University of Southern California Dissertations and Theses

Adaptive immunity in the central nervous system

(USC Thesis Other)

Adaptive immunity in the central nervous system

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2020 Kwok Wai Im
Adaptive Immunity in the Central Nervous System
by
Kwok (Chris) Wai Im

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
Neuroscience

August 2020

ii

Acknowledgements
I would like to thank the neuroscience graduate program for these past six years of
learning, creating, and building the framework for success as I pursue a career in academia. I will
never forget the beginning of this incredible journey. It was during the summer prior to starting
the PhD where I was given the opportunity to do a 10-week research internship, Bridging the
Gaps, at USC. I had always wanted to come to California to experience the sun, the fun beaches
to play volleyball, and to understand what makes Los Angeles such a special place. Dr. Joyce
Richey provided me the space to learn and to grow. Before leaving the program, she gave me a
mock interview to prepare me for PhD interviews and asked me: Where do you draw inspiration
for science when times get difficult? Instinctively, I said “my grandfather…he passed away of
Alzheimer’s disease before I graduated from the University of Texas at Dallas with a bachelor’s
degree in neuroscience.” I knew from then on, that my aspirations for science was to help the
world heal and lessen human suffering from this deadly disease…in my own way. USC gave me
more than just an education and a more inquisitive mind…it gave me a place to call home.
I will never forget the thought provoking questions Dr. Caleb Finch asked me the first
time we met during the USC interview weekend: “Do you see yourself as a glass half empty or a
glass half full? What would your research be if I gave you 3 post docs, 2 graduate students? How
would you mentor them?” Arguably the hardest questions I have ever had to answer! “I am a
glass completely empty…no clue what to do for a research direction…and the lab may as well
run itself as I have no experience of handling other peoples’ careers let alone my own, frankly.
Now, I still consider myself with an empty glass, but only because I’ve learned to keep it empty
after always learning new questions to address and having an open mind to new ideas. I have a
iii

flexible idea of how-to mentor each person in my lab to cater towards their personalities, work
ethic, and resolve. Finally, my mentorship style would be that of acceptance and nurture strong
minds of the next generation. I owe these lessons and many more to Dr. Finch, Dr. Pike, Dr.
Shao, Dr. Kaslow, and Dr. Terrence Town. The journey in this PhD has been far from simple to
say the least. I can recall at least three times during my tenure I wanted to say, “I’m done”.
However, it is with enormous gratitude that I had the support of my family, friends in Texas, and
of course, my friends here in the program and California. I attribute my reasoning as to why my
mentorship style and my outlook of science is the way it is. “Science is difficult…even when
you’ve worked to the brink of exhaustion and poured your blood, sweat, and tears into
everything, it is not enough.” However, a support of family and friends is what gets you through
even the worst of times…and to me, it has made me a stronger person. Alexis and Mariana, two
very important people that have been with me since the beginning and gave me inspiration to
strive to be even a fraction of their intellect and compassion for science and their friends. I’m far
from perfect, but we learned in this journey that our imperfections are a strength to draw upon
happy and dark times.
An incredible thank you goes out to Dr. Terrence Town. After everything that has
happened, I have the resolve to always be kind and understanding to all who take me in as a
student. It is my imperfection to be too naïve and loyal that provides me the ability to look back
and say, “Maybe I’d approach things differently…but I will not regret who I am.” Confidence
has always been my biggest weakness. I cannot thank Dr. Town enough for showing me that I
am capable and worth more than I knew I had the potential for…whether that be directly or
passively through my trials and tribulations. We may have had disagreements, arguments, and
heated discussions, but all has been taken in as a lesson to become better. I wish you the best in
iv

your endeavors. Finally, to my Grandpa. A man who fed me, raised me, and laughed with me as
I grew up in Texas. A teacher who listened to all my childish complaints, crazy ideas, and
failures. Never in our family had someone gotten a high school diploma…let alone go to college
and pursue a PhD. Though you may not have seen my graduate with that college degree, it is
with your guidance and love that made me realize half way through my education that I do this
for you and others that wish to experience the love of a grandparent in their life. You taught me
to always smile, and that life is about collaboration, not competition. I know from experience
that this habit was the hardest for me to break…comparing myself to family, to schoolmates, to
anyone. Only when I had truly learned this lesson did I start loving myself and moving forward
once you passed away. I will always remember you as my guardian angel. Thank you for
everything Grandpa.

v

Table of Contents
Acknowledgements………………………………………………………………………………ii
List of Tables…………………………………………………………………………...…..........ix
List of Figures…………………………………………………………………………………….x
Abbreviations……………………………………………………………………………..…….xii
Chapter 1. Adaptive immunity in Alzheimer’s disease………………………………………..1
Introduction………………………………………………………………………………1
1.1 Alzheimer’s disease…………………………………………………………..1
1.1.1 Alzheimer’s disease pathology…………………………………….5
1.2 Innate immunity………………………………………………………….…..7
1.2.1 Soluble mediators of innate immunity……………………………7
1.2.2 Innate immune mediators of phagocytosis…………………….…9
1.2.3 Innate immune processing and presentation of antigens………10
1.3 Adaptive immunity…………………………………………………………11
1.3.1 Innate-adaptive immune crosstalk………………………………12
1.3.2 Transendothelial migration of T cells into the CNS……………13
1.3.3 T cell influences in AD……………………………………………14
1.3.4 CD4
+
T helper cell involvement in AD…………………………..15
vi

1.3.5 T helper 17 T cells………………………………..……………….17
1.3.6 Th17 cell hypothesis in AD………………….……………………18
1.4 Conclusion…………………………………………………….…………….19
Th17 cells exacerbate Alzheimer’s disease-like pathology………………….………..22
1.5 Abstract……………………………………………….……………………..23
1.6 Results…………………………………………………………………...…..23
1.6.1 CD4
+
T cells are elevated in APP/PS1
+
mouse brains………….23
1.6.2 CD4
cre
STAT3
fl/fl
depletion of Th17 cells…………………………24
1.6.3 Decreased amyloid burden in APP/PS1
+
CD4
cre
STAT3
fl/fl

brains…………………………………………………………………….25

1.6.4 Cerebral innate immune cells phagocytose Aβ in absence of Th17
cells………………………………………………………………...…….25

1.6.5 Generation of Aβ-specific Th17 cells and mouse model………...26
1.6.6 Neutralization of IL-17 attenuates the effects of Aβ-Th17 cells on
plaque pathology………………………………………………………..27

1.6.7 Neutralization of IL-17 rescues the effect of Aβ-Th17 cells on
phagocytosis……………………………...……………………………..28

1.7 Discussion…………………………………………………………...………29
1.8 Conclusion…………………………………………………………………..33
vii

TGF-β signaling control of the T cell response to cerebral amyloid-β………………………44
1.9 Abstract……………………………………………………………………...44
1.20 Results……………………………………………………………………...45
1.20.1 Generation of a mouse model of cerebral amyloidosis expressing
a dominant negative form of TGF-β receptor in CD4
+
T
cells………………………………………………………………………45

1.20.2 Reduced amyloid burden in APP/PS1
+
CD4-DNR
+
mouse
brains…………………………………………………………………….46

1.20.3 Infiltration of T cells into brains of APP/PS1
+
CD4-DNR
+

mice………………………………………………………………………47

1.20.4 APP/PS1
+
CD4-DNR
+
mice exhibit early death……………….47
1.21 Discussion………………………………………………...………………..48
1.22 Conclusion…………………………………………………………………49
References……………………………………………………………………………….53
Chapter 2 Neural stem cell immune tolerance in graft versus host disease…………………75
HLA-A critically determines neural stem cell transplantation immune tolerance....75
2.1 Abstract……………………………………………………………………..77
2.2 Introduction…………………………………………………………………77
2.3 Results……………………………………………………………………….79
viii

2.3.1 Establishing and validating mice with human adaptive immune
systems………………………………………………….……………….79

2.3.2 Selective HLA-A expression by neural progenitors….…………80
2.3.3 Partial HLA-A matching immunologically tolerizes neural
transplants………………………………………….…………………...81

2.3.4 HLA-A impacts neuroinflammatory landscape after neural
transplantation………………………………………….………………82

2.4 Discussion………………………………………………..…………………..83
2.5 Conclusion…………………………………………………..……………….85
References……………………………………………………………….………………91
Chapter 3 Adaptive immunity in brain cancer…………………………………….…………97
Short-lived effector T cells in a mouse model of medulloblastoma………….………97
3.1 Introduction…………………………………………………………………97
References……………………………………….……..………………………………100
Chapter 4 Methods…………………………………………………………………………….104
Appendix……………………………………………………………………………………..116

ix

List of Tables
Table 1 Co-linear curve analysis ..……………………………………………….………….85

x

List of Figures
Intro Figure 1……………………………………………………………………………………20
Intro Figure 2……………………………………………………………………………………21
1.1 Increased CD4
+
T cells in APP/PS1 brains………………………………………………..34
1.2 Absence of Th17 cells in CD4
cre
STAT3
fl/fl
mice……………………………………………35
1.3 Th17 deletion reduces amyloid burden in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice……………..36
1.4 Increased Aβ phagocytosis in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice…………...……………..37
1.5 Generation of Aβ-specific Th17 cells….………………………………………………..….38
1.6 Th17-dependent exacerbation of amyloid burden…………………….…………………..39
1.7 Th17-dependent Aβ phagocytosis……………...……………………………………….….40
1.S1 Roryt
+
depletion in CD4
+
T cells………………………………………….……………...41
1.S2 Increased microgliosis in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice……………….…………….41
1.S3 Decreased astrogliosis in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice……………….…….……….42
1.S4 CD4
cre
STAT3
fl/fl
does not alter PS1 abundance…………………………….……………42
1.S5 Microglial IL23 expression in vitro……………………………………………....……….43
1.S6 Inflammatory cytokine panel for primary microglia-Th17 in vitro coculture…….…..43
1.8 Generation of CD4
+
T cell TGF-β signaling………………………………………………50
1.9 Reduced amyloid burden in APP/PS1
+
CD4-DNR
+
mice…………………………………50
xi

1.10 Infiltration of CD3
+
T cells correlated with amyloid burden…………………………51
1.11 APP/PS1
+
CD4-DNR
+
mice display early death………………………………………..52
2.1 Generating mice with human immune systems for neural stem cell transplantation..86
2.2 Human neural progenitors selectively express HLA-A………………...……………….87
2.3 Partial HLA-A matching tolerizes against human neural transplant rejection.……...88
2.4 Unsupervised transcriptomic analysis of human neural progenitors………………...90

xii

Abbreviations list
AD-Alzheimer’s disease
MCI-Mild cognitive impairment
Aβ-Amyloid beta
CSF-Cerebrospinal fluid
PBMCs-Peripheral blood mononuclear cells
PET-Positron emission tomography
MRI-Magnetic resonance imaging
MMSE-Mini mental state exam
FAD-Familial Alzheimer’s disease
EOAD-Early onset Alzheimer’s disease
APP-Amyloid precursor protein
PSEN1-Presenilin 1
PSEN2-Presenilin 2
5xFAD-Five AD-linked familial Alzheimer’s disease
3xTG-AD-Three AD-linked transgenic
APP/PS1-Amyloid precursor protein/presenilin 1 mutation
LOAD-Late onset Alzheimer’s disease
APOE4-Apolipoprotein epsilon 4
CNS-Central nervous system
ILs-Interleukins
TNF-Tumor necrosis factor
INF-Interferons
TGF-Transforming growth factor
CCL2-C-C motif chemokine ligand 2
GM-CSF-Granulocyte macrophage-colony stimulating factor
TREM2-Triggering receptor expressed on myeloid cells 2
xiii

C1Q-Complement component 1q
PAMP-Pathogen associated molecular pattern
DC-Dendritic cell
APC-Antigen presenting cell
MHC-I-Major histocompatibility complex I
MHC-II-Major histocompatibility complex II
HLA-DR-Human leukocyte antigen DR
HLA-DQ-Human leukocyte antigen DQ
BBB-Blood brain barrier
CAMs-cellular adhesion molecules
CXCR2-C-X-C motif chemokine receptor 2
Th1-T helper 1 cell
Th2-T helper 2 cell
Treg-T regulatory cell
Th17-T helper 17 cell
IFNy-Interferon gamma
IL-17A-Interluekin 17A
IL-21-Interleukin 21
IL-22-Interluekin 22
RORyt-RAR-mediated organ receptor gamma t
STAT3-Signal transducer and activator of transcription 3
EAE-Experimental autoimmune encephalomyelitis
RA-Rheumatoid arthritis
BP-Bordatella pertussis
IBD-Inflammatory bowel disease
IL-23R-Interleukin 23 receptor
Aβ-Th17-Amyloid beta T helper 17 cells
ThioS-Thioflavin S
xiv

CFA-Complete Freund’s adjuvant
SEM-Standard error of the mean
HLA-Human leukocyte antigen
hESCs-human embryonic stem cells
hNPCs-human neural progenitor cells
hHSCs-human hematopoietic cells
CoLC-Co-linear curve analysis
GVHD-Graft versus host disease
CD8-Cluster differentiation 8
PEG-PLGA-Polyethylene glycol-poly lactic acid co-glycolic acid
TCR-T cell receptor

1

Chapter 1 Adaptive immunity in Alzheimer’s disease
Introduction
1.1 Alzheimer’s disease
Alzheimer’s disease (AD) is a debilitating neurodegenerative disease that currently
affects approximately 5.8 million Americans, which is projected to increase to a
staggering 13.8 million people by 2050. Of those affected,
5.6 million people are over the age of 65 and almost two-thirds of all AD cases are
women. Staggeringly, the financial costs of the disease have been estimated to
exponentially increase to $1.1 trillion (‘2019 Alzheimer’s disease facts and figures’,
2019), which would overwhelm our current health and social care systems leading into
the 21
st
century.

Three stages have been defined in the progression of AD: 1) preclinical Alzheimer’s
disease or cognitively normal, 2) mild cognitive impairment (MCI), and 3) dementia
from Alzheimer’s disease (Bature et al., 2017). In the early phases of the disease, brain
matter shrinkage becomes apparent, and patients develop mild to severe symptoms of
cognitive impairment and increased clinical care as described in Figure 1. According
to the diagnostic criteria for AD from the Alzheimer’s Association, symptoms of
cognitive impairment include increased difficulty in planning and problem solving,
recognizing and completing familiar daily tasks, confusion in spatial awareness, poor
judgement, social withdrawal, and psychiatric instability with mood and personality
(Weller and Budson, 2018). For example, signs of Alzheimer’s disease symptoms
2

could be presented as poor judgement and decision making, inability to manage a
budget, losing track of the date or season, difficulty having conversations, or
misplacing things and being unable to retrace steps to find them. These clinical
manifestations are differentiated from the juxtaposed normal age-related changes such
as making bad decisions once in a while, missing a monthly payment, forgetting which
day it is and remembering it later, sometimes forgetting which word to use, and losing
things from time to time. According to the Diagnostic and Statistical Manual of Mental
Health Disorders, the term “dementia due to AD” is replaced with major
neurocognitive disorder and mild cognitive disorder, focusing on the decline, rather
than a deficit, in function. For example, the criteria focus less on memory impairment,
allowing for variables associated with conditions that sometimes begin with declines
in speech or language ability that ultimately lead to interfering with independence of
the individual (American Psychiatric Association, 2020). In preclinical AD, the brain
compensates initially and enables individuals to continue normal daily activities.
However, as the accumulation of damage occurs, the brain can no longer compensate
for these changes, and patients begin to show subtle decline in cognitive function
(Reddy a n d Beal, 2008).

Familial Alzheimer’s disease (FAD), also known as early-onset AD (EOAD)
comprises 5% of all AD cases (Piaceri et al., 2013). EOAD is characterized as genetic
acquisition of an autosomal-dominant inheritance that leads to early presentations of
dementia before the age of 65 (Cacace et al., 2016). These genetic mutations on three
proteins have been identified with high fidelity to cause AD: amyloid precursor protein
3

(APP), presenilin 1 (PSEN1), and presenilin 2 (PSEN2). These mutations in EOAD
occur either in the substrate (APP) or the protease (PSEN1 and PSEN 2) of the
enzymatic reaction that generates a thirty-nine to forty-two amino acid amyloid-β (Aβ)
peptide (Selkoe, 2001). Under normal conditions, APP is cleaved enzymatically by α-
secretases and forms APPα, a soluble fragment that has been associated with neuronal
development, stability, and protection (Chasseigneaux and Allinquant, 2012).
Moreover, soluble APPα has been shown to inhibit tau phosphorylation through the
modulation of glycogen synthase kinase-3 (GSK3), a ubiquitous serine/threonine
kinase that regulates an array of fundamental cell processes (Juan Deng et al., 2015)
and directly associating with beta amyloid cleaving enzyme 1 (BACE1) to decrease β-
secretase activity and Aβ generation (Obregon et al., 2012). Under pathological
conditions, APP is cleaved enzymatically by β- secretases and γ-secretases to form Aβ
peptides, leading to synaptic damage. (Zhang et al., 2011). The proposed mechanism
of synaptic damage has been postulated to be from toxic Aβ species promoting reactive
oxygen species in the microenvironment, leading to dysmorphic neuronal
morphologies and function such as Aβ inducing AMPA receptor ubiquitination a
degradation in neurons, a process that mediates synaptic transmission (Zhang et al.,
2018).

In AD, predominant Aβ species are forty and forty-two amino acids in length, Aβ40
and Aβ42 respectively. Aβ40 is prone for vascular deposition and exacerbation in
cerebral angiopathy, while the hydrophobicity of the final two amino acids in Aβ42 are
highly susceptible to aggregate and build into complex Aβ conformations over time.
4

These stages of Aβ peptide conformations are monomers, dimers, oligomers, fibrils,
to finally dense core Aβ plaques. Aβ40 neurotoxicity and aggregation contain
primarily of monomer/dimer mixtures, while Aβ42 consists a range of low-order
oligomeric species that suggest the highest toxic potential of Aβ species due to its
stochastic aggregation behavior compared to Aβ40 (Mclean, et al., 1999). The
production of Aβ40 and Aβ42 are important catalysts for neuroinflammation in the
early stages of AD pathogenesis leading to neurodegeneration (Cappai and Barnham,
2008). As a result, these toxic species have been well-studied and generated in a
multitude of animal models that recapitulate cerebral amyloidosis, such as five AD-
linked familial Alzheimer’s disease (5xFAD:APPswe, APPFlorida, APPLondon, PSEN1A>C,
PSEN1L286v) (Eimer and Vassar, 2013), three AD-linked transgenic (3xTg-AD:APP,
PSEN1M146V, MAPT) (Sterniczuk et al., 2010), and amyloid precursor protein Swedish
mutation presenilin 1 delta E9 mutation (APPswe/PS1ΔE9) (Jankowsky et al., 2004).

Late-onset AD (LOAD) represents 95% of all cases in AD and is a multifactorial
disease with a plethora of well- studied genetic, epigenetic, and environmental, risk
factors. The greatest risk factor is aging, as the prevalence of AD approximately
doubles with every five years after age 65 (Palasi et al., 2015). The most prevalent
genetic risk factor for LOAD is the ε4 allele of the cholesterol transporter
apolipoprotein E (APOE4) (Ma et al., 1996; Zhu et al., 2015; Mahley and Huang,
2006). These along with other major risk factors, such as sex, obesity, and air
pollution, do not entirely drive AD risk. Rather, there are multiple risk factors that
interact to determine the AD risk (Uchoa et al., 2016). Identifying risk factors and
5

investigating the mechanisms underlying the disease have been a major focus on
research leading into the 21
st
century (Holmes, 2013). For example, positive
correlation with AD risk has been associated with poor socioeconomic status
(Mortimer and Graves, 1993), traumatic brain injury (Lye and Shores, 2000), gut
dysbiosis (Zhao et al., 2017), and type 2 diabetes (Exalto et al., 2012; Hann, 2006).
However, greater physical exercise (Paillard et al., 2015; Radak et al., 2010),
Mediterranean diet (Scarmeas et al. 2006; Tsivgoulis et al., 2013), and meditative
practices to reduce stress (Innes and Selfe, 2014) are negatively correlated with AD
risk. Of interest, neuroinflammation has been increasingly regarded as an essential
component of AD pathogenesis and many AD risk factors impact these inflammatory
pathways. This next section will discuss the link between AD pathology and
neuroinflammation.

1.1.1 Alzheimer’s disease pathology
Alzheimer’s disease was first characterized by Alois Alzheimer in 1906 as an
“unusual disease of the cerebral cortex…causing memory loss, disorientation,
hallucinations, and mortality” that was first described in a woman in her fifties,
Auguste D. Moreover, the cerebral cortex was found to be “thinner than normal and
had presentations of senile plaques…along with neurofibrillary tangles,” shown
through immunohistological staining (Amaducci, 1996). Today, the excessive
deposition of amyloid beta (Aβ) plaques are an early hallmark of EOAD and LOAD
pathogenesis that leads to the development of intraneuronal hyperphosphorylated tau
forming neurofibrillary tangles (NFT), gliosis, and neuronal loss (Hardy and Allsop,
6

1991). The amyloid cascade hypothesis is amongst the most prevalent and posits that
Aβ accumulation is the central event and catalyst of AD. First proposed in 1992 by
John Hardy and Gerald Higgins, the amyloid cascade hypothesis states that the
accumulation of Aβ is the primary cause of AD pathogenesis and acts as a trigger for
neuroinflammation, neuronal injury, the formation of neurofibrillary tangles (NFTs),
and neuronal death (Hardy and Higgins, 1992). These events lead to the clinical
cognitive symptoms of memory loss and increased dependence as the disease worsens
(Stern et al., 1994).

Genetics of Alzheimer’s disease
The amyloid cascade hypothesis was discovered by studying EOAD and down syndrome
patients (Wisniewski et al., 1985). Down syndrome patients have a third copy on
chromosome 21, where APP gene is located. The overexpression and production of APP
leads to high prevalence of AD in down syndrome patients, wutg approximately 60-75%
of them over the age of fifty diagnosed with AD (Holland et al., 1998; Bakkar et al., 2010;
Lott, 2012). The excess production of APP and the discovery of dense core plaque
pathology in the brains of Down syndrome patients supported the hypothesis that Aβ is a
critical driving factor in AD pathogenesis. The hypothesis has since been revised to account
for amyloidosis that occurs in LOAD patients (Selkoe and Hardy, 2016). Apolipoprotein
E (ApoE) and ABCA7 protein, lipid transporters of Aβ, are some of the genes that have
been identified as a genetic locus for LOAD and that loss-of-function mutations increase
the risk of AD by threefold (Steinberg et al., 2015). Specifically, APOE4 and TREM2,
have been found to impair Aβ clearance from immune cells present in the brain that relate
7

to neuroinflammation. Taken together, the amyloid cascade hypothesis was reformed to
consider the EOAD inherited mutations driving Aβ production and the imbalance of Aβ
clearance mechanisms that lead to excess Aβ accumulation (Kline, 2012).

