo J, Ihara

Y, Saitoh T (1993) Molecular cloning of cDNA encoding an unrecognized

component of amyloid in Alzheimer disease. Proc Natl Acad Sci USA 90:11282-

11286.

Ungerstedt U (1968) 6-Hydroxy-dopamine induced degeneration of central monoamine

neurons. Eur J Pharmacol 5:107-110.

Ushach I, Zlotnik A (2016) Biological role of granulocyte macrophage colony-stimulating

factor (GM-CSF) and macrophage colony-stimulating factor (M-CSF) on cells of

the myeloid lineage. J Leukoc Biol 100:481-489.

Uversky VN, Li J, Fink AL (2001) Evidence for a partially folded intermediate in alpha-

synuclein fibril formation. J Biol Chem 276:10737-10744.

Uversky VN, Yamin G, Munishkina LA, Karymov MA, Millett IS, Doniach S, Lyubchenko

YL, Fink AL (2005) Effects of nitration on the structure and aggregation of alpha-

synuclein. Brain Res Mol Brain Res 134:84-102.

Vaday GG, Lider O (2000) Extracellular matrix moieties, cytokines, and enzymes:

dynamic effects on immune cell behavior and inflammation. J Leukoc Biol

67:149-159.

Valentine JF, Fedorak RN, Feagan B, Fredlund P, Schmitt R, Ni P, Humphries TJ (2009)

Steroid-sparing properties of sargramostim in patients with corticosteroid-

dependent Crohn's disease: a randomised, double-blind, placebo-controlled,

phase 2 study. Gut 58:1354-1362. van der Putten H, Wiederhold KH, Probst A, Barbieri S, Mistl C, Danner S, Kauffmann S,

Hofele K, Spooren WP, Ruegg MA, Lin S, Caroni P, Sommer B, Tolnay M, Bilbe 261

G (2000) Neuropathology in mice expressing human alpha-synuclein. J Neurosci

20:6021-6029.

Vasu C, Dogan RNE, Holterman MJ, Prabhakar BS (2003) Selective Induction of

Dendritic Cells Using Granulocyte Macrophage-Colony Stimulating Factor, But

Not fms-Like Tyrosine Kinase Receptor 3-Ligand, Activates Thyroglobulin-

Specific CD4+/CD25+ T Cells and Suppresses Experimental Autoimmune

Thyroiditis. The Journal of Immunology 170:5511-5522.

Vilar M, Chou HT, Luhrs T, Maji SK, Riek-Loher D, Verel R, Manning G, Stahlberg H,

Riek R (2008) The fold of alpha-synuclein fibrils. Proc Natl Acad Sci U S A

105:8637-8642.

Villani AC et al. (2017) Single-cell RNA-seq reveals new types of human blood dendritic

cells, monocytes, and progenitors. Science 356.

Vinay DS, Kwon BS (1998) Role of 4-1BB in immune responses. Semin Immunol

10:481-489.

Waller EK (2007) The role of sargramostim (rhGM-CSF) as immunotherapy. Oncologist

12 Suppl 2:22-26.

Walsh S, Finn DP, Dowd E (2011) Time-course of nigrostriatal neurodegeneration and

neuroinflammation in the 6-hydroxydopamine-induced axonal and terminal lesion

models of Parkinson's disease in the rat. Neuroscience 175:251-261.

Wang Y et al. (2012) Phosphorylated alpha-synuclein in Parkinson's disease. Sci Transl

Med 4:121ra120.

Wang ZY, Sato H, Kusam S, Sehra S, Toney LM, Dent AL (2005) Regulation of IL-10

Gene Expression in Th2 Cells by Jun Proteins. The Journal of Immunology

174:2098-2105. 262

Watanabe K, Jose PJ, Rankin SM (2002) Eotaxin-2 Generation Is Differentially

Regulated by Lipopolysaccharide and IL-4 in Monocytes and Macrophages. The

Journal of Immunology 168:1911-1918.

Watford WT, Hissong BD, Bream JH, Kanno Y, Muul L, O'Shea JJ (2004) Signaling by

IL-12 and IL-23 and the immunoregulatory roles of STAT4. Immunol Rev

202:139-156.

Wesselingh SL, Gough NM, Finlay-Jones JJ, McDonald PJ (1990) Detection of cytokine

mRNA in astrocyte cultures using the polymerase chain reaction. Lymphokine

Res 9:177-185.

Whitton PS (2007) Inflammation as a causative factor in the aetiology of Parkinson's

disease. Br J Pharmacol 150:963-976.

Wissemann WT, Hill-Burns EM, Zabetian CP, Factor SA, Patsopoulos N, Hoglund B,

Holcomb C, Donahue RJ, Thomson G, Erlich H, Payami H (2013) Association of

Parkinson disease with structural and regulatory variants in the HLA region. Am J

Hum Genet 93:984-993.