1.2 Innate immunity
The central nervous system (CNS) was once considered to be immune-privileged: an
organ system protected from peripheral immune interaction and inflammation
(Engelhardt, 2010). However, this notion has been refuted and re-evaluated,
supporting an expansive investigation on how cerebral immunity influences
homeostatic processes such as brain development (Bilbo and Schwarz, 2009), clearance
of cellular debris (Lucin and Wyss-Coray, 2009), and synaptic pruning (Bialas and
Stevens, 2013; Veerhuis et al., 2011). Current literature suggests that Aβ in the CNS
is tightly controlled, as the processing, secretion, and degradation of Aβ and its
removal from the CNS are all highly regulated processes. However, excessive
deposition of Aβ plaques is an early hallmark of AD pathogenesis (Heneka et al.,
2015).

1.2.1 Soluble mediators of innate immunity
Under normal conditions, microglia provide support for synaptic pruning,
processing/clearance of harmful pathogens, and presenting antigens to T cells to
produce antibodies from future encounters of the same pathogen (Aloisi et al., 2000).
Brain-resident macrophages, microglia, have been implicated to drive the removal of
8

Aβ from the CNS (Solito and Sastre, 2012). However increasing evidence indicates
that microglial activation can act as an independent factor at very early stages of AD,
in which cytokines and immune-related genes are the key participants in AD
pathogenesis (Zheng et al., 2016). Thus, it is imperative to understand the underlying
mechanisms of action that cytokines have on microglial function for Aβ clearance.
Cytokines are a heterogenous group of proteins with molecular weights ranging from
8 to 40 kDa (Feghali and Wright, 1997). These multifunctional molecules can be
synthesized by nearly all nucleated cells and generally act locally in a paracrine or
autocrine manner (Bornstein et al., 2004). Many of them are referred to as interleukins
(ILs), indicating that they are secreted by and act on leukocytes. Importantly, cytokines
such as tumor necrosis factors (TNFs), interferons (INFs), and transforming growth
factors (TGFs) have all been implicated in AD as either drivers or deterrents in AD
(Zheng et al., 2016). For example, once microglia encounter, engulf, and process the
pathogen, they can secrete TNF-α, IL-23, and IL-1β cytokines as a proinflammatory
signal to activate phagocytic programming or secrete C-C motif chemokine ligand 2
(CCL2) and granulocyte macrophage-colony stimulating factor (GM-CSF)
chemokines to attract T cells to the sites of pathogenic infections (Duvallet et al.,
2011). However, in AD, sustained activation of the inflammatory response leads to
poor microglial Aβ clearance (Vom Berg et al., 2012). Interestingly, deletion of IL-10
and TGF-β have been shown to rebalance immune homeostasis and reduce AD-like
pathology possibly via improving microglial phagocytosis (Guillot-Sestier et al., 2015;
Town et al., 2008). Since Aβ-induced synaptic damage, neuronal loss, and microglial
activation can explain many of the clinical hallmarks of AD, including declined
9

cognition and increased memory impairment, it is important to study the innate
immune mechanisms of microglial phagocytosis.

1.2.2 Innate immune mediators of phagocytosis
Several studies have found that microglia-mediated phagocytosis of Aβ and apoptotic
cellular debris is mediated by the following: cell surface receptors such as scavenger
receptors (e.g. CD36), (Khoury et al., 1998), triggering receptor expressed on myeloid
cells 2 (TREM2) (Yeh et al., 2017), and protein complexes such as complement
component 1q (C1Q) (Webster et al., 2000). CD36, a member of the class B scavenger
receptor family, transports fatty acids, collagen, and oxidized low-density
lipoproteins into cells (Abumrad et al., 2005). Aβ binds to the CD36 receptor on
microglia as a pathogen associated molecular pattern (PAMP) and impairs the
recycling of CD36, leading to reduced phagocytic efficiency of microglia on Aβ
(Coraci et al., 2002; Moore et al., 2002). TREM2 is expressed on osteoclasts, dendritic
cells, macrophages and microglia (Neumann and Takahashi, 2007). Mutation of
TREM2 results in defective phagocytosis and development of progressive dementias
such as Nasu- Hakola disease, frontotemporal dementia, and AD (Thrash et al., 2009)
Specifically, TREM2 is an important sensor of anionic lipids, which are accumulated
during Aβ deposition and neuronal loss; thereby, the mutation of TREM2 dampens
microglial detection of damage-associated lipids (Wang et al., 2015) . Complement
component 1q (C1q) is a key recognition protein of the C1 macromolecular complex,
which initiates the classical complement pathway for antigen presentation, danger
signaling, and induction of immunomodulatory cytokines for phagocytosis, a process
10

called opsonization (Stevens et al., 2007). Recent evidence suggests that
overexpression of C1q increases synaptic targeting and promotes excessive
elimination and synapse loss by microglia (Stephan et al., 2012). Our lab has shown
that C1q can bind to Aβ in vitro, suggesting that antigen processing and subsequent
presentation of Aβ are involved in adaptive immune T cell-mediated responses in AD
(Leung et al., 2020)

1.2.3 Innate immune processing and presentation of antigens
Antigen processing by macrophages and dendritic cells (DCs) is an immunological
process that prepares necrotic/apoptotic cells, bacteria, and viruses for presentation to
adaptive immune T cells (Blum et al., 2013) These “antigen presenting cells” (APCs)
produce antigens from phagocytosed pathogens by lysosomal (Nakagawa and
Rudensky, 1999) and proteasomal (Yang et al., 1992) degradation and present them
to adaptive immune T cells through their cell surface membrane, allowing for the
adaptive immune system to produce antibodies or perform cell-mediated responses to
destroy infectious agents or cells. (Guermonprez et al., 2002). In antibody-mediated
immune T cell responses, this process allows the immune system to mount a rapid,
efficient response to the same pathogen if encountered again, also known as
immunological memory (Janeway et al. 2001). In the context of AD, the ability of
microglial processing and presentation of Aβ to T cells appear to be impaired (Ferretti
et al. 2016).

11

Human leukocyte antigens (HLA) have been implicated in AD risk. For example,
human leukocyte antigen-DR (HLA-DR), the human locus of MHC-II in vertebrates,
is upregulated in microglia and senile/neuritic plaques in AD patients (Rogers et al.
1988). Recent case studies in AD Tunisian patients found that HLA-DQ
polymorphisms are linked to increased susceptible risk in AD (Mansouri et al., 2015).
In animal studies, two regulators of antigen processing and presentation on APCs
similar to humans, major histocompatibility complex I (MHC-I) and II (MHC-II), have
been linked to increased risk for AD pathogenesis (Kauwe et al., 2016; Tooyama et
al., 1990). For example, MHC-II immunolabel of microglia have been found within
microglia in the substantia nigra of patients with AD (Itagaki and Mcgreer, 1988).
Moreover, MHC-II expressing microglia display an “immature” phenotype in antigen
processing and presentation with the accumulation of MHC-II in their intracellular
compartments rather than the cell surface membranes near dense core plaques (Ferretti
et al., 2016). Future studies will need to clarify whether impairment in microglial
processing and presentation is a defect in lysosomal processing, proteasomal
processing, or transporter associated with antigen processing (TAP).

1.3 Adaptive immunity
The establishment and maintenance of immune responses, homeostasis, and
immunological memory depends on T cells (Kumar et al., 2018). T lymphocytes
originate from bone marrow progenitors that migrate to the thymus for maturation,
selection, and are subsequently found in peripheral draining lymph nodes as
surveillance sites for APCs and lymphocytes to meet (Ding et al.,2012). T cells
12

populate virtually every organ and tissue within the body, including primary and
secondary lymphoid tissue (Tsuji et al., 2008), mucosal and barrier sites (Mayer, 2003),
exocrine organs (Saitoh et al., 1990), fat (Nishimura et al., 2009), and the CNS
(Moalem et al., 1999; Tzartos et al., 2008; Ellwardt et al., 2016). Despite early
detection of T cells found near senile plaques in AD patients (Rogers et al. 1988), T
cell involvement in AD pathogenesis is understudied and scrutinized for the lack of or
miniscule populations found within brain parenchyma (Town et al., 2005). However,
recent evidence suggests that T lymphocytes, specifically CD4
+
T helper cells,
influence microglial clearance of Aβ and have a unique pathway to enter from the
periphery to the CNS (Togo et al., 2002; Gemechu and Bentivoglio, 2012; Browne et
al., 2013; Baruch et al., 2015).

1.3.1 Innate-adaptive immune crosstalk
Under physiological conditions, there are few CD4
+
T cells in the brain, although
they can cross the blood brain barrier (BBB) and the choroid plexus brain region
(Britschgi and Wyss-Coray, 2007; Schecter et al., 2013). Once infiltrated, CD4
+
and
CD8
+
T cells are presented with antigens from microglial cells (Jensen, 2007). CD4
+

T cells have been found to be increased in AD patients, especially in the hippocampus
and temporal cortex, leading to activated microglial/APC expression of MHC-II,
further licensing CD4
+
T cells to migrate into the CNS (Sardi et al., 2011). Moreover,
AD patients display strong Aβ-reactive CD4
+
T cell responses compared to cognitively
normal-aged subjects (Monsonego and Weiner, 2003). Interestingly, a study
introduced an Aβ- specific immunotherapy that successfully reduced amyloid burden
13

in AD patients. However, this study was discontinued after there were severe
inflammatory microglial and autoreactive T cell responses in the brains of vaccinated
AD patients (Holmes et al., 2008). Thus, in order to clear toxic Aβ, it is imperative
that innate and adaptive immune responses communicate synergistically to trigger a
balanced immune response to attenuate AD pathology (Guillot-Sestier et al., 2015;
Fiala, 2010).

1.3.2 Transendothelial migration of T cells into the CNS
Transendothelial T cell migration through the BBB is a multi-step process
characterized by a series of sequential and tightly controlled steps (Takeshita and
Ransohoff, 2012). The steps are: (i) rolling: weak adhesion of T cells to endothelial
cells mainly through interactions between selectins and their carbohydrate counter
receptors (McEver, 2015); (ii) activation: T cell activation through chemokine
stimulation and costimulatory molecules, resulting in functional activation of
adhesion molecules along their surface (Oppenheimer-Marks et al., 1998);
(iii) arrest: T cell attachment to endothelial cells through interactions between integrins
associated with T cell and cell adhesion molecules (CAMs) on endothelial cells (Muller
et al., 1993); (iv) crawling: T cell seeking preferred sites of transmigration across the
endothelium (Gerard et al., 2009; Heasman and Ridley, 2010); (v) transmigration:
migration of T cells across CNS endothelia into perivascular space and progression
across glia limitans into the brain parenchyma, a process driven in part by chemokine-
chemokine receptor interactions (Mö hle et al., 1998). Taken together, the interaction
of selectins and their ligands, integrins and CAMs, and chemokines and chemokine
14

receptors are how adaptive immune components or T cells gain entry into the CNS.

Numerous studies suggest that T cell infiltration from neurovasculature BBB
dysfunction contributes to the onset and progression of AD, proposing a link between
cerebrovascular changes and neurodegeneration (Montagne et al., 2015; Zlokovic,
2011; Farkas and Leuiten, 2001). For example, slectins and CAMs are significantly
higher in 4-month old 5xFAD mice than age-matched wild-type controls (Zenaro et
al., 2015), suggesting that BBB integrity is compromised. Interestingly, the expression
of these adhesion molecules was observed in areas burdened by Aβ plaques and
enriched with migrated T cells and other leukocytes in AD patient brains (Frohman et
al., 1991: Zenaro et al., 2015). Moreover, peripheral T cells derived from AD patients
overexpress C-X-C motif chemokine receptor 2 (CXCR2) contributing to one pathway
for transendothelial migration (Liu et al., 2010). Additionally, CXCL8, a ligand for
CXCR2, have been shown to increase expression in senile plaques of AD brains (Xia
et al., 1997).

1.3.3 T cell influences in Alzheimer’s disease
T cells have been shown to be co-localized with Aβ plaques in gray matter areas of
the cortex near microglia in AD patients (Rogers et al., 1988). Early reports on the
complete ablation of adaptive immune T and B cells via genetically crossed
immunocompromised mice (RAG2) and a mouse model of cerebral amyloidosis
(5XFAD) resulted in increased amyloid burden in the CNS and in downregulation of
microglial genes associated with Aβ clearance (Marsh et al., 2016). Furthermore, APP
15

expression and processing were unaffected in these transgenic mice, suggesting that
adaptive immune T cells may influence the clearance of Aβ from microglia. This is
further supported by our lab findings as genetic ablation of TGF-β in CD4
+
helper T
cells in a mouse model of cerebral amyloidosis improved AD-like pathology but at a
net negative cost of early death. (Im et al., 2020 in preparation). Taken together, these
studies suggest an involvement of adaptive immune T cells in AD pathogenesis,
however, it is unclear as to how specific types of T cells and their effector cytokines
influences clearance of amyloid from microglia and other phagocytes. This will be
expanded in the next section.

1.3.4 CD4
+
T helper cell involvement in AD
CD4
+
T cells are critical components of the immune system and key regulators of the
inflammatory processes to limit infection. CD4
+
T cell subtypes: T helper 1(Th1), T
helper 2 (Th2), and T regulatory (Treg), and T helper 17 (Th17), have all been implicated
in AD pathology (Arlehamn et al., 2019). Th1 cells secrete interferon-gamma (IFN-γ)
upon presentation of Aβ from microglia and have been implicated in exacerbating
prognosis in AD patients (Lambracht-Washington et al., 2011). While it is unclear
how IFN-γ secreted from Th1 cells decreased microglial clearance of Aβ, IFN-γ
secretion has been shown to upregulate glial MHC II expression (Butovsky et al.,
2003). Hence, feedback mechanisms between microglia-Th1 interactions over time
may diminish phagocytic clearance of Aβ. Indeed, adoptive transfer of antigen
specific, Aβ Th1 T cells, and subsequent anti-IFN-γ neutralization in a mouse model of
cerebral amyloidosis led to lower Aβ burden in the hippocampus and improved
16

cognition (Brown et al., 2013). Th2 cells secrete interleukin-4 (IL-4) upon presentation
of Aβ from microglia. While there is limited evidence of Th2 presence and/or IL-4
expression in AD patients, IL-4 can downregulate TNF-α and upregulate MHCII and
CD36 in microglia, potentially increasing Aβ phagocytic function in microglia
(Butovsky et al., 2005; Baruch et al., 2013). Thus, immunotherapeutic vaccine trials
have focused on boosting Th2-like immune responses in order to ameliorate disease in
AD. For example, enhanced Th2 T cell immune responses against amyloid protein by
DNA-prime-adenovirus boost regimen have been implicated in improving AD-like
pathology in mouse models of cerebral amyloidosis (Kim et al., 2007; Ghochikyan et
al., 2006). Studies have observed elevated CD4
+
CD25
+
Treg cells in AD patients
compared to healthy control (Pellicano et al., 2009). Treg influence in AD remains
unclear, however AD patients’ PBMCs and CSF samples have demonstrated a positive
correlation of total Treg numbers with tauopathy (Oberstein et al., 2018). Furthermore,
transient depletion of Foxp3
+
Treg cells in a mouse model of cerebral amyloidosis
improved cognitive function and clearance of Aβ through the increased licensing of
peripheral immune cells into the CNS via the choroid plexus (Baruch et al., 2015).
Genetic depletion of either TGF-β and or IL-10 improves in Aβ clearance; however,
it is unknown if Treg cells provide these cytokines during these events (Town et al.,
2008; Guillot-Sestier et al., 2015). It is well established of the relationship of the
development of Tregs vs. Th17 cells based on the differential dependence of TGFβ. Our
lab has shown that depletion of TGFβ signaling in CD4
+
cells ameliorates AD-like
pathology, yet it is unclear of the magnitude each of these cell types influence
amyloidosis in AD. Additionally, Th17 cells found in AD patient CSF and PBMC
17

samples correlate with amyloidopathy (Oberstein et al., 2018). Little is known about
the role Th17 cells in AD; However, multiple studies in autoimmunity have suggested
their role in pathogenicity to draw hypotheses.

1.3.5 T helper 17 T cells
Recent evidence suggests that total Th17 cell numbers are positively correlated with
cerebral amyloidopathy in AD patients (Oberstein et al., 2018). Th17 cells are a T cell
population that secretes effector cytokines interleukin-17A (IL-17A), interleukin-21
(IL-21), interleukin-22 (IL-22), and granulocyte macrophage-colony stimulating factor
(GM-CSF), classically distinct from Th1, Th2, and Treg cells (Zuniga et al., 2013).
Physiologically, IL-17A secreted from Th17 cells stimulates the mobilization and de
novo generation of neutrophils by GM-CSF, thereby bridging innate and adaptive
immune processes (McCarthay et al., 2014). Il-17A facilitates neutrophil recruitment
as a host defense mechanism against extracellular bacteria such as Klebsiella
pneumonaie (Chen et al., 2011; Muranski et al., 2014) or Bacteroides fragilis
(Omenetti et al., 2015; Troy et al., 2010), and Candida albicans (Conti et al., 2010;
Shao et al., 2019; Schimke et al., 2010). IL-21 signaling promotes and sustains Th17
differentiation in an autocrine manner and induces expression of RAR-mediated organ
receptor gamma t (RORγt) and signal transducer and activator of transcription 3
(STAT3), both of which are necessary transcription factors in the development and
lineage commitment of Th17 cells (Wei et al., 2007). IL-22, an IL-10 family member,
coexpresses with IL-17A from Th17 cells and tightly regulates neutrophil recruitment
to infection sites (Liang et al., 2006). Finally, GM- CSF is essential for the ability of
18

Th17 cells to drive inflammation in the CNS as GM-CSF expression promotes cell
survival, proliferation, differentiation, mobilization, and activation in macrophages
(McGeachy 2011). Taken together, Th17-mediated immunity provides critical support
for scenarios of acute bacterial and fungal infections. However, chronic inflammatory
conditions in autoimmunity have shown that systemic activation of Th17 leads to
exacerbation of pathology and gliosis.

1.3.6 Th17 hypothesis in AD
In the context of disease, prolonged Th17 cell involvement can exacerbate pathology
as shown in experimental autoimmune encephalomyelitis (EAE) (McGinley et al.,
2018; Butcher et al., 2018; Aranami et al., 2008) rheumatoid arthritis (RA) (Kugyelka
et al., 2016; Boniface et al., 2013;, Azizi et al. 2013), Bordetella pertussis (BP)
(Warfel et al., 2013; Gates et al., 2017, and inflammatory bowel disease (IBD)
(Kempski et al., 2017; Hou, 2020). Studies suggest that Th17-mediated secretion of IL-
17A and macrophage secretion of IL-23 are detrimental in the initial phases of these
diseases. Elevated levels of IL-23 promotes Th17 survival while IL-17 increases
microglial activation. This progression leads to dysregulated macrophage maturation,
demyelination, and sepsis (Bunte et al., 2019). Specifically, IL-23 has been shown to
stimulate the expression of GM-CSF in Th17 cells. In contrast, the depletion of GM-
CSF in Th17 cells resulted in suppression of disease pathology in virtually all animal
models of inflammation and autoimmunity that have been tested (Hamilton, 2015).
Taken together, GM-CSF is a crucial mediator of Th17 cell encephalitogenicity and has
been shown to be a promising therapeutic intervention for a multitude of diseases
19

(Maddur, 2012). Interestingly, Il-23 receptor (IL-23R) polymorphisms and elevated
IL-23 expression are associated with increased AD risk (Liu et al., 2014; Shahrokhi et
al., 2018, Sarasella et al., 2014). Moreover, increased IL-17A expression in CSF
(Alhaidary et al., 2014) and increased Th17 related transcription factor RORγt
(Saresella et al., 2011) are associated with an increased risk for AD. Taken together,
systemic activation of Th17 cells with microglial IL-23 expression may diminish
microglial capabilities to perform Aβ clearance (Figure 2).

1.4 Conclusion
While there have been studies to dissect the roles of Th1, Th2, and Treg cells
involvement in AD pathology, Th17 involvement remains unknown. This section will
test the hypothesis that Th17 cells exacerbate cerebral amyloidosis and diminish
microglial clearance of Aβ. We found that depletion of Th17 cells in our APP/PS1
mouse model of cerebral amyloidosis rescued AD-like pathology, restored cognitive
functions, and increased phagocytic uptake of Aβ within microglial phagolysosomes in
12-month-old animals. Moreover, adoptive transfer of Aβ-specific Th17 (Aβ-Th17)
cells in APP/PS1 mice at 11 months of age exacerbated AD-like pathology and
decreased phagocytosis. Strikingly, weekly administration with αIL-17 neutralizing
antibody reversed pathological hallmarks from adoptively transferred Aβ-Th17 cells,
suggesting that Th17 cells play a major role in the clearance of Aβ and interaction with
microglia. These assays provide insight into potential therapeutic interventions to slow
or effectively inhibit AD pathology.

20

Figure 1. (adapted from Jack et al. 2010) Dynamic biomarkers of the Alzheimer’s disease pathological cascade. Aβ is
identified by CSF Aβ42 or positron emission tomography (PET) amyloid imagining. Tau-mediated neuronal injury and
dysfunction is identified by CSF tau or fluorodeoxyglucose-PET. Brain structure is measured by use of structural MRI. Aβ=β-
amyloid. MCI=mild cognitive impairment.

21

Figure 2. Schematic diagram of hypothesis: Th17 T cells exacerbate microglial amyloid beta clearance in Alzheimer’s
disease through the IL-17/IL-23 pathway. Under normal conditions (Left), innate immune ramified microglia survey the central
nervous system (CNS) to identify pathogens and self-antigens while the blood brain-barrier (BBB) restricts peripheral adaptive
immune T helper 17 (Th17) cells from entry. In Alzheimer’s disease (AD) (right), accumulation of amyloid beta (Aβ) forming
plaques are phagocytosed by activated microglia and secrete interleukin-23 (IL-23) in the immune microenvironment.
Simultaneously, Th17 cells infiltrate the CNS, and secrete interleukin-17A (IL-17A) after binding to activated microglia. IL-
17/IL-23 interaction has been shown to be highly pathogenic in AD pathogenesis and been identified as a driver of the disease.
22

Th17 cells exacerbate Alzheimer’s disease-like pathology
Kwok W. Im,
1
†
,2,3
Mariana F. Uchoa,
1
Marie V. Guillot-Sestier,
1,2
† Orla M. Finucane,
2

Brian P. Leung,
1,4
† Dejie Zhen,
1
Cole J. Miller,
1
Alex Vesling,
1
Javier Rodriguez,
1

Kingston H. Mills,
3
Marina A. Lynch,
2
and Terrence Town
1*

1
Zilkha Neurogenetic Institute, Department of Physiology and Biophysics,
Keck School of Medicine of the University of Southern California, 1501 San
Pablo Street. Los Angeles, CA 90089-2821, USA
2
Trinity Biomedical Sciences Institute, Department of Physiology, Trinity
College Institute of Neuroscience, Trinity College Dublin, College Green
Dublin 2, Ireland
3
School of Biochemistry and Immunology, Trinity Biomedical Sciences
Institute, Trinity College Dublin, Dublin 2, Ireland
4
Hong Kong University of Science and Technology, Clear Water Bay, Kowloon,
Hong Kong, China
*Corresponding author: ttown@usc.edu
†Current address
One Sentence Summary: T helper 17 cells exacerbate Alzheimer’s disease-like
pathology, and treatment with IL-17 neutralizing antibody reverses pathological
clearance of Alzheimer’s β-amyloid.

23

1.5 Abstract
T helper 17 (Th17) cells are present in Alzheimer’s disease (AD) and in mouse
models of the disease. However, the impact of Th17 cells on neuroinflammation and
AD pathogenesis is unknown. To address this, we crossed the APP/PS1 mouse model
of cerebral amyloidosis with a cre-lox recombinant mouse deficient in Th17 cells
(APP/PS1
+
CD4
cre
-Stat3
f/f
). Th17 depletion in APP/PS1 mice decreased cerebral
amyloid burden and increased Aβ uptake and clearance by mononuclear phagocytes
by quantitative in silico 3D modelling. Further, adoptive transfer of Aβ-specific Th17
cells into APP/PS1 mice decreased Aβ phagocytosis and exacerbated cerebral
amyloidosis. Finally, administration of anti-IL-17 neutralizing antibody prior to
adoptive transfer reversed these phenotypes. Our results suggest that Th17 cells
negatively instruct mononuclear phagocytes, precluding Aβ clearance. Harnessing
Th17-mediated innate immunity may be therapeutically relevant for AD.

1.6 Results
1.6.1 CD4
+
T cells are elevated in APP/PS1 mouse brains
To assess the abundance of Th17 cells present in the brain of cerebral amyloidosis, we
used the APPswePSEN1ΔE9 (APP/PS1) (Jankowsky et al., 2004) and performed flow
cytometry on whole brain homogenates in 12-month- old APP/PS1 mice and WT
littermate controls (Fig. 1A). We found that there were CD4
+
T cells present in both
WT and APP/PS1 mice but a significant increase in total number of cells in APP/PS1
mice (Fig. 1B). Intracellular staining revealed that a proportion of CD4
+
cells stained
positive for IL-17 was significantly increased in APP/PS1 mice compared to WT in
24

both absolute numbers and percentages of cells sorted from CD4
+
T cells (Fig. 1, C and
D). No changes were found at 6 months of age in either WT or APP/PS1 mice (data
not shown).

1.6.2 CD4
cre
Stat3
fl/fl
depletion of Th17 cells
Because IL-17 producing Th17 cells have been implicated in the pathogenesis of
chronic inflammatory and autoimmune diseases (Butcher et al., 2018; Aranami et al.,
2008), we hypothesized that the Th17 cells drive cerebral amyloidosis and chronic
inflammation. To test this, we crossed the APP/PS1 mouse model of cerebral
amyloidosis with a Cre-lox mouse line deficient in STAT3 signaling on CD4 T cells
(CD4
cre
STAT3
fl/fl
). We chose to focus our study on depleting STAT3 signaling in CD4
T cells as an established mouse model known to inhibit Th17 T cell development
(Egwuagu, 2009). To identify T cell subset manipulations from these transgenic mice,
we performed flow cytometry and ELISA on draining cervical lymph nodes and spleen
(not shown) from 3-6-month-old CD4
cre
STAT3
fl/fl
mice to WT littermate controls (Fig.
2A). Cells were homogenized and stimulated with PMA/ionomycin and Brefeldin A
for 4h at 37°C prior to ELISA and FACS. Representative FACS plots from
intracellularly stained IFN-γ and IL-10 on sorted CD4
+
cells revealed no changes in
CD4
cre
STAT3
fl/fl
mice or WT littermate controls (Fig 2A, bottom and middle plots).
However, CD4
cre
STAT3
fl/fl
mice displayed significant reduction in IL-17-producing
CD4
+
cells (Fig. 2A, top right) compared to littermate controls (Fig. 2A, top left).
These results were further supported by ELISA collected from the supernatant (Fig.
2B) and the reduction in RoRγt
+
CD4
+
cells in CD4
cre
STAT3
fl/fl
mice compared to WT
25

in FACS (Sfig. 1, A and B).

1.6.3 Decreased amyloid burden in APP/PS1
+
CD4
cre
Stat3
fl/fl
brains
Accumulation of Aβ is the central event that acts as a trigger for neuroinflammation
(Hardy and Higgins, 1992). At 12 months of age, APP/PS1
+
CD4
cre
stat3
fl/fl
mice
manifested significantly reduced Aβ deposition in hippocampal regions as measured
by thioflavinS (ThioS) histochemistry versus APP/PS1 mice (Fig. 3A). Quantitatively,
percentages of ThioS staining and total number of plaques observed were also
significantly reduced in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice (Fig 3., B and C). Cingulate
cortex and entorhinal cortex staining for ThioS also displayed these changes (data not
shown). Water-soluble forms of Aβ, such as toxic Aβ oligomers, have been shown to
significantly correlate with neuritic plaques (McLean et al., 1999). We assessed Aβ1-
40 and Aβ1-42 soluble and insoluble fractions and observed a two-fold reduction in triton-
soluble (Fig. 3D) and guanidine- fractionated insoluble (Fig. 3E) from
APP/PS1
+
CD4
cre
STAT3
fl/fl
mice compared to APP/PS1 mice.

1.6.4 Cerebral innate immune cells phagocytose Aβ in absence of Th17 cells
We hypothesized that excessive Aβ accumulation drives neuroinflammatory
conditions that diminish microglial Aβ clearance as seen in previous literature (Vom
Berg et al., 2012; Guillo-Sestier et al., 2015). To assess the Aβ phagocytic capabilities
of Th17 sufficient versus Th17 deficient mice in APP/PS1 transgene, we utilized
quantitative in silico 3D modeling (Fig. 4A (Guillot-Sestier et al., 2016). 12-month-
26

old brains of APP/PS1
+
CD4
cre
STAT3
fl/fl
mice displayed increased volumetric uptake
of Aβ-labeled 6E10
+
immunostaining within CD68
+
phagolysosomes of Iba1
+

macrophages compared to APP/PS1
+
STAT3
fl/fl
mice (Fig. 4B) and increased
percentage occupied of 6E10
+
within CD68
+
phagolysosomes near plaques (Fig. 4C).
Increased Iba1
+
immunolabeling for macrophages were significantly increased in
APP/PS1
+
CD4
cre
STAT3
fl/fl
compared to APP/PS1
+
STAT3
fl/fl
mice (Sfig. 2, A and B).
However, GFAP
+
immunolabeled cells were decreased in APP/PS1
+
CD4
cre
STAT3
fl/fl

mice compared to APP/PS1
+
STAT3
fl/fl
mice (Sfig. 3, A and B). To rule out the
possibility of an effect on cerebral amyloidosis due to altered APP/Swe or PS1Δe9
transgene expression, western blot and quantitative real-time reverse transcriptase
PCR (qPCR) were performed on protein and RNA extracted from hippocampal cortex
of APP/PS1
+
STAT3
fl/fl
and APP/PS1
+
CD4
cre
STAT3
fl/fl
mice. No differences were
found between groups on PS1 protein on mRNA levels (Sfig. 4. A, B, and C).

1.6.5 Generation of Aβ-specific Th17 cells and mouse model
Aβ-specific Th1 cells producing IFN-γ have been shown to increase AD-like
pathology and can be rescued by neutralizing IFN-γ antibodies in 6-month-old
APP/PS1 mice (Browne et al., 2013). Furthermore, presentation of IL-17 producing T
cells elevated in an age and disease-severity dependent manner (McManus et al., 2014).
To generate antigen-specific, Aβ, Th17 (Aβ-Th17) cells, we immunized APP/PS1 mice
with aggregated Aβ1-42 and complete Freund’s adjuvant (CFA) emulsion via i.p. for 8-
10 days prior to extracting cervical lymph nodes and splenocytes to in vitro to expand
Aβ-Th17 cells. Lymphocyte and splenocyte co-cultures were treated with polarizing
27

cytokines of IL-1β and IL-6 and recall stimulated with 20μg/ml of Aβ1-42 for 3 days.
FACS plots display sorting of CD4
+
cells and intracellular cytokine staining for IL-17
in non-specific antigen recall, MOG35-55, versus Aβ1-42 recall stimulation (Fig. 5A).
Aβ1-42 recall stimulated Th17 cells increased 3-fold in total numbers compared to
MOG35- 55 recall stimulation (Fig. 5B). ELISA collected from supernatant found
comparable IL-17 expression of Aβ1-42 recall stimulated and positive control
PMA/ionomycin treatment while no changes were observed in T cell- mediated
cytokine production of IFN-γ or IL-10 (Fig. 5C). After one round of Aβ1-42 stimulation,
surviving Th17 cells gated on CD4
+
IL-17
+
antibodies were washed and injected i.v.
(15x10
6
cells/mouse) into 11-month-old APP/PS1 mice.

1.6.6 Neutralization of IL-17 attenuates the effects of Aβ-Th17 cells on plaque pathology
Having shown genetic manipulation of Th17 cells ameliorates AD-like pathology and
increased microglial clearance of Aβ, we assessed the role of key Th17 cytokine, IL-
17, by treating APP/PS1 mice with a neutralizing antibody prior to, and following, Aβ-
Th17 cell transfer (see methods section for details). At 12 months of age, APP/PS1
mice adoptively transferred with Aβ-Th17 cells observed 2-fold increase in ThioS
+

immunostaining compared to APP/PS1 mice (Fig. 6A). Interestingly, neutralizing IL-
17 in APP/PS1 mice reversed ThioS
+
plaque pathology in percentages and total plaque
numbers in APP/PS1 mice adoptively transferred with Aβ-Th17 cells (Fig. 6, B and
C). These data were supported with two-fold increases in triton-soluble (Fig. 6D) and
insoluble (Fig. 6E) of Aβ1-40 and Aβ1-42 fractions in APP/PS1 mice adoptively
transferred with Aβ-Th17 compared to APP/PS1 mice and a reversal in attenuation with
28

neutralizing IL-17 antibodies.

1.6.7 Neutralization of IL-17 rescues the effect of Aβ-Th17 cells on phagocytosis
Our data suggest that anti-IL-17 attenuated the effects of Aβ-Th17 mediated effects
on plaque deposition and concentration of Aβ1-40 and Aβ1-42 fractions in APP/PS1
mice. We hypothesized a strong correlation between Th17 mediated influences
affecting microglial clearance of Aβ. Quantitative in silico 3D modeling revealed
diminished Aβ-labeled 6E10
+
immunostaining in CD68
+
phagolysosomes in iba1
+

microglia in APP/PS1 mice adoptively transferred with Aβ-Th17 cells compared to
APP/PS1 mice (Fig. 7A). However, administration of anti- IL-17 in APP/PS1 mice
rescues phagocytic uptake and percentage of phagolysosome occupied inhibition from
Aβ-Th17 mice (Fig. 7, B and C). Our lab previously demonstrated that IL-10 treatment
shifts microglial activation away from Aβ phagocytosis via increasing activation of
STAT3 signaling (Guillot- Sestier et al., 2015). STAT3 signaling is a critical regulator
for Th17 development and activation of IL-17. To directly investigate the effects of
Th17 modulation on microglial phagocytosis, adult primary microglial cultures were
established from CD4
cre
STAT3
fl/fl
mice and STAT3
fl/fl
mice, and Aβ1-42 phagocytosis
was evaluated. Both groups expressed IL-23 mRNA expression in microglial cultures
alone after exposure to LPS or Aβ1-42 (Sfig. 5). Supernatant collected for ELISA
displayed complete inhibition of IL-17 expression in all treatments with neutralizing
IL-17 ab (Sfig. 6A). Pre-incubation of Aβ-Th17 cells and Aβ1-42 phenocopied
significant reductions in Aβ1-42-labeled Aβcy555 uptake in CD68
+
phagolysosomes of
29

CD11b
+
microglia compared to PBS-pre-incubation control as in our in vivo studies
(Fig. 7, D and E). However, these effects were reversed when treated with anti- IL-17
preincubation. Supernatant collected from these co-cultures revealed significant
increases in TNF-α expression in pre-incubated Aβ-Th17 cells (Sfig. 6B). No
significance was observed in other inflammatory markers (Sfig. 6, C, D, E, and F).

1.7 Discussion
The IL-23/IL-17 axis impacts the survival and differentiation of highly pathogenic
Th17 cells in a multitude of chronic inflammatory diseases. Thus, blocking the IL-
23/IL-17 axis has been shown to suppress the development of diseases such as psoriatic
arthritis, inflammatory bowel disease, and multiple sclerosis (Furue et al., 2018; Singh
et al., 2015; Kong et al., 2016). This has prompted an impetus to uncover a deeper
understanding of innate- adaptive immune crosstalk. Recent evidence has implicated
an imbalance of circulating Th17 cells in AD patients and are highly correlated with
cerebral amyloidopathy (Oberstein et al., 2018). Moreover, Polymorphisms of IL- 23R
(Liu et al., 2014) and increased serum levels of IL-23 and IL-17 (Chen et al., 2014) in
AD patients have been found to be associated with high risk for late onset AD. While
the role of brain innate immune microglia in AD has been extensively studied, the
influence of IL-17
+
-producing Th17 cell populations in this disease has been largely
overlooked. Here we investigate the role of IL-23/IL-17 axis-elicited responses in AD
mediated between microglia and Th17 cells respectively. Results showed increased IL-
17
+
-producing CD4
+
T cells in 12-month-old APP/PS1 mouse brains similar to
30

observations made at 6 months of age (Browne et al., 2013) and 12 months of age in
3xTg-AD mice compared to WT control littermates (St-Amour et al., 2019).
Moreover, Cristiano and colleagues reported neutralization of IL-17 rescues amyloid-
β-induced neuroinflammation by significantly reducing microglial proliferative
responses and activation (Cristiano et al., 2019). While studies have quantified high
expression of Th17 polarizing cytokines of IL-2, TNF-α, IL-17, and GM-CSF, there
is no definitive evidence to demonstrate Th17 cell involvement in AD. To address this
knowledge gap, we generated APP/PS1 mice deficient for STAT3 signaling in CD4
+

T cells, to inhibit the development of Th17 cells, and assessed AD-like pathology.
Mean fluorescent intensities demonstrated significantly reduced CD4
+
RoRyt
+
T cell
populations with a slight decrease in CD4
+
FoxP3
+
T cells. This was somewhat
unexpected, as we expected that Treg populations would be increased at the same
magnitude as Th17 reduction. However, this could be due to other T cell subsets that
may compensate these findings of slight reduction to Treg populations. Results showed
reduced amyloid pathology in these animals, and remaining plaques were associated
with activated microglia. Interestingly, remaining plaque deposition resembled “moth-
eaten” morphology, as seen from our previous observations, APP transgenic mice after
Aβ1-42 immunizations, and in brains of AD patients (Guillot- Sestier et al., 2015; Bard
et al., 2000; Nicoll et al., 2006; Nicoll et al., 2003; Schenk et al., 1999; Zotova et al.,
2011). We did not observe histological evidence of brain-infiltrating peripheral
mononuclear phagocytes in APP/PS1 mice depleted of Th17 cells. Furthermore, Th17
depletion did not change the abundance of CD45
hi
versus CD45
int
mononuclear
phagocytes in APP/PS1
+
STAT3
fl/fl
versus APP/PS1
+
CD4
cre
STAT3
fl/fl
mice,
31

suggesting that brain-resident microglia are likely the major population responsible
for Aβ clearance. Moreover, it is unclear as to why there was an observation of
increased CD68
+
cells and decreased astroglia in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice. It
is well established that impairment in microglial phagocytic function near plaques
correlates with Aβ plaque accumulation (Krabbe et al., 2013). The results we report
here show that in APP/PS1 mice genetically depleted of Th17 cells increased Aβ
uptake in CD68
+
phagolysosomes. In this regard, Th17 depletion in APP/PS1 mice
seems to restore Aβ microglial phagocytic capabilities. These findings add to the
complexity of Th17-mediated cytokines such as GM-CSF (Kiyota et al., 2018) and IL-
2 (Alves et al., 2017) administration ameliorating Aβ burden. However, neutralizing
antibodies against IL-17 (Cristiano et al., 2019) and TNF-α (Tobinick et al., 2006;
Ekert et al., 2018) have shown promise as a therapeutic target in AD. As a caveat,
Th17 cells and microglia are not the sole mediators of pathogenicity, as neutrophils,
gamma-delta cells, dendritic cells, and even astrocytes have been shown to be involved
in the IL-23/IL-17 axis for autoimmunity and chronic inflammatory diseases.
Regardless, we report that Th17 depletion in APP/PS1 mice is sufficient to ameliorate
AD-like pathology and rescues phagocytic function in microglia. To evaluate the
impact of Th17 on plaque burden in the brain, we adoptively transferred Aβ-specific
Th17 cells into 11-12-month-old mice. Transfer of Aβ-Th17 cells increased Aβ plaque
burden in hippocampal tissue compared to APP/PS1 mice. This suggests that Aβ-Th17
cells play a role in the development of Aβ plaques in the brain. This was supported by
the treatment of mice with a neutralizing anti-IL-17A neutralizing antibody, which
attenuated the effect of Th17 cells on Aβ accumulation. Interestingly, Aβ-Th17 cells
32

were detected within the brain parenchyma via FACS sorting and was effectively
decreased in mice administered with α-IL17. We hypothesized that these infiltrating
Aβ-Th17 cells influence plaque associated microglial phagocytosis. Indeed, APP/PS1
mice administered with Aβ-Th17 cells observed decreased Aβ uptake within CD68
+

phagolysosomes compared to APP/PS1. For the first time, we show that neutralizing
antibody against IL-17A rescues Aβ-Th17 phenotype, suggesting that neutralizing IL-
17A on pathogenic Th17 cells ameliorate AD-like pathology. Studies report anti-IL-
17 and anti-IL-23 antibodies to be potential therapeutic targets to curtail Aβ
accumulation (Katayama, 2019; Vom Berg et al., 2012). However, it is unclear if
Th17-mediated responses drive microglial Aβ phagocytosis. We tested Th17
interaction with microglial phagocytosis in in vitro cocultures. We observed
significant reductions in Aβ uptake in Aβ-Th17 microglial cocultures compared to Aβ
exposure in microglia. These data suggest that Aβ-Th17-mediated cellular responses
influence microglial Aβ phagocytosis. Indeed, cocultures treated with neutralizing
anti-IL-17 antibodies significantly increased Aβ uptake compared to Aβ-Th17
microglial cocultures. TNF-α expression has been shown to promote Th17
differentiation and subsequent pathogenicity in chronic inflammatory diseases (Zheng
et al., 2014). Aβ-Th17 microglial cocultures displayed increased TNF-α expression and
was subsequently reversed when administered neutralizing IL-17 antibodies. We do not
have a definitive explanation for how microglial phagocytic mechanisms are modified
from Th17 cells. However, multiple studies in chronic inflammatory diseases such as
IBD and Crohn’s disease have observed that the magnitude and outcome of tissue
damage is mostly dependent on distinct CD4
+
T cell cytokine profiles (Monteleone et
33

al., 2009). While we did not histologically observe Aβ-Th17-microglial interactions
near plaques in vivo, nevertheless, the magnitude and abundance of IL-17 producing
Th17 cells sufficiently exacerbated AD- like pathology and impaired microglial
phagocytosis. Future studies will need to focus on investigating distinct T cell subsets
and how each influence innate immune signaling in AD pathogenesis.

1.8 Conclusion
Altogether, our findings suggest that pathogenicity of the IL-23/IL-17 axis is relevant
in AD pathogenesis. Genetic blockade and neutralizing IL-17 on Th17 cells promote
microglial phagocytic Aβ clearance and amelioration of AD-like pathology.
Importantly, our data suggests that systemic IL-17 producing Th17 cells in AD
pathogenesis leads to impaired microglial function. Therefore, investigating the IL-
23/IL-17 axis is imperative to promoting more efficacious treatment therapies for AD.

34

Figure 1. Increased CD4
+
T cells in APP/PS1 brains. Flow cytometric analysis of brain
mononuclear cells from 12-month- old APP/PS1 vs. WT mice. After stimulation with
PMA/ionomycin and brefeldin A for 4 h at 37°C, cells were immunostained for surface CD4 and
intracellular IL-17. (a) Representative dot plots and quantification of (b) absolute CD4
+
T cell
numbers, (c) percentage of CD4
+
IL-17
+
cells, and (d) absolute cell counts for CD4
+
IL-17
+
cells in
brains of APP/PS1 and WT mice. *p<.05, **p<0.01, ***p<0.001 by Student’s t test for independent
means (n = 4-6). Data are representative of three independent experiments. Error bars are ± SEM.
35

Figure 2. Absence of Th17 cells in CD4
cre
STAT3
fl/fl
mice. Flow cytometric analysis of mononuclear cells
in draining lymph nodes from 3-6mo old CD4
cre
STAT3
fl/fl
vs. WT mice stimulated with PMA/ionomycin
and Brefeldin A for 4h at 37 degrees Celsius. (a) Cells were immunostained for CD4 and intracellular IL-
17, IFN-y, IL-10 and analyzed by flow cytometry. (b) Supernatants were collected and analyzed by ELISA
for IL-17, IFN-y, and IL-10 cytokine secretion. ***p<0.01 by Student t test for each cytokine (n=6); n.s,
not significant. Data representative of three independent experiments. Error bars denote ± SEM.
36

Figure 3. Th17 deletion reduces amyloid burden in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice. Thioflavin S staining was
performed on brain sections from 12-month-old WT, Th17KO, APP/PS1, or APP/PS1+Th17KO mice. (a)
Photomicrographs showing Thioflavin S staining in the hippocampus were blindly assessed (b) for β-amyloid
burden (% immunolabeled area) and (c) total plaque numbers; n = 4-5 with 3 images per animal. Separate ELISA
analysis for Aβ1-40 or Aβ1-42 biochemistry was assessed in (d) triton-soluble or (e) guanidine-soluble brain
homogenate fractions from n = 4-5 mice per group. *p<.05, **p<0.01, ***p<0.001 by two-way ANOVA; n.d, not
detectable. Error bars are ± 1 SEM. Scale bar denotes 30 μm.