Wong BR, Josien R, Lee SY, Sauter B, Li HL, Steinman RM, Choi Y (1997) TRANCE

(tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF

family member predominantly expressed in T cells, is a dendritic cell-specific

survival factor. J Exp Med 186:2075-2080.

Worbs T, Forster R (2007) A key role for CCR7 in establishing central and peripheral

tolerance. Trends Immunol 28:274-280.

Wu DC, Jackson-Lewis V, Vila M, Tieu K, Teismann P, Vadseth C, Choi DK,

Ischiropoulos H, Przedborski S (2002) Blockade of microglial activation is

neuroprotective in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mouse model

of Parkinson disease. J Neurosci 22:1763-1771. 263

Xu Y, Hunt NH, Bao S (2008) The role of granulocyte macrophage-colony-stimulating

factor in acute intestinal inflammation. Cell Res 18:1220-1229.

Yamagata T, Nishida J, Sakai R, Tanaka T, Yazaki Y, Hirai H (1997) Of the GATA-

binding proteins, only GATA-4 selectively regulates the human IL-5 gene

promoter in IL-5 producing cells which express multiple GATA-binding proteins.

Leukemia 11:501-502.

Yamagata T, Nishida J, Sakai R, Tanaka T, Honda H, Hirano N, Mano H, Yazaki Y, Hirai

H (1995) Of the GATA-binding proteins, only GATA-4 selectively regulates the

human interleukin-5 gene promoter in interleukin-5-producing cells which express

multiple GATA-binding proteins. Mol Cell Biol 15:3830-3839.

Yamin G, Uversky VN, Fink AL (2003) Nitration inhibits fibrillation of human α-synuclein

in vitro by formation of soluble oligomers. FEBS Letters 542:147-152.

Yang J, Wang N (2015) Genome-wide expression and methylation profiles reveal

candidate genes and biological processes underlying synovial inflammatory

tissue of patients with osteoarthritis. Int J Rheum Dis 18:783-790.

Yi H, Zhao Y (2007) Chemokines, chemokine receptors and CD4+CD25+ regulatory T

cells. Expert Rev Clin Immunol 3:343-349.

Yoritaka A, Hattori N, Uchida K, Tanaka M, Stadtman ER, Mizuno Y (1996)

Immunohistochemical detection of 4-hydroxynonenal protein adducts in

Parkinson disease. Proc Natl Acad Sci U S A 93:2696-2701.

Yoshie O, Matsushima K (2015) CCR4 and its ligands: from bench to bedside. Int

Immunol 27:11-20.

Zamarron BF, Chen W (2011) Dual Roles of Immune Cells and Their Factors in Cancer

Development and Progression. Int J Biol 7:651-658.

Zhang JM, An J (2007) Cytokines, inflammation, and pain. Int Anesthesiol Clin 45:27-37. 264

Zhang W, Wang T, Pei Z, Miller DS, Wu X, Block ML, Wilson B, Zhang W, Zhou Y, Hong

JS, Zhang J (2005) Aggregated alpha-synuclein activates microglia: a process

leading to disease progression in Parkinson's disease. FASEB J 19:533-542.

Zhao C, Ling Z, Newman MB, Bhatia A, Carvey PM (2007) TNF-alpha knockout and

minocycline treatment attenuates blood-brain barrier leakage in MPTP-treated

mice. Neurobiol Dis 26:36-46.

Zheng Y, Chaudhry A, Kas A, deRoos P, Kim JM, Chu TT, Corcoran L, Treuting P, Klein

U, Rudensky AY (2009) Regulatory T-cell suppressor program co-opts

transcription factor IRF4 to control T(H)2 responses. Nature 458:351-356.

Zhou TT, Zu G, Wang X, Zhang XG, Li S, Liang ZH, Zhao J (2015) Immunomodulatory

and neuroprotective effects of ginsenoside Rg1 in the MPTP(1-methyl-4-phenyl-

1,2,3,6-tetrahydropyridine) -induced mouse model of Parkinson's disease. Int

Immunopharmacol 29:334-343.

Zhu J, Paul WE (2008) CD4 T cells: fates, functions, and faults. Blood 112:1557-1569.

Ziegler-Heitbrock L, Lotzerich M, Schaefer A, Werner T, Frankenberger M, Benkhart E

(2003) IFN- Induces the Human IL-10 Gene by Recruiting Both IFN Regulatory

Factor 1 and Stat3. The Journal of Immunology 171:285-290.

Ziegler-Heitbrock L, Ancuta P, Crowe S, Dalod M, Grau V, Hart DN, Leenen PJ, Liu YJ,

MacPherson G, Randolph GJ, Scherberich J, Schmitz J, Shortman K, Sozzani S,

Strobl H, Zembala M, Austyn JM, Lutz MB (2010) Nomenclature of monocytes

and dendritic cells in blood. Blood 116:e74-80.