37

Figure 4. Increased Aβ phagocytosis in APP/PS1
+
CD4
cre
STAT3
fl/fl
mice. (a) Twelve-month-old
APP/PS1Stat3
fl/fl
(left) or APP/PS1 CD4
cre
Stat3
fl/fl
(right) hippocampi were stained for Iba1 (blue), CD68 (red), or
6E10 (green) for 3D quantitative in situ modeling (yellow inset) to analyze Aβ encapsulated in phagolysosomes.
Immunolabeling with CD68
+
quantified the volume of Aβ in phagolysosomes (b) and % phagolysosomes occupied
by Aβ (c); n = 10 images per group. **p<.01 by one-way ANOVA . Error bars are ± 1 SEM. Scale bar denotes 5
µm.

38

Figure 5. Generation of Aβ-specific Th17 cells. CD45.1 mice were subcutaneously injected with emulsified
complete Freund’s adjuvant (4 mg/ml) and aggregated Aβ1-42 (50 µg/mouse) for 8 days. Single cell suspensions
were made from lymph nodes and spleens for in vitro expansion of Aβ-Th17 cells by Aβ recall stimulation (20
µg/ml) in presence of Th17 polarizing cytokines, IL-1β (10 ng/ml) and IL-23 (10 ng/ml) for 4 days. Representative
dot plots (a) and flow cytometric analysis of mononuclear cells in co-culture were prepared and stained for CD4
+

and IL-17
+
(b). Supernatants were collected and stimulated with MOG35-55, PMA/ionomycin, or Aβ1-42 for 4 h and
analyzed by ELISA for IL-17, IFN-γ, or IL-10 production (c). ***p<0.01 by two-way ANOVA; n.s, not significant.
Error bars are ± 1 SEM.

39

Figure 6. Th17-dependent exacerbation of amyloid burden. Quantitative measurement of Thioflavin
S staining was performed in 12-month-old mouse hippocampus for each genotype. (a) Photomicrographs
showing Thioflavin S staining in the hippocampus were blindly assessed for (b) β-amyloid burden (%
immunolabeled area) and (c) total plaque numbers; n = 4-5 brains per group. Separate ELISA analysis
for Aβ1-40 or Aβ1-42 biochemistry was assessed in (d) triton-soluble or (e) guanidine-soluble brain
homogenate fractions from n = 4-5 mice per group. *p<.05, **p<0.01 by two-way ANOVA; n.d, not
detectable; n.s, not significant. Error bars are ± 1 SEM. Scale bar denotes 30 μm.
40

Figure 7. Th17-dependent Aβ phagocytosis. (a) Twelve-month-old mouse hippocampi were stained for Iba1 (blue),
CD68 (red), and 6E10 (green) from each indicated genotype for 3D quantitative in situ modeling (yellow inset) to analyze
Aβ encapsulated in phagolysosomes. Immunolabeling with CD68+ quantified the volume of Aβ in phagolysosomes (b)
and % phagolysosomes occupied by Aβ (c). *p<.05, **p<0.01, ***p<0.001; n = 5-6 per group. In vitro assay on adult
primary microglial-Th17 cocultures preincubation treatments followed by 30min wash and 1.5h Aβcy3 incubation for
3Dqism on volume of Aβ in phagolysosomes (d) and % phagolysosomes occupied by Aβ (e); n = 6 animals per treatment
group averaged from three separate experiments. *p<.05, **p<0.01, ***p<0.001 compared to Aβ1-42 treatment alone
by one-way ANOVA; n.s, not significant; n.d, not detectable. Error bars are ± 1 SEM. Scale bar denotes 5 µm.
41

Supplementary figure 1. Rorγt
+
depletion in CD4
+
T cells. Mouse draining lymph nodes were extracted and
homogenized from all four mouse genotypes and stimulated with PMA/ionomycin and Brefeldin A for 4 h and stained
with CD4, Roryγt and Foxp3 antibodies (a). Histograms show isotype control (red) vs. various genotypes (blue; b).
Supplementary figure 2. Increased microgliosis in APP/PS1
+
CD4
cre
STAT3fl/fl mice. Representative
photomicrographs of Iba1+ immunolabeled cells (a) in 12-month-old mouse hippocampal sections from each indicated
genotype. Percent immunolabeled Iba1+ area was assessed (b). For a-b, n=6 animals per group; **p<.01 compared to
STAT3fl/fl littermate controls; +++p<.001 compared to APP/PS1+STAT3fl/fl mice by one-way ANOVA. Error bars are
± SEM. Scale bar denotes 30 μm.
42

Supplementary figure 3. Decreased astrogliosis in APP/PS1
+
CD4
cre
STAT3fl/fl mice. Representative
photomicrographs of GFAP+ immunolabeled cells (a) in 12-month-old mouse hippocampal sections from each
indicated genotype. Percent immunolabeled GFAP+ area was assessed (b). For a-b, n=6 animals per group;
***p<.001 compared to STAT3fl/fl littermate controls; +++p<.001 compared to APP/PS1+STAT3fl/fl mice by two-
way ANOVA. Error bars are ± SEM. Scale bar denotes 30 μm.
Supplementary figure 4. CD4
cre
STAT3
fl/fl
does not alter PS1 abundance. Western blots of human (h) hPS1
holoprotein levels in frontal cortex homogenates from APP/PS1 mice with the indicated genotypes. β-actin is
shown as a loading control (a). Data are represented as mean ±SEM for n=5 samples for each group, with
APP/PS1
+
STAT3
fl/fl
signal normalized to 100%. Quantitative-PCR analysis of PS1 mRNA levels in frontal
cortex from APP/PS1
+
STAT3
fl/fl
mice versus APP/PS1
+
CD4
cre
STAT3
fl/fl
mice (b). Quantitation of hPS1 protein
levels in frontal cortex homogenates from APP/PS1 mice with the indicated genotypes. Expression levels are
normalized to β-actin (c). mRNA levels are normalized to HPRT, and n=5 per group.
43

Supplementary figure 5. Microglial Il23 expression in vitro. Adult primary
microglia were extracted and cultured from C57BL/6 mice and treated for 24 h prior
to RNA collection for Il23 qPCR. *p<.05 by one-way ANOVA. Error bars are ± SEM.
Supplementary figure 6. Inflammatory cytokine panel for primary microglia-T
h
17 in vitro coculture
assay. ELISA quantitation of IL-17 (a), TNF-α (b), IFN-γ (c), IL-10 (d), IL-1β (e), and IL-6 (f) expression
from supernatants collected after 24 h pretreatment followed by 30 min PBS wash and then 1.5 h incubation
with Aβ
cy3.
For each group, n=6 mice averaged from three separate experiments. PBS-control refers to
microglial-Th
17
cultures not preincubated with Aβ
1-42
but which received Aβ
cy3
for 1.5 h. *p<.05, **p<0.01,
***p<0.001 compared to Aβ
1-42
treatment alone by one-way ANOVA. Error bars are ± SEM.
44

TGF-β signaling control of the T cell response to cerebral amyloid-β
1
Im, K.W.,
1,2
Guillot-Sestier, M.V.,
1
Yuen, J.,
1
Rezai-Zadeh, K.,
1
Rodriguez, J.,
1
Town, T.

1
Zilkha Neurogenetic Institute, Department of Physiology & Biophysics, Keck School
of Medicine of the University of Southern California,1501 San Pablo Street, Los
Angeles, CA 90089-2821, USA
2
Trinity Biomedical Sciences Institute, Department of Physiology, Trinity College
Institute of Neuroscience, Trinity College Dublin, College Green Dublin 2, Ireland
Correspondence: ttown@usc.edu

1.9 Abstract

Previous data from our lab suggests that breaking immune tolerance in innate
immune cells via blocking TGFβ signaling attenuates AD-like pathology in mice and
rat models of AD. However, TGFβ signaling on adaptive immune T cell in response
to Aβ and implications for AD etiopathology remain unclear. We bred the
APPswePSEN1ΔE9 mouse model of cerebral amyloidosis (APP/PS1) with a dominant-
negative transgenic mouse that expresses an inhibitory form of TGF-β receptor type II
in CD4
+
T cells. Strikingly, our data reveal that APP/PS1
+
CD4-DNR
+
bitransgenic
mice present reduction of cerebral amyloid burden and CAA but have the net negative
45

consequence of early death; likely due to overly exuberant brain inflammation. Indeed,
reduced amyloid burden in APP/PS1
+
CD4-DNR
+
brains occur with increased CD4
+

T-cell numbers and increased Iba1 immunoreactivity. Furthermore, we observed
significantly increased numbers of CD45
+
CD3
+
T cells in parenchyma and blood
vessels in cerebral cortex and hippocampus of APP/PS1
+
CD4-DNR
+
mice. Our results
suggest that inhibition of T-cell TGF-β signaling induces brain influx of peripheral T
cells, recruitment of microglia and cerebral Aβ clearance. This raises the intriguing
possibility that infiltrating T cells may instruct microglia to restrict amyloid burden.

1.20 Results

1.20.1 Generation of a mouse model of cerebral amyloidosis expressing a
dominant negative form of TGF-β receptor in CD4
+
T cells.
Previously we had shown that blockade of innate immune TGF-β signaling mitigated
AD-like pathology in transgenic mice (Town et al., 2008); therefore, we asked whether
inhibiting the TGF-β pathway in adaptive immune T cells might also impact AD-like
pathology. To begin to address this, we crossed a mouse model of cerebral amyloidosis
(Jankowsky et al., 2004) with CD4-DNR animals that expresses the dnTGFβRII
transgene under regulatory control of the CD4 promoter (Gorelik and Flavell, 2000).
Fig. 8A and 8B shows a diagram of the CD4-dnTGFβRII construct used to generate
the CD4-DNR transgenic mice. The black box indicates the mouse CD4 promoter
placed upstream of the dnTGFβRII transgene, which was cloned into exon 3 of the
rabbit β-globin gene (including a downstream polyadenylation sequence).
46

Importantly, the dnTGFβRII transgene lacks the C-terminal signaling domain but
remains tethered to the cell membrane, where it acts as a dead-end decoy receptor to
bind exogenous TGF-β ligands without coupling to intracellular signaling.

1.20.2 Reduced amyloid burden in APP/PS1
+
CD4-DNR
+
mouse brains
To begin to understand the putative impact of brain T cell infiltration on cerebral
amyloidosis, we analyzed APP/PS1
+
CD4-DNR
+
vs. APP/PS1
+
CD4-DNR
-
mouse
brains at 6 months of age. APP/PS1
+
CD4-DNR
+
mice manifested significantly
reduced amyloid deposition by 43-58% across all three brain regions examined, as
revealed by 4G8 staining (Fig. 9A; CC, cingulate cortex; HC, hippocampus; EC,
entorhinal cortex; reductions versus APP/PS1
+
CD4-DNR
-
mice by t-test; * p<0.05).
To validate histological observations, we performed biochemical analyses of Aβ
species in brain homogenates from the same animals as above. When considering
detergent-soluble Aβ1-40 and Aβ1-42, we observed significant reductions in
APP/PS1
+
CD4-DNR
+
mice vs. APP/PS1
+
CD4-DNR
-
littermates from 22-30% (Fig
9B; t-test, * p<0.05). Additionally, extraction of detergent- insoluble Aβ peptides with
the chaotropic agent guanidine-HCl revealed 42% decreased Aβ1-40 and statistically
significant 59% reduced Aβ1-42 abundance (Fig. 9C,D; all mice vs. APP/PS1
+
CD4-
DNR
-
; by t-test, † trend p<0.10, * p<0.05).

47

1.20.3 Infiltration of T cells into brains of APP/PS1
+
CD4-DNR
+
mice
To begin understanding if brain T-cell infiltration impacted cerebral amyloidosis, we
analyzed APP/PS1
+
CD4- DNR
+
vs. APP/PS1
+
CD4-DNR
–
brains at 6 months of age.
Interestingly, reduced cerebral amyloid burden in APP/PS1
+
CD4-DNR
+
mice
correlates with increased CD4
+
T-cell numbers (Fig. 10A) and increased Iba1
immunoreactivity (Fig. 10B) in the entorhinal cortex, raising the intriguing possibility
that infiltrating T cells may instruct brain mononuclear phagocytes to restrict amyloid
burden. We performed preliminary confocal microscopy analyses for the pan-
leukocyte marker CD45 and for the T cell-specific marker CD3 in conjunction with
4G8 antibody to reveal Aβ deposits. We observe significantly increased numbers of
CD3
+
T cells in the choroid plexus of bitransgenic mice (Fig. 10C; by t-test, ** p <
0.01). Importantly, we detect CD3
+
T cells in parenchyma and blood vessels in the
cerebral cortex Fig. 10D,; by t-test, † a trend of p < 0.10) and in the hippocampus (Fig.
10E; by t-test, * p < 0.05) of APP/PS1
+
CD4-DNR
+
vs. APP/PS1
+
CD4-DNR
–
mice.

1.20.4 APP/PS1
+
CD4-DNR
+
mice exhibit early death
Four groups of littermates (APP/PS1
+
CD4-DNR
+
, APP/PS1
+
CD4-DNR
-
, APP/PS1
-
CD4-DNR
+
, and APP/PS1
-
CD4-DNR
-
cohorts) were monitored for up to 1 year.
Strikingly, APP/PS1
+
CD4-DNR
+
bitransgenic progeny die significantly earlier
compared to all three control littermates that have 25% or less death (and do not
significantly differ from one another) (Fig. 11; *** p < 0.001, ** p ≤ 0.01 vs.
APP/PS1
+
CD4-DNR
+
). Early death of bitransgenic mice is dependent on the APP/PS1
48

transgenes, which drive CNS β-amyloidosis.

1.21 Discussion
TGF-β family of proteins is a large group of secreted pleiotropic cytokines that have
a wide range of functional properties. These cytokines initiate signaling events in target
cells that affect cell-fate decisions, proliferation and survival by modifying cellular
transcription programs. Our lab previously reported that blocking TGF-β- SMAD2/3
innate immune signaling mitigates AD-like pathology (Town et al., 2008).
Specifically, these mice observed marked increases in the infiltration of peripheral
macrophages near cerebral vessels and Aβ plaques. These results suggested that
peripheral immune cells represent a new therapeutic target for AD. Thus, we
hypothesized that TGF-β inhibition on adaptive immune signaling also ameliorates AD-
like pathology in the same vein. To test this, we generated an APP/PS1 transgenic
mouse model deficient of TGF-β signaling on CD4
+
T cells and assessed AD-like
pathology at 6 months of age. Results show that APP/PS1
+
CD4-DNR
+
mice have
reduced amyloid burden compared to APP/PS1 mice. These results may be the result of
increased infiltration of peripheral immune cells. Indeed, infiltration of CD4
+
cells
were negatively correlated with 4G8
+
burden in APP/PS1
+
CD4- DNR
+
mice. Iba1
+

immunolabeling in APP/PS1
+
CD4-DNR
+
was negatively correlated with 4G8
+

burden. No observations of peripheral immune macrophages were found in the same
areas of the parenchyma, blood vessels, or choroid plexus area in these mice. These
results suggest that inhibition of TGF-β signaling on CD4
+
T cells enables infiltration
of peripheral immune T cells to infiltrate into plaque associated areas. We cannot
49

distinctly identify these CD4
+
T cell subsets in the CNS. However, one can postulate
that these could be the infiltration of Th1 and Th2 cells as both have been seen to
ameliorate and/or exacerbate AD-like pathology. Future studies will need to dissect
CD4
+
T cell involvement in AD pathogenesis as reports have indicated T cell mediated
influences on microglial phagocytosis of Aβ (Marsh et al., 2016). Indeed, while
inhibition of TGF-β in CD4
+
T cells ameliorated AD-like pathology from our previous
studies on innate immune signaling (Town et al., 2008), these results came at the net
negative cost of early death. Approximately 50% of APP/PS1
+
CD4-DNR
+
mice
survived by 3 months of age compared to littermate controls. This may suggest that
TGF-β signaling in CD4
+
T cells is essential for survival in chronic inflammatory
settings. This was supported by an approximate 75% survival probability in APP/PS1
-
CD4-DNR
+
mice. TGF-β signaling is involved in restricting Th1 and Th2 cells while
controlling Treg and Th17 cell polarization. The early death in chronic inflammatory
settings could be from the lack of an adaptive immune compartment necessary for
survival, as Th17 cells have been involved in protection against extracellular
pathogens.

1.22 Conclusion
Taken together, we report a critical role for TGF-β signaling in CD4
+
T cells in the
context of AD. While inhibition of TGF-β signaling in CD4
+
T cells ameliorated AD-
like pathology, survival probability in these mice was compromised. This suggests that
TGF-β on adaptive immune signaling is far more complex than expected. Further
studies investigating the balance between Th1, Th2, Tregs, and Th17 cells will elucidate
50

potential therapeutic efficacy in the treatment of AD.

Figure 8. Generation of CD4
+
T cell TGF-β signaling. Diagram showing (a) construction of CD4-DNR mice and (b)
mode of dnTGFβRII. CD4 promoter is placed upstream of the dominant negative TGF-βII receptor transgene. Dominant
negative TGF-βII receptor transgene lacks the C-terminal signaling domain but remains tethered to the cell membrane,
acting as a dead-end decoy receptor to bind to exogenous TGF-β ligands without coupling to intracellular signaling.
Figure 9. Reduced amyloid burden in APP/PS1
+
CD4-DNR
+
mice compared to APP/PS1
+
CD4-DNR
-
. (A)
Quantitation of amyloid burden using 4G8 labeling. (B) Quantitative analysis of CAA severity using ThioflavinS
labeling. EISA analysis of frontal cortex detergent-soluble (C) or guanidine-HCL-extracted (D) Aβ1-40 and Aβ1-42
species from APP/PS1
+
CD4-DNR
-
(white bars) and APP/PS1
+
CD4-DNR
+
(black bars) mice. Data are shown as mean
± SEM † p 0.1, * p 0.05, ** p<0.01; by t-test.
51

Figure 10. Infiltration of CD3
+
T cells correlated with amyloid burden. (a) Greater number of CD4
+
T
cells correlates with reduced amyloid burden evaluated by 4G8 immunoreactivity in brains of APP/PS1
+
CD4-
DNR
+
mice. The correlation is significant (**p<0.01). (b) increased activated microglial abundance evaluated
by Iba1 immunoreactivity (IR) correlates with reduced amyloid burden in the brain of APP/PS1
+
CD4-DNR
+

mice. The correlation is also significant (*p<0.05). Greater number of CD45
+
CD3
+
T cells in the choroid
plexus (c), cingulate cortex (d), hippocampus € and entorhinal cortex (f) of APP/PS1
+
CD4-DNR
+
mice (black
bars) compared to APP/PS1
+
CD4-DNR
-
mice (white bars). (d-f) T cells are found in the parenchyma (P) and
blood vessels (V). (g) representative confocal images of CD3
+
(green) T cells observed in APP/PS1
+
CD4-
DNR
+
mice. Amyloid plaques are labeled with 4G8 and nuclei with DAPI. Inserts show the morphology of
CD3
+
(green) and CD45
+
(red) double -positive T cells.
52

Figure 11. APP/PS1
+
CD4-DNR
+
mice display early death. Diagrams showing (a) construction of CD4-
DNR mice and (b) mode of dnTGFβRII action. (c) the four groups of progenies resulting from crossbreeding
of APP/PS1
+
and CD4-DNR
+
mice were followed for 1 year. Survival curves and Kaplan-Meier statistics
show early death in APP/PS1
+
CD4-DNR
+
mice.
53

References
1. Alzheimer's Association. (2019). 2019 Alzheimer's disease facts and figures.
Alzheimer's & Dementia, 15(3), 321-387.
2. Bature, F., Guinn, B. A., Pang, D., & Pappas, Y. (2017). Signs and symptoms
preceding the diagnosis of Alzheimer’s disease: a systematic scoping review of
literature from 1937 to 2016. BMJ open, 7(8), e015746.
3. Weller, J., & Budson, A. (2018). Current understanding of Alzheimer’s disease
diagnosis and treatment. F1000Research, 7.
4. Reddy, P. H., & Beal, M. F. (2008). Amyloid beta, mitochondrial dysfunction and
synaptic damage: implications for cognitive decline in aging and Alzheimer's
disease. Trends in molecular medicine, 14(2), 45- 53.
5. Johnson, K. A., Minoshima, S., Bohnen, N. I., Donohoe, K. J., Foster, N. L.,
Herscovitch, P., ... & Carrillo, M. C. (2013). Update on appropriate use criteria for
amyloid PET imaging: dementia experts, mild cognitive impairment, and
education. Journal of Nuclear Medicine, 54(7), 1011-1013.
6. Maxim, V., Şendur, L., Fadili, J., Suckling, J., Gould, R., Howard, R., & Bullmore,
E. (2005). Fractional Gaussian noise, functional MRI and Alzheimer's disease.
Neuroimage, 25(1), 141-158.
7. Mattsson, N., Zetterberg, H., Hansson, O., Andreasen, N., Parnetti, L., Jonsson, M.,
... & Rich, K. (2009). CSF biomarkers and incipient Alzheimer disease in patients
with mild cognitive impairment. Jama, 302(4), 385- 393.
8. Piaceri, I., Nacmias, B., & Sorbi, S. (2013). Genetics of familial and sporadic
Alzheimer’s disease. Front Biosci (Elite Ed), 5(1), 167-177.
54

9. Cacace, R., Sleegers, K., & Van Broeckhoven, C. (2016). Molecular genetics of
early-onset Alzheimer's disease revisited. Alzheimer's & Dementia, 12(6), 733-748.
10. Selkoe, D. J. (2001). Alzheimer's disease: genes, proteins, and therapy. Physiological
reviews, 81(2), 741-766.
11. Chasseigneaux, S., & Allinquant, B. (2012). Functions of Aβ, sAPPα and sAPPβ:
similarities and differences. Journal of neurochemistry, 120, 99-108.
12. Zhang, Y. W., Thompson, R., Zhang, H., & Xu, H. (2011). APP processing in
Alzheimer's disease. Molecular brain, 4(1), 3.
13. McLean, C. A., Cherny, R. A., Fraser, F. W., Fuller, S. J., Smith, M. J.,
Vbeyreuther, K., ... & Masters, C. L. (1999). Soluble pool of Aβ amyloid as a
determinant of severity of neurodegeneration in Alzheimer's disease. Annals of
neurology, 46(6), 860-866.
14. Cappai, R., & Barnham, K. J. (2008). Delineating the mechanism of Alzheimer’s
disease Aβ peptide neurotoxicity. Neurochemical research, 33(3), 526-532.
15. Eimer, W. A., & Vassar, R. (2013). Neuron loss in the 5XFAD mouse model of
Alzheimer’s disease correlates with intraneuronal Aβ 42 accumulation and
Caspase-3 activation. Molecular neurodegeneration, 8(1), 2.
16. Sterniczuk, R., Dyck, R. H., LaFerla, F. M., & Antle, M. C. (2010).
Characterization of the 3xTg-AD mouse model of Alzheimer's disease: part 1.
Circadian changes. Brain research, 1348, 139-148.
17. Jankowsky, J. L., Slunt, H. H., Gonzales, V., Jenkins, N. A., Copeland, N. G., &
Borchelt, D. R. (2004). APP processing and amyloid deposition in mice haplo-
insufficient for presenilin 1. Neurobiology of aging, 25(7), 885-892.
55

18. Palasí, A., Gutiérrez-Iglesias, B., Alegret, M., Pujadas, F., Olabarrieta, M., Liébana,
D., ... & Boada, M. (2015). Differentiated clinical presentation of early and late-
onset Alzheimer’s disease: is 65 years of age providing a reliable threshold?
Journal of neurology, 262(5), 1238-1246.
19. Ma, J., Brewer Jr, H. B., & Potter, H. (1996). Alzheimer Aβ neurotoxicity:
promotion by antichymotrypsin, apoE4; inhibition by Aβ-related peptides.
Neurobiology of aging, 17(5), 773-780.
20. Zhu, L., Zhong, M., Elder, G. A., Sano, M., Holtzman, D. M., Gandy, S., ... & Cai,
D. (2015). Phospholipid dysregulation contributes to ApoE4-associated cognitive
deficits in Alzheimer’s disease pathogenesis. Proceedings of the National
Academy of Sciences, 112(38), 11965-11970.
21. Mahley, R. W., & Huang, Y. (2006). Apolipoprotein (apo) E4 and Alzheimer's
disease: unique conformational and biophysical properties of apoE4 can modulate
neuropathology. Acta Neurologica Scandinavica, 114, 8- 14.
22. Uchoa, M. F., Moser, V. A., & Pike, C. J. (2016). Interactions between
inflammation, sex steroids, and Alzheimer’s disease risk factors. Frontiers in
neuroendocrinology, 43, 60-82.
23. Casey, D. A., Antimisiaris, D., & O’Brien, J. (2010). Drugs for Alzheimer’s
disease: are they effective? Pharmacy and Therapeutics, 35(4), 208.
24. Holmes, C. (2013). Systemic inflammation and Alzheimer's disease.
Neuropathology and applied neurobiology, 39(1), 51-68.
25. Mortimer, J. A., & Graves, A. B. (1993). Education and other socioeconomic
determinants of dementia and Alzheimer's disease. NEUROLOGY-
56

MINNEAPOLIS-, 43, 39-39.
26. Lye, T. C., & Shores, E. A. (2000). Traumatic brain injury as a risk factor for
Alzheimer's disease: a review. Neuropsychology review, 10(2), 115-129.
27. Jiang, C., Li, G., Huang, P., Liu, Z., & Zhao, B. (2017). The gut microbiota and
Alzheimer’s disease. Journal of Alzheimer's Disease, 58(1), 1-15.
28. Exalto, L. G., Whitmer, R. A., Kappele, L. J., & Biessels, G. J. (2012). An update
on type 2 diabetes, vascular dementia and Alzheimer's disease. Experimental
gerontology, 47(11), 858-864.
29. Haan, M. N. (2006). Therapy Insight: type 2 diabetes mellitus and the risk of late-
onset Alzheimer's disease. Nature clinical practice Neurology, 2(3), 159-166.
30. Paillard, T., Rolland, Y., & de Souto Barreto, P. (2015). Protective effects of
physical exercise in Alzheimer's disease and Parkinson's disease: a narrative
review. Journal of clinical neurology, 11(3), 212-219.
31. Radak, Z., Hart, N., Sarga, L., Koltai, E., Atalay, M., Ohno, H., & Boldogh, I.
(2010). Exercise plays a preventive role against Alzheimer's disease. Journal of
Alzheimer's disease, 20(3), 777-783.
32. Scarmeas, N., Stern, Y., Tang, M. X., Mayeux, R., & Luchsinger, J. A. (2006).
Mediterranean diet and risk for Alzheimer's disease. Annals of Neurology: Official
Journal of the American Neurological Association and the Child Neurology
Society, 59(6), 912-921.
33. Tsivgoulis, G., Judd, S., Letter, A. J., Alexandrov, A. V., Howard, G., Nahab, F., ... &
Wadley, V. G. (2013). Adherence to a Mediterranean diet and risk of incident cognitive
impairment. Neurology, 80(18), 1684-1692.
57

34. Innes, K. E., & Selfe, T. K. (2014). Meditation as a therapeutic intervention for
adults at risk for Alzheimer’s disease–potential benefits and underlying
mechanisms. Frontiers in psychiatry, 5, 40.
35. Amaducci, L. (1996). Alzheimer's original patient. Science, 274(5286), 324-329.
36. Hardy, J., & Allsop, D. (1991). Amyloid deposition as the central event in the
etiology of Alzheimer's disease. Trends in pharmacological sciences, 12, 383-388.
37. Hardy, J. A., & Higgins, G. A. (1992). Alzheimer's disease: the
amyloid cascade hypothesis. Science, 256(5054), 184-186.
38. Egwuagu, C. E. (2009). STAT3 in CD4+ T helper cell differentiation and
inflammatory diseases. Cytokine, 47(3), 149-156.
39. Stern, Y., Albert, S. M., Sano, M., Richards, M., Miller, L., Folstein, M., ... &
Lafleche, G. (1994). Assessing patient dependence in Alzheimer's disease. Journal
of Gerontology, 49(5), M216-M222.
40. Wisniewski, K. E., Wisniewski, H. M., & Wen, G. Y. (1985). Occurrence of
neuropathological changes and dementia of Alzheimer's disease in Down's
syndrome. Annals of Neurology: Official Journal of the American Neurological
Association and the Child Neurology Society, 17(3), 278-282.
41. Holland, A. J., Hon, J., Huppert, F. A., Stevens, F., & Watson, P. (1998).
Population-based study of the prevalence and presentation of dementia in adults
with Down's syndrome. The British Journal of Psychiatry, 172(6), 493-498.
42. Bakkar, R. M., Luo, G., Webb, T. A., Crutcher, K. A., & de Courten-Myers, G. M.
(2010). Down's Syndrome with Alzheimer's Disease-Like Pathology: What Can It
Teach Us about the Amyloid Cascade Hypothesis? International Journal of
58

Alzheimer’s Disease, 2010.
43. Lott, I. T. (2012). Neurological phenotypes for Down syndrome across the life
span. In Progress in brain research (Vol. 197, pp. 101-121). Elsevier.
44. Selkoe, D. J., & Hardy, J. (2016). The amyloid hypothesis of Alzheimer's disease at
25 years. EMBO molecular medicine, 8(6), 595-608.
45. Steinberg, S., Stefansson, H., Jonsson, T., Johannsdottir, H., Ingason, A., Helgason, H.,
... & Kong, A. (2015). Loss-of-function variants in ABCA7 confer risk of Alzheimer's
disease. Nature genetics, 47(5), 445-447.
46. Kline, A. (2012). Apolipoprotein E, amyloid-ß clearance and therapeutic
opportunities in Alzheimer's disease. Alzheimer's research & therapy, 4(4), 32.
47. Engelhardt, B. (2010). T cell migration into the central nervous system during
health and disease: different molecular keys allow access to different central
nervous system compartments. Clinical and Experimental Neuroimmunology, 1(2),
79-93.
48. Bilbo, S. D., & Schwarz, J. M. (2009). Early-life programming of later-life brain
and behavior: a critical role for the immune system. Frontiers in behavioral
neuroscience, 3, 14.
49. Lucin, K. M., & Wyss-Coray, T. (2009). Immune activation in brain aging and
neurodegeneration: too much or too little? Neuron, 64(1), 110-122.
50. Bialas, A. R., & Stevens, B. (2013). TGF-β signaling regulates neuronal C1q
expression and developmental synaptic refinement. Nature neuroscience, 16(12),
1773-1782.
51. Veerhuis, R., Nielsen, H. M., & Tenner, A. J. (2011). Complement in the brain.
59

Molecular immunology, 48(14), 1592-1603.
52. Heneka, M. T., Carson, M. J., El Khoury, J., Landreth, G. E., Brosseron, F.,
Feinstein, D. L., ... & Herrup, K. (2015). Neuroinflammation in Alzheimer's
disease. The Lancet Neurology, 14(4), 388-405.
53. Solito, E., & Sastre, M. (2012). Microglia function in Alzheimer’s disease. Frontiers in
pharmacology, 3, 14.
54. Aloisi, F., Ria, F., & Adorini, L. (2000). Regulation of T-cell responses by CNS
antigen-presenting cells: different roles for microglia and astrocytes. Immunology
today, 21(3), 141-147.
55. Zheng, C., Zhou, X. W., & Wang, J. Z. (2016). The dual roles of cytokines in
Alzheimer’s disease: update on interleukins, TNF-α, TGF-β and IFN-γ.
Translational neurodegeneration, 5(1), 7.
56. Feghali, C. A., & Wright, T. M. (1997). Cytokines in acute and chronic
inflammation. Front Biosci, 2(1), d12- d26.
57. Bornstein, S. R., Rutkowski, H., & Vrezas, I. (2004). Cytokines and
steroidogenesis. Molecular and cellular endocrinology, 215(1-2), 135-141.
58. Duvallet, E., Semerano, L., Assier, E., Falgarone, G., & Boissier, M. C. (2011).
Interleukin-23: a key cytokine in inflammatory diseases. Annals of medicine, 43(7),
503-511.
59. Vom Berg, J., Prokop, S., Miller, K. R., Obst, J., Kälin, R. E., Lopategui-Cabezas,
I., ... & Winter, Y. (2012). Inhibition of IL-12/IL-23 signaling reduces Alzheimer's
disease–like pathology and cognitive decline. Nature medicine, 18(12), 1812.
60. Guillot-Sestier, M. V., Doty, K. R., Gate, D., Rodriguez Jr, J., Leung, B. P., Rezai-
60

Zadeh, K., & Town, T. (2015). Il10 deficiency rebalances innate immunity to
mitigate Alzheimer-like pathology. Neuron, 85(3), 534- 548.
61. Town, T., Laouar, Y., Pittenger, C., Mori, T., Szekely, C. A., Tan, J., ... & Flavell,
R. A. (2008). Blocking TGF-β–Smad2/3 innate immune signaling mitigates
Alzheimer-like pathology. Nature medicine, 14(6), 681- 687.
62. Baik, S. H., Kang, S., Lee, W., Choi, H., Chung, S., Kim, J. I., & Mook-Jung, I.
(2019). A breakdown in metabolic reprogramming causes microglia dysfunction in
Alzheimer's disease. Cell metabolism, 30(3), 493- 507.
63. El Khoury, J., Hickman, S. E., Thomas, C. A., Loike, J. D., & Silverstein, S. C.
(1998). Microglia, scavenger receptors, and the pathogenesis of Alzheimer’s
disease. Neurobiology of aging, 19(1), S81-S84.
64. Yeh, F. L., Hansen, D. V., & Sheng, M. (2017). TREM2, microglia, and
neurodegenerative diseases. Trends in molecular medicine, 23(6), 512-533.
65. Webster, S. D., Park, M., Fonseca, M. I., & Tenner, A. J. (2000). Structural and
functional evidence for microglial expression of C1qRP, the C1q receptor that
enhances phagocytosis. Journal of leukocyte biology, 67(1), 109-116.
66. Abumrad, N. A., Ajmal, M., Pothakos, K., & Robinson, J. K. (2005). CD36
expression and brain function: does CD36 deficiency impact learning ability?
Prostaglandins & other lipid mediators, 77(1-4), 77-83.
61

67. Coraci, I. S., Husemann, J., Berman, J. W., Hulette, C., Dufour, J. H., Campanella,
G. K., ... & El Khoury, J.B. (2002). CD36, a class B scavenger receptor, is
expressed on microglia in Alzheimer's disease brains and can mediate production
of reactive oxygen species in response to β-amyloid fibrils. The American journal
of pathology, 160(1), 101-112.
68. Moore, K. J., El Khoury, J., Medeiros, L. A., Terada, K., Geula, C., Luster, A. D.,
& Freeman, M. W. (2002). A CD36-initiated signaling cascade mediates
inflammatory effects of β-amyloid. Journal of Biological Chemistry, 277(49),
47373-47379.
69. Neumann, H., & Takahashi, K. (2007). Essential role of the microglial triggering
receptor expressed on myeloid cells-2 (TREM2) for central nervous tissue
immune homeostasis. Journal of neuroimmunology, 184(1-2), 92-99.
70. Thrash, J. C., Torbett, B. E., & Carson, M. J. (2009). Developmental regulation of
TREM2 and DAP12 expression in the murine CNS: implications for Nasu-Hakola
disease. Neurochemical research, 34(1), 38-45.
71. Wang, Y., Cella, M., Mallinson, K., Ulrich, J. D., Young, K. L., Robinette, M. L.,
... & Holtzman, D. M. (2015). TREM2 lipid sensing sustains the microglial
response in an Alzheimer’s disease model. Cell, 160(6), 1061-1071.
72. Stevens, B., Allen, N. J., Vazquez, L. E., Howell, G. R., Christopherson, K. S., Nouri, N.,
... & Sher, A. (2007). The classical complement cascade mediates CNS synapse
elimination. Cell, 131(6), 1164-1178.
73. Stephan, A. H., Barres, B. A., & Stevens, B. (2012). The complement system: an
unexpected role in synaptic pruning during development and disease. Annual
62

review of neuroscience, 35, 369-389.
74. Leung, B. P., Doty, K. R., Uchoa, M. F., Im, K. W., Miller, C. J., Vesling, A., Rodriguez.
J., Tenner, A., Town, T. (2020). TREM2/C1q interaction in a mouse model of cerebral
amyloidosis (in preparation)
75. Blum, J. S., Wearsch, P. A., & Cresswell, P. (2013). Pathways of antigen
processing. Annual review of immunology, 31, 443-473.
76. Nakagawa, T. Y., & Rudensky, A. Y. (1999). The role of lysosomal proteinases in
MHC class Il‐mediated antigen processing and presentation. Immunological
reviews, 172(1), 121-129.
77. Yang, Y., Waters, J. B., Frueh, K., & Peterson, P. A. (1992). Proteasomes are
regulated by interferon gamma: implications for antigen processing. Proceedings
of the National Academy of Sciences, 89(11), 4928-4932.
78. Guermonprez, P., Valladeau, J., Zitvogel, L., Théry, C., & Amigorena, S. (2002).
Antigen presentation and T cell stimulation by dendritic cells. Annual review of
immunology, 20(1), 621-667.
79. Janeway, C. A. (2001). How the immune system works to protect the host from
infection: a personal view. Proceedings of the National Academy of Sciences,
98(13), 7461-7468.
80. Ferretti, M. T., Merlini, M., Späni, C., Gericke, C., Schweizer, N., Enzmann, G.,
... & Nitsch, R. M. (2016). T-cell brain infiltration and immature antigen-presenting
cells in transgenic models of Alzheimer’s disease- like cerebral amyloidosis. Brain,
behavior, and immunity, 54, 211-225.
81. Ridge, P. G., Hoyt, K. B., Boehme, K., Mukherjee, S., Crane, P. K., Haines, J. L., ... &
63

Kauwe, J. S. (2016). Assessment of the genetic variance of late-onset Alzheimer's
disease. Neurobiology of aging, 41, 200-e13.
82. Tooyama, I., Kimura, H., Akiyama, H., & McGeer, P. L. (1990). Reactive
microglia express class I and class II major histocompatibility complex antigens in
Alzheimer's disease. Brain research, 523(2), 273-280.
83. Rogers, J., Luber-Narod, J., Styren, S. D., & Civin, W. H. (1988). Expression of
immune system-associated antigens by cells of the human central nervous system:
relationship to the pathology of Alzheimer's disease. Neurobiology of aging, 9,
339-349.
84. Kloetzel, P. M. (2004). The proteasome and MHC class I antigen processing.
Biochimica et Biophysica Acta (BBA)-Molecular Cell Research, 1695(1-3), 225-
233.
85. Cebrián, C., Zucca, F. A., Mauri, P., Steinbeck, J. A., Studer, L., Scherzer, C. R., ...
& Zecca, L. (2014). MHC- I expression renders catecholaminergic neurons
susceptible to T-cell-mediated degeneration. Nature communications, 5(1), 1-14.
86. Itagaki, S., McGeer, P. L., & Akiyama, H. (1988). Presence of T-cytotoxic
suppressor and leucocyte common antigen positive cells in Alzheimer's disease
brain tissue. Neuroscience letters, 91(3), 259-264.
87. Gehrmann, J., Banati, R. B., & Kreutzberg, G. W. (1993). Microglia in the immune
surveillance of the brain: human microglia constitutively express HLA-DR
molecules. Journal of neuroimmunology, 48(2), 189-198.
88. Mansouri, L., Klai, S., Gritli, N., Fekih-Mrissa, N., Messalmani, M., Bedoui, I., ...
& Mrissa, R. (2015). Association of HLA-DR/DQ polymorphism with Alzheimer’s
64

disease. The American journal of the medical sciences, 349(4), 334-337.
89. Kumar, B. V., Connors, T. J., & Farber, D. L. (2018). Human T cell development,
localization, and function throughout life. Immunity, 48(2), 202-213.
90. Ding, Y., Xu, J., & Bromberg, J. S. (2012). Regulatory T cell migration during an
immune response. Trends in immunology, 33(4), 174-180.
91. Tsuji, M., Suzuki, K., Kitamura, H., Maruya, M., Kinoshita, K., Ivanov, I. I., ... &
Fagarasan, S. (2008). Requirement for lymphoid tissue-inducer cells in isolated follicle
formation and T cell-independent immunoglobulin A generation in the gut. Immunity,
29(2), 261-271.
92. Mayer, L. (2003). Mucosal immunity. Pediatrics, 111(Supplement 3), 1595-1600.
93. Saitoh, T., Fujiwara, M., & Asakura, H. (1990). Ductal lesions of exocrine glands
and insulitis induced by L3T4+ T cells following graft-versus-host reaction due to
major histocompatibility complex class II disparity. Clinical immunology and
immunopathology, 57(3), 339-350.
94. Nishimura, S., Manabe, I., Nagasaki, M., Eto, K., Yamashita, H., Ohsugi, M., ...
& Yoshimura, K. (2009). CD8+ effector T cells contribute to macrophage
recruitment and adipose tissue inflammation in obesity. Nature medicine, 15(8),
914.
95. Moalem, G., Leibowitz–Amit, R., Yoles, E., Mor, F., Cohen, I. R., & Schwartz, M.
(1999). Autoimmune T cells protect neurons from secondary degeneration after
central nervous system axotomy. Nature medicine, 5(1), 49-55.
96. Tzartos, J. S., Friese, M. A., Craner, M. J., Palace, J., Newcombe, J., Esiri, M.
M., & Fugger, L. (2008). Interleukin-17 production in central nervous system-
65

infiltrating T cells and glial cells is associated with active disease in multiple
sclerosis. The American journal of pathology, 172(1), 146-155.
97. Ellwardt, E., Walsh, J. T., Kipnis, J., & Zipp, F. (2016). Understanding the role of
T cells in CNS homeostasis. Trends in immunology, 37(2), 154-165.
98. Town, T., Tan, J., Flavell, R. A., & Mullan, M. (2005). T-cells in Alzheimer’s
disease. Neuromolecular medicine, 7(3), 255-264.
99. Togo, T., Akiyama, H., Iseki, E., Kondo, H., Ikeda, K., Kato, M., ... & Kosaka, K.
(2002). Occurrence of T cells in the brain of Alzheimer's disease and other
neurological diseases. Journal of neuroimmunology, 124(1- 2), 83-92.
100. Gemechu, J. M., & Bentivoglio, M. (2012). T cell recruitment in the brain
during normal aging. Frontiers in cellular neuroscience, 6, 38.
101. Browne, T. C., McQuillan, K., McManus, R. M., O’Reilly, J. A., Mills, K.
H., & Lynch, M. A. (2013). IFN-γ Production by Amyloid β–Specific Th1 Cells
Promotes Microglial Activation and Increases Plaque Burden in a Mouse Model of
Alzheimer’s Disease. The Journal of Immunology, 190(5), 2241-2251.
102. Baruch, K., Rosenzweig, N., Kertser, A., Deczkowska, A., Sharif, A. M.,
Spinrad, A., ... & Schwartz, M. (2015). Breaking immune tolerance by targeting
Foxp3+ regulatory T cells mitigates Alzheimer’s disease pathology. Nature
communications, 6(1), 1-12.
103. Britschgi, M., & Wyss‐Coray, T. (2007). Systemic and acquired immune
responses in Alzheimer's disease. International review of neurobiology, 82, 205-
233.
104. Shechter, R., London, A., & Schwartz, M. (2013). Orchestrated leukocyte
66

recruitment to immune- privileged sites: absolute barriers versus educational gates.
Nature Reviews Immunology, 13(3), 206-218.
105. Sardi, F., Fassina, L., Venturini, L., Inguscio, M., Guerriero, F., Rolfo, E., &
Ricevuti, G. (2011).

Alzheimer's disease, autoimmunity and inflammation. The good, the bad and
the ugly. Autoimmunity reviews, 11(2), 149-153.
106. Monsonego, A., & Weiner, H. L. (2003). Immunotherapeutic
approaches to Alzheimer's disease. Science, 302(5646), 834-838.
107. McEver, R. P. (2015). Selectins: initiators of leucocyte adhesion and
signaling at the vascular wall. Cardiovascular research, 107(3), 331-339.
108. Holmes, C., Boche, D., Wilkinson, D., Yadegarfar, G., Hopkins, V., Bayer,
A., ... & Zotova, E. (2008). Long-term effects of Aβ42 immunisation in
Alzheimer's disease: follow-up of a randomised, placebo- controlled phase I trial.
The Lancet, 372(9634), 216-223.
109. Guillot-Sestier, M. V., Doty, K. R., & Town, T. (2015). Innate immunity
fights Alzheimer's disease. Trends in neurosciences, 38(11), 674-681.
110. Fiala, M. (2010). Re-balancing of inflammation and Aβ immunity as a
therapeutic for Alzheimer's disease- view from the bedside. CNS & Neurological
Disorders-Drug Targets (Formerly Current Drug Targets-CNS & Neurological
Disorders), 9(2), 192-196.
111. Takeshita, Y., & Ransohoff, R. M. (2012). Inflammatory cell trafficking across
the blood–brain barrier: chemokine regulation and in vitro models. Immunological
67

reviews, 248(1), 228-239.
112. Oppenheimer-Marks, N., Brezinschek, R. I., Mohamadzadeh, M., Vita, R.,
& Lipsky, P. E. (1998). Interleukin 15 is produced by endothelial cells and
increases the transendothelial migration of T cells in vitro and in the SCID mouse-
human rheumatoid arthritis model in vivo. The Journal of clinical
investigation, 101(6), 1261-1272.
113. Muller, W. A., Weigl, S. A., Deng, X. I. A. O. H. U. I., & Phillips, D. M.
(1993). PECAM-1 is required for transendothelial migration of leukocytes. The
Journal of experimental medicine, 178(2), 449-460.
114. Gérard, A., van der Kammen, R. A., Janssen, H., Ellenbroek, S. I., & Collard,
J. G. (2009). The Rac activator Tiam1 controls efficient T-cell trafficking and route
of transendothelial migration. Blood, The Journal of the American Society of
Hematology, 113(24), 6138-6147.
115. Heasman, S. J., & Ridley, A. J. (2010). Multiple roles for RhoA during T cell
transendothelial migration. Small GTPases, 1(3), 553-563.
116. Mö hle, R., Bautz, F., Rafii, S., Moore, M. A., Brugger, W., & Kanz, L. (1998).
The chemokine receptor CXCR-4 is expressed on CD34+ hematopoietic
progenitors and leukemic cells and mediates transendothelial migration induced by
stromal cell-derived factor-1. Blood, The Journal of the American Society of
Hematology, 91(12), 4523-4530.
117. Montagne, A., Barnes, S. R., Sweeney, M. D., Halliday, M. R., Sagare, A. P., Zhao,
Z., ... & Harrington, M. G. (2015). Blood-brain barrier breakdown in the aging human
hippocampus. Neuron, 85(2), 296-302.
68

118. Zlokovic, B. V. (2011). Neurovascular pathways to neurodegeneration in
Alzheimer's disease and other disorders. Nature Reviews Neuroscience, 12(12),
723-738.
119. Farkas, E., & Luiten, P. G. (2001). Cerebral microvascular pathology in aging
and Alzheimer's disease. Progress in neurobiology, 64(6), 575-611.
120. Zenaro, E., Pietronigro, E., Della Bianca, V., Piacentino, G., Marongiu, L.,
Budui, S., ... & Montresor, A. (2015). Neutrophils promote Alzheimer's disease–
like pathology and cognitive decline via LFA-1 integrin. Nature medicine, 21(8),
880.
121. Frohman, E. M., Frohman, T. C., Gupta, S., de Fougerolles, A., & van den
Noort, S. (1991). Expression of intercellular adhesion molecule 1 (ICAM-1) in
Alzheimer's disease. Journal of the neurological sciences, 106(1), 105-111.
122. Liu, Y. J., Guo, D. W., Tian, L., Shang, D. S., Zhao, W. D., Li, B., ... & Chen,
Y. H. (2010). Peripheral T cells derived from Alzheimer's disease patients
overexpress CXCR2 contributing to its transendothelial migration, which is
microglial TNF-α-dependent. Neurobiology of aging, 31(2), 175-188.
123. Marsh, S. E., Abud, E. M., Lakatos, A., Karimzadeh, A., Yeung, S. T.,
Davtyan, H., ... & Inlay, M. A. (2016). The adaptive immune system restrains
Alzheimer’s disease pathogenesis by modulating microglial function. Proceedings
of the National Academy of Sciences, 113(9), E1316-E1325.
124. Im, K. W., Guillot-Sestier, M. V., Doty, K. R., Rodriguez, J., Town, T. (2020)
TGF-β signaling on CD4
+
T cells attenuate AD-like pathology in a mouse model
of cerebral amyloidosis. (in preparation)
69

125. Arlehamn, C. S. L., Garretti, F., Sulzer, D., & Sette, A. (2019). Roles for the
adaptive immune system in Parkinson’s and Alzheimer’s diseases. Current opinion
in immunology, 59, 115-120.
126. Lambracht-Washington, D., Qu, B. X., Fu, M., Anderson, L. D., Stüve, O.,
Eagar, T. N., & Rosenberg, R. N. (2011). DNA immunization against amyloid beta
42 has high potential as safe therapy for Alzheimer’s disease as it diminishes
antigen-specific Th1 and Th17 cell proliferation. Cellular and molecular
neurobiology, 31(6), 867.
127. Butovsky, O., Talpalar, A. E., Ben-Yaakov, K., & Schwartz, M. (2005).
Activation of microglia by aggregated β-amyloid or lipopolysaccharide impairs
MHC-II expression and renders them cytotoxic whereas IFN-γ and IL-4 render
them protective. Molecular and Cellular Neuroscience, 29(3), 381-393.
128. Baruch, K., Ron-Harel, N., Gal, H., Deczkowska, A., Shifrut, E., Ndifon, W.,
... & Berkutzki, T. (2013). CNS-specific immunity at the choroid plexus shifts
toward destructive Th2 inflammation in brain aging. Proceedings of the
National Academy of Sciences, 110(6), 2264-2269.
129. Kim, H. D., Jin, J. J., Maxwell, J. A., & Fukuchi, K. I. (2007). Enhancing Th2
immune responses against amyloid protein by a DNA prime-adenovirus boost
regimen for Alzheimer's disease. Immunology letters, 112(1), 30-38.
130. Ghochikyan, A., Mkrtichyan, M., Petrushina, I., Movsesyan, N., Karapetyan,
A., Cribbs, D. H., & Agadjanyan, M. G. (2006). Prototype Alzheimer's disease
epitope vaccine induced strong Th2-type anti-Aβ antibody response with Alum to
Quil A adjuvant switch. Vaccine, 24(13), 2275-2282.
70

131. Jonuleit, H., & Schmitt, E. (2003). The regulatory T cell family: distinct
subsets and their interrelations. The Journal of Immunology, 171(12), 6323-6327.
132. Pellicanò, M., Larbi, A., Goldeck, D., Colonna-Romano, G., Buffa, S., Bulati, M.,
... & Derhovanessian,

E. (2012). Immune profiling of Alzheimer patients. Journal of neuroimmunology, 242(1-
2), 52-59.
133. Oberstein, T. J., Taha, L., Spitzer, P., Hellstern, J., Herrmann, M., Kornhuber,
J., & Maler, J. M. (2018). Imbalance of circulating Th17 and regulatory T cells in
Alzheimer’s disease: a case control study. Frontiers in immunology, 9, 1213.
134. Zúñiga, L. A., Jain, R., Haines, C., & Cua, D. J. (2013). Th17 cell development:
from the cradle to the grave. Immunological reviews, 252(1), 78-88.
135. McCarthy, M. K., Zhu, L., Procario, M. C., & Weinberg, J. B. (2014). IL-17
contributes to neutrophil recruitment but not to control of viral replication during
acute mouse adenovirus type 1 respiratory infection. Virology, 456, 259-267.
136. Chen, K., Fisher, C., Zheng, M., Weaver, C., & Kolls, J. (2011). Immunization
of heat-killed Klebsiella pneumoniae promotes strong mucosal Th17 responses and
protects mice from live infection (59.4).
137. Muranski, P., Franco-Colon, M., Chinnasamy, D., Chen, K., Stegemann, S.,
Sui, J., ... & Kolls, J. (2014). Ex Vivo Generation of CD4+ Th17 Cells to Prevent
and Treat Infection from Antibiotic-Resistant Klebsiella Pneumoniae in
Immunocompromised Patients.
138. Omenetti, S., & Pizarro, T. T. (2015). The Treg/Th17 axis: a dynamic balance
71

regulated by the gut microbiome. Frontiers in immunology, 6, 639.
139. Troy, E. B., & Kasper, D. L. (2010). Beneficial effects of Bacteroides fragilis
polysaccharides on the immune system. Frontiers in bioscience: a journal and
virtual library, 15, 25.
140. Conti, H. R., & Gaffen, S. L. (2010). Host responses to Candida albicans: Th17
cells and mucosal candidiasis. Microbes and Infection, 12(7), 518-527.
141. Shao, T. Y., Ang, W. G., Jiang, T. T., Huang, F. S., Andersen, H., Kinder, J.
M., ... & Hershey, G. K. K. (2019). Commensal Candida albicans positively
calibrates systemic Th17 immunological responses. Cell host & microbe, 25(3),
404-417.
142. Schimke, L. F., Sawalle-Belohradsky, J., Roesler, J., Wollenberg, A., Rack,
A., Borte, M., ... & Reichenbach, J. (2010). Diagnostic approach to the hyper-IgE
syndromes: immunologic and clinical key findings to differentiate hyper-IgE
syndromes from atopic dermatitis. Journal of allergy and clinical immunology,
126(3), 611-617.
143. Wei, L., Laurence, A., Elias, K. M., & O'Shea, J. J. (2007). IL-21 is produced
by Th17 cells and drives IL-17 production in a STAT3-dependent manner. Journal
of biological chemistry, 282(48), 34605-34610.
144. Liang, S. C., Long, A. J., Bennett, F., Whitters, M. J., Karim, R., Collins,
M., ... & Fouser, L. A. (2007). An IL-17F/A heterodimer protein is produced
by mouse Th17 cells and induces airway neutrophil recruitment. The Journal
of Immunology, 179(11), 7791-7799.
145. McGeachy, M. J. (2011). GM-CSF: the secret weapon in the T H 17 arsenal.
72

Nature immunology, 12(6), 521.
146. McGinley, A. M., Edwards, S. C., Raverdeau, M., & Mills, K. H. (2018). Th17
cells, γδ T cells and their interplay in EAE and multiple sclerosis. Journal of
autoimmunity, 87, 97-108.
147. Butcher, M. J., & Zhu, J. J. (2018). Plasticity and functionality of Th17 subsets in
EAE.
148. Aranami, T., & Yamamura, T. (2008). Th17 Cells and autoimmune
encephalomyelitis (EAE/MS). Allergology International, 57(2), 115-120.
149. Kugyelka, R., Kohl, Z., Olasz, K., Mikecz, K., Rauch, T. A., Glant, T. T., &
Boldizsar, F. (2016). Enigma of IL-17 and Th17 cells in rheumatoid arthritis and in
autoimmune animal models of arthritis. Mediators of inflammation, 2016.
150. Boniface, K., Moynet, D., & Mossalayi, M. D. (2013). Role of Th17 cells in
the pathogenesis of rheumatoid arthritis. World Journal of Rheumatology, 3(3), 25-
31.
151. Azizi, G., Jadidi‐Niaragh, F., & Mirshafiey, A. (2013). Th17 Cells in
Immunopathogenesis and treatment of rheumatoid arthritis. International journal of
rheumatic diseases, 16(3), 243-253.
152. Warfel, J. M., & Merkel, T. J. (2013). Bordetella pertussis infection induces a
mucosal IL-17 response and long-lived Th17 and Th1 immune memory cells in
nonhuman primates. Mucosal immunology, 6(4), 787- 796.
153. Gates, H. A., Smith, A. A., Jakoush, T. A., Silva, E., You, H., Wu, V., ... &
Segal-Abramson, T. (2017). Regulatory T cells facilitate Th17 responses to
Bordetella pertussis and aid in resolution of whooping cough in mice.
73

154. Kempski, J., Brockmann, L., Gagliani, N., & Huber, S. (2017). TH17 cell and
epithelial cell crosstalk during inflammatory bowel disease and carcinogenesis.
Frontiers in immunology, 8, 1373.
155. Hou, G., & Bishu, S. (2020). Th17 Cells in Inflammatory Bowel Disease: An
Update for the Clinician. Inflammatory Bowel Diseases.
156. Hamilton, J. A. (2015). GM-CSF as a target in inflammatory/autoimmune
disease: current evidence and future therapeutic potential. Expert review of clinical
immunology, 11(4), 457-465.
157. Maddur, M. S., Miossec, P., Kaveri, S. V., & Bayry, J. (2012). Th17 cells:
biology, pathogenesis of autoimmune and inflammatory diseases, and
therapeutic strategies. The American journal of pathology, 181(1), 8-18.
158. Liu, M., Hu, X., Wang, Y., Chen, X., & Wu, J. (2014). Association of IL-23
and its receptor gene single- nucleotide polymorphisms with multiple sclerosis in
Chinese southern population. International Journal of Neuroscience, 124(12), 904-
907.
159. Shahrokhi, V. M., Ravari, A., Mirzaei, T., Zare-Bidaki, M., Asadikaram,
G., & Arababadi, M. K. (2018). IL-17A and IL-23: plausible risk factors to
induce age-associated inflammation in Alzheimer's disease. Immunological
Investigations, 47(8), 812-822.
160. Saresella, M., Marventano, I., Calabrese, E., Piancone, F., Rainone, V., Gatti,
A., ... & Clerici, M. (2014). A complex proinflammatory role for peripheral
monocytes in Alzheimer's disease. Journal of Alzheimer's Disease, 38(2), 403-413.
161. Alhaidary, A. F., Adhia, A. H., Melconian, A. K., & Hamdan, A. A. (2014). The
74

level of IL-1α, IL-10 and IL-17A in Alzheimer's disease patients: Comparative
study. Baghdad Science Journal, 11(4), 1486-1492.
162. Saresella, M., Calabrese, E., Marventano, I., Piancone, F., Gatti, A., Alberoni, M., ...
& Clerici, M. (2011). Increased activity of Th-17 and Th-9 lymphocytes and a
skewing of the post-thymic differentiation pathway are seen in Alzheimer’s disease.
Brain, behavior, and immunity, 25(3), 539-547.

75

Chapter 2 Neural stem cell immune tolerance in graft versus host
disease

Human leukocyte antigen-A neural transplant tolerance in
humanized mice
1
Kwok W. Im,
1,2
Kevin R. Doty,
1,3
Juan Biancotti,
1,4
David Gate,
1,5
Brian P. Leung,
6
George Y. Liu and

1
Terrence Town*

1
Zilkha Neurogenetic Institute, Department of Physiology & Neuroscience, Keck
School of Medicine of the University of Southern California,1501 San Pablo Street,
Los Angeles, CA 90089-2821, USA
2
Present address: Amgen, One Amgen Center Drive Thousand Oaks, CA 91320-1799, USA

3
Present address: Johns Hopkins University, Department of Surgery, School of
Medicine, 1721 E. Madison Ross Bldg., Room 733 Baltimore, MD 21205, USA
4
Present address: Stanford Neurosciences Institute, Department of Neurology and
Neurological Sciences, Stanford University School of Medicine, 1201 Welch Road,
MSLS Building, Room P205, Stanford, CA 94305- 5489, USA
76

5
Present address: Hong Kong University of Science and Technology, Clear Water Bay,
Kowloon, Hong Kong, China
6
Cedars-Sinai Medical Center, Department of Biomedical Sciences, Infectious &
Immunological Diseases Research Center, 8700 Beverly Blvd. Davis Building, Room
4094G Los Angeles, CA 90048, USA

*Correspondence: ttown@usc.edu
77

2.1 Abstract

Transplantation of non-matched fetal brain tissue can successfully treat Parkinson’s
disease. Yet, lack of animal models to study human immune responses to neural stem
cell transplants has broadly limited bench to bedside translation for neurological
disorders. Here, we report a ‘humanized’ mouse model to investigate neural transplant
tolerance dependent on human leukocyte antigens (HLAs). We demonstrate that a
single allelic match at HLA- A−irrespective of HLA-B, -C, -DR, or -DQ
haplotype−promotes immune tolerance to transplanted human neural progenitors.
Accordingly, human genes driving neural development and immune tolerance are
enriched at the transplant site, supporting that partial HLA-A matching promotes neural
progenitor engraftment. Together, our results establish humanized mice as a pre-
clinical tool to assess human neural transplant therapy.

2.2 Introduction
Recent advances in human embryonic stem cell (hESC) transplantation have raised
the potential for this emerging treatment. Studies in animals and humans are addressing
clinical concerns including functional differentiation of stem cell pools, tumorigenesis,
and immunological rejection
1–3
. Furthermore, contemporary improvements in
stabilization, purification, and establishment of diverse hESC banks give good reason
to continue clinical development of stem cell-based therapies
4–6
.
78

One of the most studied hESC therapies is transplantation of non-matched fetal brain
tissue into Parkinson’s disease patients
7–10
. To promote allograft acceptance,
immunosuppressive drugs such as FK506 or Cyclosporin A
11,12
are prescribed. While
generally thought to reduce acute transplant rejection, immunosuppressive drug
dosing and duration is not well-understood
13
, and chronic immunosuppression can lead
to adverse events such as infection and malignancy
14,15
. One approach to moderate
immunosuppressive therapy by attenuating rejection is matching donor to recipient
tissues. Human leukocyte antigens (HLAs) comprise human major histocompatibility
complex (MHC); an important set of histocompatibility molecules. HLA-A, -B, and -
C encode class I, while -DR and -DQ are the main class II loci. HLAs are highly
polymorphic: there are > 12,200 HLA alleles, with > 9,200 variants in HLA class I
alone
16,17
. HLA alleles are in linkage disequilibrium, and are co-inherited as
haplotypes
18
. Large-scale HLA diversity necessitates banking hESCs from donors with
plethora HLA haplotypes to afford greatest chances of donor-recipient
histocompatibility. Yet, chance for a partial HLA match from 150 donors is only
18.5%, and more complete matching is increasingly difficult with only incremental
benefit
19
. But how many degrees of freedom from a complete HLA match are allowed
for transplant acceptance?

Immune system differences between species make it difficult to model transplanting
human cells into mice. ‘Humanized’ mice that replace mouse endogenous immunity
with human adaptive immune systems are emerging to address this
20–22
. We generated
humanized mice to investigate immune responses to human neural progenitor cell
79

(hNPC) transplants. Specifically, we HLA haplotyped and differentiated hNPCs for
transplantation into NOD.Cg-Prkdc
scid
Il2rg
tm1Wjl
/SzJ (NSG) mice reconstituted with
CD34
+
human hematopoietic stem cells (hHSCs). Humanized mice demonstrated
robust hHSC proliferation and differentiation into adaptive immune cells. Further,
HLA-A was selectively expressed by hNPCs. We conducted HLA haplotype ‘mix and
match’ adoptive transfer assays by transplanting hNPCs into the brain. Cerebral
immune responses were significantly reduced in humanized mice transplanted with
hNPCs that were partially matched at HLA-A. Finally, partial HLA- A matching altered
the transcriptional landscape at the transplant site consistent with neural engraftment
and immune tolerance. In summary, we report using humanized mice to evaluate
neural transplantation therapy.

2.3 Results

2.3.1 Establishing and validating mice with human adaptive immune systems
Human HSC transplantation into neonatal immunodeficient NSG mice increases
thymic output and survival
1
. To generate mouse models with human immune systems
(Fig. 1a), we utilized NSG mice bearing null alleles of the Il-2 receptor common
gamma chain and the scid mutation in the DNA repair complex protein Prkdc,
rendering them B, NK, and T cell deficient
23
. Newly born NSG pups were irradiated
to deplete endogenous mouse HSCs, thereby creating space for hHSC engraftment.
80

CD34
+
hHSCs were isolated from umbilical cord blood, sorted by flow cytometry until
purity was ≥ 95% and CD3
+
T cells were ≤ 0.1%, and injected into the liver
24,25
(Fig.
1a and b). Human leukocytes organized into anatomically distinct white pulp and
marginal zones in the spleen (data not shown). Five-to-six months after hHSC
transplantation, we assayed for the pan-leukocyte marker, CD45. Enumerating
percentages of mouse (m) vs. human (h) CD45 disclosed robust splenic engraftment
of hCD45
+
immune cells (Fig. 1c and d, ***p < 0.001; n = 16). Evidencing hHSC
differentiation, lymph nodes from reconstituted NSG mice contained CD45
+
, CD3
+
, T
cell receptor alpha/beta (TcRα/β
+
), CD4
+
and CD8
+
human T-cells (Fig. 1e and f, ***p
< 0.001; n = 16). These results demonstrate successful reconstitution of NSG mice
with human adaptive immune cells.

2.3.2 Selective HLA-A expression by neural progenitors
HLA expression by neural stem cells is not well-understood, although studies suggest
that HLAs may impact immune tolerance to hESC-derived progenitors
26,27
. Therefore,
we analyzed HLA expression profiles in hESC- derived hNPCs. Human ESCs
typically matured to NPCs as evidenced by Sox2, Tuj1 and Nestin proteins (Fig. 2a).
To assess HLA expression by hNPCs, we developed a co-linear curve (CoLC) analysis
strategy. CoLC tests deviation from a line, such that a non-significant p value indicates
linearity and therefore, a positive result (Table 1). Analyses were performed for
hNPCs compared to human microglia (positive control) using the GAPDH
housekeeping gene as a reference (Fig. 2b). CoLC revealed statistical significance for
HLA-B, -C, -DR, and -DQ, indicating non-linearity and therefore lack of gene
81

expression. However, a similar analysis of HLA-A in NPCs was non-significant,
indicating expression (Fig. 2b). Together, these results demonstrate that HLA-A is the
predominant isotype expressed by hNPCs.

2.3.3 Partial HLA-A matching immunologically tolerizes neural transplants
Histocompatibility mismatch between donor and recipient drives alloimmune
reactions that can lead to graft-vs- host disease (GVHD)
28
. Because HLA-A was
selectively expressed by hNPCs, we sought to examine its role in neural
transplantation using humanized mice. To test this, we ‘mix and matched’ HLA
halpotypes between transplanted hNPCs and humanized mouse immune systems (Fig.
3a). Human immune responses to hNPC engraftment were tested by immunostaining
brain sections through the transplant site with hCD45, two weeks after hNPC
transplantation. Fig. 3b demonstrates reduced brain-infiltrating human leukocytes that
are partially (one allele; 50% vs. fully mis-matched (no alleles; 0%) at HLA-A.
Blockade of infiltrating leukocytes were complete, because partial matching at HLA-
A did not differ from controls that received phosphate buffered saline (PBS) injections
in lieu of hNPCs (Fig. 3c, *p<0.05, ***p<0.001; PBS controls, n=5; 0% match, n=10;
50% match, n=10).
To assess mouse innate immune responses to neural progenitor grafts, we
immunostained for ionized calcium- binding receptor 1 (Iba1; a structural marker of
reactive brain macrophages) that showed reduced reactivity in 50% vs. 0% HLA-A
matches (Fig. 3d, e, *p < 0.05, ***p < 0.001). Immunoreactivity for cleaved (active)
82

caspase 3 (casp-3), a marker of apoptosis, was reduced at the transplant site when
comparing partially to fully HLA-Amis-matched (Fig. 3f, g, ** p < 0.01). Taken
together, these results suggest that partial matching at HLA-A promotes immune
tolerance to human neural transplants.

2.3.4 HLA-A impacts neuroinflammatory landscape after neural transplantation
Examining gene expression changes after transplantation has aided our
understanding of intermediate GVHD phenotypes
29
. To investigate impact of HLA-A
matching on the transcriptome after neural transplantation, we isolated the hNPC
transplant site two weeks after engraftment and performed RNA sequencing. Because
the transplant site contains both mouse and human DNA, we developed a technique to
separate transcripts based on inter-species sequence variability. From a total of 5923
differentially expressed genes, we observed robust changes when comparing either
complete mismatch at HLA-A vs. PBS injected controls or 50% to 0% HLA-A matches.
Strikingly, only modest transcriptome changes were observed between the 50%
matched group and PBS controls (Fig. 4a), suggesting brain immune tolerance to
transplanted hNPCs. This was supported by a similar pattern of gene expression
profiles in hierarchical clustered heat maps from the 50% HLA-A match group vs. PBS
controls (Fig. 4b). Together with histological findings at the transplant site, these
global transcriptome changes show that partial HLA-A matching promotes immune
tolerance to transplanted human neural progenitors.

83

To further test this, we conducted Gene Set Expression Analysis (GSEA) and
Ingenuity Pathway Analysis (IPA) to profile immune and inflammatory transcripts at
the hNPC transplant site. Remarkably, we observed significant gene enrichment for
tolerance induction, T cell selection, and antigen processing and presentation in 50%
vs. 0% HLA-A match conditions (Fig. 4c, d). IPA revealed similar pathway changes
for stem cell development, and robust downregulation of inflammatory and
tumorigenic pathways in 50% vs. 0% HLA-A matches (Fig. 4e). Together, these results
show that matching a single allele at the HLA-A locus, irrespective of HLA-B, -C, -DR,
or -DQ, promotes human immune tolerance to neural transplants.

2.4 Discussion

Histocompatibility by HLA matching between donors and recipients is a widely
recognized determinant of transplant success. With recent advances in immune
suppressors (e.g., cyclosporin, FK506 and methotrexate), HLA mismatched transplants
have become more common in diabetes, ischemic heart failure, leukemia,
cytomegalovirus hepatitis, and Parkinson’s disease
9,30–33
. However, the extent to
which HLA matching leads to successful clinical outcomes depends on tissue-specific
microenvironment and disease
34
. Furthermore, the contribution of HLA matching to
neural transplant tolerance remains unclear and represents a knowledge gap in the
factors responsible for CNS engraftment. This problem has been challenging to
address, because modeling the human immune system in a rodent is difficult. In this
report, we generated mice bearing human immune systems to examine the role of
84

HLAs in neural transplantation. Consistent with previous reports of up to 60-80%
engraftment of human immune cells
35,36
, we successfully reconstituted NSG mice with
purified CD34
+
human cord blood cells that proliferated and differentiated into human
CD4
+
and CD8
+
T lymphocytes that matured into CD3
+
TCRαβ
+
T-cells.
Transplant donor selection is typically based on HLA class I (HLA-A, -B, -C) and II
(HLA-DRB1 and -DQB1), requiring matching at least 8 of 10 antigens for a ‘go’
decision
37
. But which HLAs are necessary for allograft acceptance, and how many
degrees of freedom from a perfect match are permissible for immune tolerance?
Previous studies implicate HLA class I in allograft acceptance
38
, and we found that
hNPCs selectively expressed HLA-A. This result suggested that HLA-A may be the key
HLA locus for allograft acceptance. Indeed, immunogenicity to 50% HLA-A matched
hNPCs was reduced both in terms of human CD45
+
leukocytes and mouse Iba1
+
brain
macrophages at the transplant site. Additionally, 50% HLA-A matching supported
hNPC survival as evidenced by diminished active casp-3 (an apoptosis marker) at the
transplant site. Strikingly, the transcriptomic landscape at the neural transplant site
was comparable between 50% HLA-A match and PBS control. Specifically, gene
ontology enrichment revealed lack of immune responses (including antigen
presentation and leukocyte activation) in both groups. As a positive control, GSEA
and IPA analytic methods revealed differential expression of genes involved in
leukocyte extravasation and PI3K signaling when comparing 50% to 0% HLA-A
matches. Deeper analysis demonstrated robust downregulation of glioblastoma
multiforme, interleukin-8, mTOR, and glycolysis signaling in 50% vs. 0% HLA-A
matches, consistent with other reports on immune tolerance to allografts
39,40
. Although
85

our current findings suggest that 50% HLA-A matching confers immune tolerance, it
is noteworthy that these changes occur acutely after hNPC transplantation. It has
been reported that mismatched alleles other than HLA-A can lead to GVHD in long-
term transplants, increasing mortality
34,41
. Future studies investigating the long-term
effects of mismatching at other HLA loci on neural transplants will be important to
understand chronic allograft acceptance.
2.5 Conclusion

For the first time, this study reports a mouse model to investigate human immune
responses to human neural transplantation. Together, our results suggest that 50%
matching at the HLA-A locus critically determines the immune response to neural
transplants, suggesting that humanized mice are an important tool to assess HLA-
dependent neural transplant tolerance. Understanding human immune responses to
transplantation is vital, and our results offer a pre-clinical model system to do this.

HLA expression

hNPCs microglia F p value
HLA-A 0.231 0.221 0.011 n.s
HLA-B 0.241 0.126 19.66 0.004
HLA-C 1.384 0.575 1.59E+07 0.0001
HLA-DR 2.008 0.442 9.71E+07 0.0001
HLA-DQ 0.46 0.176 9.44E+07 0.0001
Table 1: Figures obtained from co-linear curve (CoLC) analysis.
Values under the hNPCs and microglia columns are line slopes.

86

87

Figure 2: Human neural progenitors selectively express HLA-A. (a) Representative photomicrographs of human
neural progenitor cells (hNPCs) expressing Sox2 (green), Tuj1 (red), and Nestin (red; DAPI in blue). Scale bars denote
100 µm. (b) Co-linear curve analysis of HLA expression by qPCR in hNPCs (red dotted line) vs. human microglial
cells (positive control for all HLAs; blue dotted line). All samples were normalized to GAPDH (black dotted line) (n
= 6, ***p < 0.001, n.s = not significant).

88

89

Figure 3: Partial HLA-A matching tolerizes against human neural transplant rejection. (a) Mix and match
haplotype assay. Human hematopoietic stem cells (HSCs) were haplotyped and engrafted for 5-6 months to
generate humanized mice. Human NPCs were haplotyped and unilaterally transplanted into the right striatum
for two weeks. (b) Representative images of infiltrating human immune cells at the transplant site in 50% vs. 0%
HLA-A matched groups using the human-specific pan-leukocyte marker, hCD45 (red). Scale bars denote 50 μm.
(c) Quantitation of % hCD45 immunolabeled area. (d) Representative images of immunoreactive brain
macrophages (Iba1, purple) at the hNPC transplant site in the right striatum (marked by HuNu, green). Scale
bars denote 50 μm. (e) Quantitation of % Iba1 immunolabeled area. (f) Representative images of cleaved (active)
caspase-3 (casp-3, red) immune-reactivity at the hNPC transplant site (HuNu, green). Scale bars denote 30 μm.
(g) Quantitation of % casp-3 immunolabeled area. Data are means ± SEMs; **p<0.01, ***p<0.001; PBS, n = 5;
50%, n = 10, 0%, n = 10; n.s. = not significant; n.d. = not detectable.
90

91

References
1. Myers, L. A., Patel, D. D., Puck, J. M. & Buckley, R. H. Hematopoietic
stem cell transplantation for severe combined immunodeficiency in the
neonatal period leads to superior thymic output and improved survival.
Blood 99, 872–878 (2002).
2. Quintana, E. et al. Efficient tumor formation by single human
melanoma cells. Nature 456, 593– 598 (2008).
3. Piccirillo, S. G. M. et al. Bone morphogenetic proteins inhibit the
tumorigenic potential of human brain tumor-initiating cells. Nature 444,
761–765 (2006).
4. Steinbeck, J. A. et al. Optogenetics enables functional analysis of
human embryonic stem cell derived grafts in a Parkinson’s disease
model. Nat. Biotechnol. 33, (2015).
5. Kriks, S. et al. Dopamine neurons derived from human ES cells
efficiently engraft in animal models of Parkinson’s disease. Nature
(2011). doi:10.1038/nature10648
6. Mayor, N. P. et al. HLA Typing for the Next
Generation. PLoS One (2015).
doi:10.1371/journal.pone.0127153
7. Kordower, J. H., Chu, Y., Hauser, R. A., Freeman, T. B. &
Olanow, C. W. Lewy body-like pathology in long-term embryonic
nigral transplants in Parkinson’s disease. Nat. Med. (2008).
doi:10.1038/nm1747
92

8. Kordower, J. H. et al. Neuropathological evidence of graft survival and striatal
reinnervation after
the transplantation of fetal mesencephalic tissue in a patient with
Parkinson’s disease. The New England journal of medicine (1995).
doi:10.1056/NEJM199504273321702
9. Li, J.-Y. et al. Lewy bodies in grafted neurons in subjects with
Parkinson’s disease suggest host to-graft disease propagation. Nat.
Med. (2008). doi:10.1038/nm1746
10. Lindvall, O. Developing dopaminergic cell therapy for
Parkinson’s disease-give up or move forward? Movement Disorders
(2013). doi:10.1002/mds.25378
11. Lindvall, O. & Kokaia, Z. Stem cells in human neurodegenerative
disorders - Time for clinical translation? Journal of Clinical
Investigation (2010). doi:10.1172/JCI40543
12. Li, D. W. & Duncan, I. D. The immune status of the myelin deficient rat
and its immune responses to transplanted allogeneic glial cells. J.
Neuroimmunol. (1998). doi:10.1016/S0165
5728(98)00006-X
13. Ajithkumar, T. V., Parkinson, C. A., Butler, A. & Hatcher, H. M.
Management of solid tumours in organ-transplant recipients. Lancet
Oncology (2007). doi:10.1016/S1470-2045(07)70315-7
14. Moss, P. & Rickinson, A. Cellular immunotherapy for viral infection after HSC
transplantation. Nat. Rev. Immunol. (2005). doi:10.1038/nri1526
93

15. Amariglio, N. et al. Donor-derived brain tumor following neural
stem cell transplantation in an ataxia telangiectasia patient. PLoS Med.
(2009). doi:10.1371/journal.pmed.1000029
16. Mittal, K. K. The HLA polymorphism and susceptibility to disease. Vox Sang. (1976).
17. Hughes, A. L. & Yeager, M. Natural selection at major
histocompatibility complex loci of vertebrates. Annu. Rev. Genet.
(1998). doi:10.1146/annurev.genet.32.1.415
18. Cao, K. et al. Analysis of the frequencies of HLA-A,
B, and C alleles and haplotypes in the five major ethnic
groups of the United States reveals high levels of
diversity in these loci and contrasting distribution
patterns in these populations. in Human Immunology
(2001). doi:10.1016/S0198-8859(01)00298-1
19. Taylor, C. J. et al. Banking on human embryonic stem cells:
Estimating the number of donor cell lines needed for HLA matching.
Lancet (2005). doi:10.1016/S0140-6736(05)67813-0
20. Lan, P., Tonomura, N., Shimizu, A., Wang, S. & Yang, Y. G.
Reconstitution of a functional human immune system in immunodeficient
mice through combined human fetal thymus/liver and CD34+
cell transplantation. Blood (2006). doi:10.1182/blood-2005-11-4388
21. Tonomura, N., Habiro, K., Shimizu, A., Sykes, M. & Yang, Y.-G.
Antigen-specific human T-cell responses and T cell-dependent production
of human antibodies in a humanized mouse model. Blood (2008).
94

doi:10.1182/blood-2007-11-121319
22. Rajesh, D. et al. Th1 and Th17 immunocompetence in humanized
NOD/SCID/IL2r??null mice. Hum. Immunol. (2010). doi:10.1016/j.humimm.2010.02.019
23. Shultz, L. D. et al. Human Lymphoid and Myeloid Cell Development in
NOD/LtSz-scid IL2R null Mice Engrafted with Mobilized Human
Hemopoietic Stem Cells. J. Immunol. 174, 6477–6489 (2005).
24. Traggiai, E. et al. Development of a human adaptive immune
system in cord blood cell transplanted mice. Science 304, 104–
107 (2004).
25. Nikolic, W. V et al. Peripherally administered human umbilical cord blood cells reduce
parenchymal and vascular beta-amyloid deposits in Alzheimer mice.
Stem Cells Dev. 17, 423– 439 (2008).
26. Drukker, M. et al. Characterization of the expression of MHC proteins
in human embryonic stem cells. Proc. Natl. Acad. Sci. U. S. A. (2002).
doi:10.1073/pnas.142298299
27. Grinnemo, K.-H. et al. Human embryonic stem cells are
immunogenic in allogeneic and xenogeneic settings. Reprod.
Biomed. Online 13, 712–724 (2006).
28. Petersdorf, E. W., Malkki, M., Gooley, T. A., Martin, P. J. & Guo, Z.
MHC haplotype matching for unrelated hematopoietic cell transplantation.
PLoS Med. 4, 0059–0068 (2007).
29. Vandiedonck, C. et al. human major histocompatibility complex
Pervasive haplotypic variation in the spliceo-transcriptome of the human
95

major histocompatibility complex. Genome Res. 21, 1042–1054 (2011).
30. Scharp, D. W. et al. Insulin independence after islet transplantation into type I diabetic
patient. Diabetes 39, 515–518 (1990).
31. Butts, R. J. et al. Association of human leukocyte antigen donor-
recipient matching and pediatric heart transplant graft survival. Circ.
Hear. Fail. 7, 605–611 (2014).
32. Szydlo, R. et al. Results of allogeneic bone marrow transplants for
leukemia using donors other than HLA-identical siblings. J. Clin.
Oncol. 15, 1767–1777 (1997).
33. Mañez, R. et al. the Influence of Hla Matching on
Cytomegalovirus Hepatitis and Chronic Rejection After Liver
Transplantation. Transplantation 55, 1067–1071 (1993).
34. Kanda, J. Effect of HLA mismatch on acute graft-versus-host
disease. Int. J. Hematol. 98, 300– 308 (2013).
35. Audigé, A. et al. Long-term leukocyte reconstitution in NSG mice transplanted with
human cord blood hematopoietic stem and progenitor cells. BMC Immunol. 18, 28 (2017).
36. Misharin, A. V et al. Development of a new humanized mouse model to
study acute inflammatory arthritis. J. Transl. Med. 10, 190 (2012).
37. Jagasia, M. et al. Risk factors for acute GVHD and
survival after hematopoietic cell transplantation. Blood 119,
296–307 (2012).
38. Garrido, F. et al. Natural history of HLA expression during tumour
development. Immunol. Today 14, 491–499 (1993).
96

39. Sibbring, J. S., Sharma, A., McDicken, I. W., Sells, R. A. & Christmas,
S. E. Localization of C-X C and C-C chemokines to renal tubular
epithelial cells in human kidney transplants is not confined
to acute cellular rejection. Transpl. Immunol. 6, 203–208 (1998).
40. Lee, C. F. et al. Preventing Allograft Rejection by Targeting Immune
Metabolism. Cell Rep. 13, 760–770 (2015).
41. Woolfrey, A. et al. HLA-C Antigen Mismatch Is Associated with
Worse Outcome in Unrelated Donor Peripheral Blood Stem Cell
Transplantation. Biol. Blood Marrow Transplant. 17, 885–892 (2011).
42. Kim, D., Pertea, G., Trapnell, C. et al. TopHat2: accurate
alignment of transcriptomes in the presence of insertions, deletions
and gene fusions. Genome Biol 14, R36 (2013) doi:10.1186/gb 2013-
14-4-r36
43. Trapnell, C., Roberts, A., Goff, L. et al. Differential gene and transcript expression
analysis of RNA seq experiments with TopHat and Cufflinks. Nat Protoc 7, 562–578
(2012)
doi:10.1038/nprot.2012.016
44. Gentleman, R.C., Carey, V.J., Bates, D.M. et al. Bioconductor: open
software development for computational biology and bioinformatics.
Genome Biol 5, R80 (2004) doi:10.1186/gb-2004-5 10-r80

97

Chapter 3 Adaptive immunity in Brain Cancer
Short-lived effector T cells in a mouse model of
medulloblastoma
3.1 Introduction

Each year 4,200 more children (11.5 each day) are diagnosed with a pediatric brain
tumor in the USA [2]. Pediatric brain tumors are the most common solid brain tumors
in children and are the leading cause of cancer- related deaths in children 18 and under
[3]. Even with the advent of modern therapeutic approaches, the treatment itself,
including surgery, radiation and chemotherapy, carries great risk of brain injury to the
developing brain with long-term neurological, intellectual, hormonal, and
psychological damage [4-7]. For those children less than 3 years of age or with
recurrent disease, the outlook for long-term survival is grim [8, 9]. These inherent risks
necessitate the need for alternative therapies.
Pediatric brain cancer is different from adult forms of the disease. Firstly, there are
differences in the underlying genetic etiology. For example, most pediatric gliomas do
not have the oncogene p53 as a contributing factor [10]. In fact, recent studies indicate
that a unique set of recurrent mutations are found in pediatric glioma but not in adult
forms of the disease [11, 12]. While gliomas are represented by a spectrum of
pathologies that range from low- to high-grade in character and behavior, most
pediatric brain tumors are low-grade tumors, as opposed to adult glioblastoma [13,
14]. Thus, a large fraction of pediatric gliomas results from stepwise transformation
98

of lower grade tumors [14]. Finally, while considerable focus has been given to adult
forms of glioma, many types of pediatric brain cancer are, by comparison,
understudied. Importantly, the underlying differences between pediatric and adult
CNS tumors often mean that successful adult glioma treatments are ineffective in
children [15, 16]. One of the main anti-cancer approaches has been to target the
immune system [17]. The concept is to increase the body’s own response to the tumor
itself. In this regard, the cardinal immunosuppressive molecule, TGF-β, remains one
of the most well-studied pathways in cancer biology [18, 19]. Yet, the pleiotropic
functions of TGF- β signaling in almost all cell types have obscured its role in tumor
development and have made cell-specific therapeutic approaches difficult. One of the
main immune cells affected by TGF-β signaling is the CD8 T cell, which is responsible
for long-lasting anti-tumor responses [1, 20]. This critical antitumor role of CD8 T
cells in pediatric brain cancers is an under investigated area, and we are uniquely
positioned to fill this knowledge gap.

In order to test the CD8 T cell-specific contribution of TGF-β signaling to
development of pediatric brain cancer, an innovative approach will be utilized. The
model is a germline-encoded mutant SmoA1 transgenic MB mouse that will be treated
with next-generation nanoparticles to specifically block TGF-β signaling in CD8 T
cells. The SmoA1 transgenic mouse is one of the only published and validated models
of pediatric brain cancer [25], and we have experience working with these animals for
the past 7 years [1]. Therefore, a second innovative element is pharmacological
targeting of the peripheral adaptive immune system to bring about tumor therapeutic
99

effects within the CNS. Importantly, this approach circumvents the issue of blood-
brain-barrier penetrance of an anti- tumor drug or biological. The third innovative
aspect centers on a novel nanoparticle-based approach with high translational
potential. Specifically, we have developed a method of using PEG-PLGA
nanoparticles encapsulating the small molecule TGF-β-SMAD 2/3 signaling inhibitor
(SB-505124). Over the past few decades, biodegradable polyesters, such as PLGA,
have been extensively studied for a wide variety of pharmaceutical and biomedical
applications. The biodegradable polyester family has been regarded as one of the few
synthetic biodegradable polymers with controllable biodegradability, excellent
biocompatibility, and a high degree of safety. The need for a variety of drug
formulations for different drugs and delivery pathways has resulted in development of
various types of block copolymers consisting of biodegradable polyesters and PEG.
This work will utilize the unique properties of PEG-PLGA block copolymers to target
TGF-β signaling in CD8 T cells. Based on over 20 years of favorable safety data in
the clinic, the PEG-PLGA formulation has very high translational potential. Therefore,
we believe this to be a significant value-added to our proposal.

Significance

The novel therapeutic strategy outlined in this study has significance for both basic
science and cancer immunotherapy. Our data suggest that co-treatment with SB and
RGD leads to synergistically reduced TGF-β signaling at the tumor as compared with
SB treatment alone. A key advantage of our nanoparticle co-treatment approach with
SB/RGD is that pharmacologic manipulation of TGF-β production/signaling occurs
100

both at the intra- and extra-cellular levels. Moreover, the next-generation nanoparticles
used for this study represent a novel therapeutic and preventative modality for tumor
formation, as they can selectively and efficiently deliver SB- 505124 to tumor cells
with integrin αvβ3-inhibitor RGD via a minimally invasive peripheral route and are
backed by over 25 years of clinical safety data.

References

1. Gate D., et al. T-cell TGF-beta signaling abrogation restricts medulloblastoma
progression. Proc Natl Acad Sci U S A. 2014;111(33):E3458-66.
2. 2012 CBTRUS Statistical Report: Primary Brain and Central Nervous System Tumors
Diagnosed in the United States in 2004-2008, 2012.
3. Ries, L.A.G. and SEER Program (National Cancer Institute (U.S.)), Cancer incidence
and survival among children and adolescents : United States SEER program 1975-
1995 /[edited by Lynn A. Gloecker Ries ... et al.]. SEER pediatric monograph1999,
Bethesda, MD: National Cancer Institute, SEER Program. vi, 182 p.
4. Lannering, B., et al., Long-term sequelae after pediatric brain tumors: Their effect on
disability and quality of life. Med Pediatr Oncol, 1990. 18(4): p. 304-310.
5. Carey, M.E., et al., Neuropsychological functioning and social functioning of
survivors of pediatric brain tumors: evidence of nonverbal learning disability. Child
Neuropsychol, 2001. 7(4): p. 265-72.
6. Nagel, B.J., et al., Early patterns of verbal memory impairment in children treated for
101

medulloblastoma. Neuropsychology, 2006. 20(1): p. 105-12.

7. Macedoni-Luksic, M., B. Jereb, and L. Todorovski, Long-term sequelae in children
treated for brain tumors: impairments, disability, and handicap. Pediatr Hematol
Oncol, 2003. 20(2): p. 89-101.
8. Tomlinson, F.H., et al., Medulloblastoma: I. Clinical, diagnostic, and therapeutic
overview. J Child Neurol, 1992. 7(2): p. 142-55.
9. Gajjar, A. and B. Pizer, Role of High-Dose Chemotherapy for Recurrent
Medulloblastoma and Other CNS Primitive Neuroectodermal Tumors. Pediatr
Blood Cancer, 2010. 54(4): p. 649-651.
10. Felix, C.A., et al., p53 gene mutations in pediatric brain tumors. Med Pediatr Oncol,
1995. 25(6): p. 431-6.
11. Schwartzentruber, J., et al., Driver mutations in histone H3.3 and chromatin
remodelling genes in paediatric glioblastoma. Nature, 2012. 482(7384): p. 226-31.
12. Wu, G., et al., Somatic histone H3 alterations in pediatric diffuse intrinsic
pontine gliomas and nonbrainstem glioblastomas. Nat Genet, 2012. 44(3): p. 251-3.
13. Riemenschneider, M.J. and G. Reifenberger, Molecular neuropathology of low-
grade gliomas and its clinical impact Low-Grade Gliomas, J. Schramm, Editor
2010, Springer Vienna. p. 35-64.
14. Broniscer, A., et al., Clinical and molecular characteristics of malignant
transformation of low-grade glioma in children. J Clin Oncol, 2007. 25(6): p. 682-
9.
15. Hargrave, D.R. and S. Zacharoulis, Pediatric CNS tumors: current treatment and
102

future directions. Exp Rev Neurother, 2007. 7(8): p. 1029-1042.
16. Siegel, M.J., J.L. Finlay, and S. Zacharoulis, State of the art chemotherapeutic
management of pediatric brain tumors. Exp Rev Neurother, 2006. 6(5): p. 765-779.
17. Wojtowicz-Praga, S., Reversal of tumor-induced immunosuppression: a new
approach to cancer therapy. J Immunother, 1997. 20(3): p. 165-77.
18. Weller, M. and A. Fontana, The failure of current immunotherapy for malignant
glioma. Tumor-derived TGF- beta, T-cell apoptosis, and the immune privilege of
the brain. Brain Res Brain Res Rev, 1995. 21(2): p. 128-51.
19. Platten, M., W. Wick, and M. Weller, Malignant glioma biology: role for TGF-
beta in growth, motility, angiogenesis, and immune escape. Microsc Res Tech,
2001. 52(4): p. 401-10.
20. Gorelik, L. and R.A. Flavell, Immune-mediated eradication of tumors through
the blockade of transforming growth factor-beta signaling in T cells. Nat Med,
2001. 7(10): p. 1118-22.
21. Town, T., et al., Blocking TGF-beta-Smad2/3 innate immune signaling mitigates
Alzheimer-like pathology. Nat Med, 2008. 14(6): p. 681-7.
22. Laouar, Y., et al., TGF-beta signaling in dendritic cells is a prerequisite for the
control of autoimmune encephalomyelitis. Proc Natl Acad Sci U S A, 2008.
105(31): p. 10865-70.
23. Allen, S.J., et al., Adaptive and innate transforming growth factor beta signaling
impact herpes simplex virus 1 latency and reactivation. J Virol, 2011. 85(21): p.
11448-56.
103

24. Gorelik, L. and R.A. Flavell, Abrogation of TGFbeta signaling in T cells leads to
spontaneous T cell differentiation and autoimmune disease. Immunity, 2000. 12(2):
p. 171-81.
25. Hatton, B.A., et al., The Smo/Smo model: hedgehog-induced medulloblastoma
with 90% incidence and leptomeningeal spread. Cancer Res, 2008. 68(6): p. 1768-
76.
26. Liu L., et al., Visualization and quantification of T cell-mediated cytotoxicity
using cell-permeable fluorogenic caspase substrates. Nat Med, 2002 8(2):185-189.

104

Chapter 4 Methods
Chapter 1 Methods

Animals
B6.Cg-Tg (APPswePSEN1ΔE9)85Dbo/Mmjax MMRRC stock #034832 transgenic
mice (APP/PS1) and Tg(CD4- cre)1Cwi/BfluJ stock #017336 mice were bred
individually with B6.129S1-Stat3
tm1Xyfu
/J (Stat3
flox/flox
) stock #016923 mice for at least
3 generations to completely flox Stat3. APP/PS1Stat3
f/f
mice and CD4-creStat3
f/f
mice
were then bred to generate 1. APP/PS1
-
CD4-cre
-
Stat3
f/f
(WT) 2. APP/PS1
-
CD4-
cre
+
Stat3
f/f
(KO) 3. APP/PS1
+
CD4-cre
-
Stat3
f/f
(AD) 4. APP/PS1
+
CD4-
cre
+
Stat3
f/f
(ADTh17KO). APP/PS1 mice were crossed with CD4dnTGFBRII stock
#0055511. All mice strains are on the C57BL/6 background and were purchased from
Jackson Laboratory. All experimental procedures were monitored and documented
regarding age- and sex- matched controls. All mice were housed under standardized
conditions (access to food and water ad libitum) approved by the University of
Southern California Institutional Animal Care and Use Committee and performed
under the National Institute of Health guidelines. CD45.1
+
and CD45.2
+
on a C57BL/6
background were obtained from Jackson Laboratories and housed in the Bioresources
Unit in Trinity College Dublin. Unless otherwise indicated, data were obtained from
CD45.2
+
(WT) mice or APP/PS1 mice. All mice were maintained in controlled
conditions (temperature 22 to 23°C, 12-h light- dark cycle, and food and water ad
libitum), under veterinary supervision and experimentation was carried out under the
license granted by the HPRA and with the approval of the Local Ethics committee. All
105

animal groups were age- and sex-matched.
Generation of Aβ- specific Th17 cell lines, in vivo transfer and preparation of tissues
CD45.1
+
mice were immunized on the flank with Aβ1-42 peptide (75µg/ mouse)
emulsified in complete Freund’s adjuvant (CFA) containing 4mg/ml (0.4mg/ mouse)
of heat killed M. Tuberculosis. Mice were sacrificed 8-10 days later; Spleens and
popliteal lymph nodes were harvested and re-stimulated with Aβ1-42 aggregated
peptide (25µg/ml) in the presence of IL-1β (10ng/ml) and IL-23 (10ng/ml) to generate
A 𝛽 Th17 cells. Cells were suspended in RPMI-1640 culture medium, and incubation
continued for a further 4 days. Supernatant was collected for ELISA and cells were
washed and injected i.v. into 14-15-month-old APP/ PS1 mice (15x10
6

cells/mouse) or in 300µl serum-free medium. Control APP/PS1 mice and WT mice
received 300 µl serum- free medium alone. In a separate set of experiments, 11 month
old APP/PS1 and WT mice were initially injected i.p, with 𝛂 -il17 Ab (BioxCell
#BE0173) or a control IgG1 isotype control Ab (#BE0083) at 300ug and after 24h
were injected i.v. with A 𝛽 Th17 Mice were sacrificed with CO2 gas and transcardially
perfused with ice-cold PBS. Brains were extracted and quartered with a sterile surgical
blade.

Tissue Processing
Mice were perfused with sterile, ice-cold PBS and extracted for blood, plasma,
spleen, draining cervical lymph nodes, CSF, spinal cord, and brain. Brains were
extracted similar to previously published methods (Tan et al. 2002, Town et al. 2015).
106

Anterior three quarters of the brain were either snap frozen for protein isolation or
immersion-fixed in 4% paraformaldehyde in a graded series of sucrose diluted in PBS
(15% to 30% each incubation step at 4º C overnight) and embedded in optimal cutting
temperature compound (Sakura cat#25608-930) for cryosectioning. Protein was either
used for western blot and ELISA or extracted to make cDNA libraries for RNAseq.
RNA was extracted with Trizol (Life Technologies), from hippocampal homogenate.
Extracted DNA from these samples were purified using RNAeasy Mini Kit (Qiagen)
and then reverse-transcribed using SuperScript III first strand cDNA synthesis system
(Life Technologies).

Flow cytometry gating strategy
Cells from draining cervical lymph nodes, brain, and spleen were isolated and
stimulated with PMA/Ionomycin and Brefeldin A for standard intracellular staining
protocols. After 4h stimulation and intracellular staining of 24h, cells were gated on
singlets vs. doublets, side-scatter vs. forward-scatter lymphocyte gating populations,
and live/dead staining prior to gating next for CD45
+
CD3
+
cells. Next, cells from these
populations were gated on CD4
+
vs. IL-17
+
, IFN-y
+
, IL-10
+
, RORyt
+
, and FoxP3
+
and
compared to fluorescent minus one of each antibody, unstained, and all stained cell
samples for quantification.

Aβ and inflammatory panel for ELISA
After 15 min incubation on ice, homogenates were centrifuged at 14,000 g for 15 min
at 4°C; the supernatant contains triton soluble Aβ peptides. The triton-insoluble pellet
107

was subsequently extracted using 10 volumes of 5M guanidine-HCl (diluted in 50 mM
tris-HCl, pH 8.0), and pellets were briefly re-homogenized and shaken for 4 h at room
temperature to promote extraction of insoluble Aβ. Homogenates were centrifuged for
5min at 8,000 g. Triton-soluble and guanidine-HCl-soluble fractions were used as
input for ELISA detection of Aβ1-38, Aβ1-40 or Aβ1-42 peptides (Mesoscale discovery
cat# K15200G) according to the manufacturer’s recommendations. For the custom
inflammatory panel, cytokine antibodies for ifn-γ, il-1β, il-6, il-10, tnf-α, mcp- 1, mip-
3α, il-21, il-22, il23, il-17a (Mesoscale discovery cat#K152A0H-2). Total protein
concentrations were determined in each fraction by BCA protein assay (Thermo
Scientific cat#23225) and used for normalization of Aβ or cytokine levels. For in vitro
recall stimulation, coculture of splenocytes and lymphocytes from draining cerival
lymph nodes were recall stimulated with MOG35-55, PMA/Ionomycin, Aβ1-42, or no
stimulation for 4h prior to collection of supernatant for ELISA.

Immunohistochemistry and Thioflavin S staining
Sections were incubated in 10% normal horse serum (NHS), 0.1% Triton-X 100, and
phosphate buffer saline (PBS) for 1h at 20°C to block non-specific binding. Primary
antibodies were directed against CD11b (CD11b/Integrin αM (M-19), goat polyclonal,
1:1000, Santa Cruz), CD68 (rat; 1:1000, Serotec), 6e10 (mouse, 1:1000), CD3 (rabbit
monoclonal ,1:500, Novus Biologicals), GFAP (rabbit, 1:1000, Sigma) and SNAP 25
(mouse; 1:1000, Sigma) at 4°C overnight. After washing, slides were incubated (1h,
20°C) with appropriate Alexa Fluor 488- (1:1000), Alexa Fluor 594- (1:1000), Alexa
Fluor 633 (1:1000) - conjugated secondary antibodies (Life Technologies) to visualise
108

amyloid beta plaques (6e10), microglia (CD11b) and T cells (CD3) or phagocytic
activity (CD68) respectively. Sections were washed in water and incubated (1%
aqueous Thioflavin S (ThioS); 8 min, RT) to stain for aggregated protein. Finally,
sections were washed (50% ethanol, 2 x 5min; water, 3 x 5min),
air- dried and mounted with Prolong Gold containing DAPI (Invitrogen- Molecular
Probes). Sample fluorophores were imaged in separate channels using Nikon A1 laser
scanning confocal microscope and plaques were quantified using Image J software
(U.S. National Institutes of Health, Bethesda, MD). ThioS staining which reflects Aβ
plaque deposition was expressed as a percentage of the assessed area of the
hippocampus, cingulate cortex and entorhinal cortex and by the number of amyloid
plaques greater than 150 microns.

Preparation of tissue for PCR
The frontal cortex was homogenized in (1% β-mercaptoethanol in RA1 buffer) to
prevent RNA degradation. The resultant lysate was filtered using NucleoSpin® filters
and centrifuged (11,000 rpm, 1min). The filters were discarded, ethanol (70%, 350µl)
was added and the samples were placed in new collection tubes with silica containing
filters and centrifuged (11,000 rpm, 30 sec) to bind RNA to the membrane. Membrane
desalting buffer (350 µl) was added to the tubes to ensure salt removal and the
samples were centrifuged (11,000 rpm, 1min). rDNase reaction mixture (95 µl) was
added directly onto the center of the silica membrane to facilitate the removal of
contaminated DNA and the samples were incubated (RT, 15 min). The membrane was
washed with buffer RAW2 (200 µl), centrifuged (11,000 rpm, 30 sec) to inactivate
109

rDNase, washed twice with buffer RA3 (600 µl and 250 µl respectively) and
centrifuged (11,000, 30 sec; 11,000, 2 min) to dry the membrane completely. RNA was
eluted in RNase- free H2O (60 µl) and centrifuged (11,000, 1 min).

RNA Isolation, cDNA Synthesis and Relative Quantitative Real- Time PCR
RNA quality was determined by NanoDrop 2000 spectrophotometer. The ratio of
absorbance at 260/280 nm was used to assess the purity of RNA, with a value of ~ 2.0
indicating pure RNA. RNase-free water was added to the purified RNA samples to
give a final volume of 20 µl with a concentration of 500 ng total RNA in each sample.
RNA was reverse -transcribed into cDNA using the High-Capacity cDNA Reverse
Transcription kit (Applied Biosystems). cDNA master mix (20 µl per sample; 10X RT
buffer (4 µl) , 25X dNTPase (1.6 µl), 10X RT random primers (4 µl), multi-scribe
reverse transcriptase (2 µl), nuclease-free H2O (8.4 µl)) was added to each sample to
give a final volume of 40 µl and samples were run in a PTC-200 Thermal Cycler for 2h.
cDNA (2.5 µl) was added to each well in a Thermo-Fast 96 PCR Detection Plate
(Thermo Scientific), to which PCR mastermix (17.5 µl per well; target Primer (1 µl),
β-actin primer (1 µl), KappaProbe (10 µl), Dye Rox (0.4 µl), RNase-free H2O (5.5 µl))
was also added. Duplicate samples were assessed for each gene of interest by relative
quantitative real-time PCR (qPCR) (7300 Real-Time PCR system; Applied
Biosystem). Actin expression was used as a housekeeping gene.

PCR Quantification
Analysis of relative gene expression was measured by the comparative CT method.
110

The genes of interest were compared with the expression of actin mRNA which was
the endogenous control. The values were normalized to the WT group and the
differences between the groups were expressed as fold changes.

RNAseq
Total RNA was collected from frozen tissue sections containing the injection site, 2
sections per animal, and isolated using the FFPE RNeasy protocol (Qiagen).
Ribosomal RNA was depleted from 500 ng input using RiboMinus Eukaryote Kit with
concentration module (Ambion). Stranded mRNA libraries were generated from 50 ng
ribosomal depleted RNA using the TruSeq library preparation kit (Illumina) and
sequenced on the HiSeq 2000 platform. Single end 62 base pair reads were generated
and aligned to mm10 (mouse) reference genome with STAR, restricting the
alignment to only uniquely mapping reads with 1 possible mismatch
permitted (Kim et al., 2013; Trapnell et al., 2012). Several Bioconductor packages in
R (www.r-project.org) were used for the analysis of these data (Gentleman et al.,
2004).

Statistical Analysis
GraphPad prism version 6.07 was utilized for all statistics. Multi-group comparisons
were performed by one-way analysis of variance followed by standard post hoc tests
of significance of Dunnett’s or Tukey’s. All other tests were performed with Student’s
t test. In all scenarios, p≤0.05 was considered to be statistically significant All data are
111

presented as means ± SEM.

Chapter 2 Methods

CD34
+
hematopoietic stem cell isolation from human umbilical cords
Human umbilical cord blood samples were obtained from the Cedars-Sinai
Department of Obstetrics and Gynecology. Samples were obtained from mothers
screening negative for heritable diseases and giving birth without complications.
CD34
+
hematopoietic stem cells (HSCs) were Automacs sorted using the Miltenyi
Biotec human CD34 MicroBead Kit. HSC isolation was verified by flow cytometry,
and samples were enriched until CD34 purity was ≥ 95% and CD3
+
T cell abundance
was ≤ 0.1%, as previously described
23,24
.
Animals
Gamma irradiated NOD.Cg-Prkdc
scid
Il2rg
tm1Wjl
/SzJ (NSG) mice on the non-obese
diabetic/ShiLtJ background were obtained from the Jackson Laboratory (stock
#00057). All mice were housed in a barrier facility under specific pathogen-free
conditions and were given ad libitum access to food and water. All animal experiments
were approved by the University of Southern California Institutional Animal Care and
Use Committee (IACUC protocol #12023), and all animal studies were performed in
accordance with National Institutes of Health guidelines and recommendations from
the Association for Assessment and Accreditation of Laboratory Animal Care
International.

112

Haplotyping
HSCs and human neural progenitor cells (hNSCs) were haplotyped using the
Invitrogen AllSet+™ Gold HLA ABDRDQ SSP Kit (stock #54360D). DNA samples
were taken for HLA haplotyping analyses, and GAPDH was used as an internal
reference control. Gel electrophoresis was performed to validate AllSet+™ Gold HLA
data and interpreted from specific amplification patterns with UniMatch® Plus
software.

Cell culture and transplantation
We generated stem cell lines from HSCs and passed them into EZ media containing
epidermal growth factor (25 ng/ml), fibroblast growth factor (25 ng/ml), 0.2% heparin,
and 1% penicillin-streptomycin supplemented with Stemregenin 1. Cells were
maintained in culture no longer than 1 week to prevent differentiation. NSG pups (1-
2 days old) were irradiated with a 3-4 h interval at 2 x 2 Gy from a Cesium 137 source
at 3.75 Gy/min. At 4-12 h post-irradiation, mice were intra-hepatically injected with 7
x 10
5
HSCs in 25 µl of PBS, and mice were given 5- 6 months for human immune
system reconstitution. Human embryonic stem cells (hESCs) were maintained in EZ
media, and aggregated neurospheres were dissociated and plated onto poly-ornithine
plus laminin-coated wells to generate hNPCs. Human NPCs (3 x 10
5
cells/mouse)
were stereotaxically injected at +0.5mm bregma, +1.5-2mm lateral, and +3-3.5mm
depth. Humanized mice were euthanized 2 weeks after neural transplantation.

113

Tissue handling
Mice were sacrificed with CO2 and transcardially perfused with ice-cold PBS. Brains
were extracted and quartered using a mouse brain slicer as previously described
25
.
Brains, spleens and livers were either snap frozen for protein isolation or immersion-
fixed in 4% paraformaldehyde (PFA) in a graded series of sucrose diluted in PBS (10%
to 20% to 30%; each incubation step at 4º C overnight) and embedded in optimal
cutting temperature compound (OCT, Tissue Tek Sakura) for cryosectioning.
Immunohistochemistry
Neurospheres derived from hESCs were dissociated with 1% trypsin and DNase I (1
mg/ml), and single cell suspensions were transferred to plates coated with poly-
ornithine plus laminin in EZ media. Cells were fixed with 4% (PFA) for 10 min at
ambient temperature and washed in phosphate-buffered saline (PBS) at 4ºC. After 1 h
of blocking, neurospheres were incubated at 4
o
C overnight with primary antibodies
directed against early progenitor markers including Nestin (mouse, 1:100; Chemicon
International), Sox2 (goat, 1:100; Santa Cruz Biotech), and Tuj1 (mouse, 1:1000;
Covance). Mouse brains and spleens were cryosectioned at 20-30 μm using a Leica
Model CM1850 freezing microtome, applied to Superfrost Plus Gold slides, and
allowed to air dry for 10 min at ambient temperature. Immunofluorescence was
performed by first blocking with a solution containing 10% normal donkey serum,
0.3% Triton-X 100, and PBS for 1 h at 20ºC. Subsequently, sections were reacted with
primary antibodies directed against hCD45 (mouse biotinylated, 1:200;
BDPharmingen), mCD45 (rat polyclonal, 1:200; AbD Serotec), cleaved caspase-3
(1:300; Biolegend), human nuclear antigen (mouse HuNu, 1:300; Genway), and Iba1
114

(goat, 1:200; Lifespan Biosciences) at 4ºC overnight. After 3 washes for 5 min each,
slides were incubated for 1 h at 20ºC with appropriate Alexa Fluor 488-, Alexa Fluor
594-, and Alexa Fluor 647- conjugated secondary antibodies (Life Technologies).
After 3 additional washes with PBS at 20ºC, slides were air-dried and mounted with
Prolong Gold containing DAPI (Invitrogen-Molecular Probes).
Confocal image analysis
Slides were imaged in separate channels with a Nikon A1 laser scanning confocal
microscope. We acquired images of spleen sections (n = 12-14 per mouse) that were
stained with mCD45 and hCD45 to measure human immune cell reconstitution.
Images of mouse brain sections (n = 10-12 per right striatum) containing the transplant
site were immunostained with Iba1, hCD45, mCD45 and cleaved caspase-3 to quantify
humanized mouse immune responses to transplanted neural stem cells. Data are
reported as percentage of positive pixels divided by total area captured for each image.

Quantitative real-time PCR
Messenger RNA was extracted from hNPCs using the Trizol reagent (Life
Technologies). Trace contaminating DNA was digested with DNAse I and RNA was
purified using the RNAeasy Mini Kit (Qiagen) and reverse- transcribed using the
SuperScript III first strand cDNA synthesis system (Life Technologies). Human gene
expression was assessed by Taqman quantitative real-time PCR assays. Primers were
designed to target HLA-A, -B, -C, -DR and -DQ (Ref 10025636, Bio-Rad), and
GAPDH was used as an internal reference control.
RNAseq
We extracted total RNA from hNPC transplants (n = 2-4 per brain) using frozen tissue
115

sections and the fixed- formalin paraffin-embedded RNeasy protocol that can be used
for frozen sections (Qiagen). Ribosomal RNA was depleted from 500 ng of input
sample using the RiboMinus Eukaryote Kit with the concentration module (Ambion).
Stranded mRNA libraries were generated from 50 ng of ribosomal-depleted RNA
using the TruSeq library preparation kit (Illumina) and sequenced on the HiSeq 2000
platform. Single-end 62 base pair reads were generated and aligned to both mm10
(mouse) and hg19 (human) reference genomes with Tophat2; restricting the alignment
to only uniquely mapping reads with one possible mismatch permitted
42,43
. We used
Bioconductor packages written in R (www.r-project.org) for data assembly and
analysis
44
. We quantified reads on a per RefSeq gene basis using SeqMonk
(https://www.bioinformatics.babraham.ac.uk/projects/seqmonk/).

Statistical Analysis
GraphPad Prism, SPSS, and R software were used for statistical analysis of the data as
appropriate. We developed co-linear curve analysis (CoLC) to determine HLA gene
expression by hNPCs. One-way analysis of variance (ANOVA) was performed
followed by Dunnett’s T3 or Tukey’s post hoc tests, where appropriateness of the post-
hoc test was determined by Levine’s test for equality of the variance. In all cases, data
were considered statistically significant if p < 0.05. All data are presented as means ±
SEMs.

116

Appendix
Chapter 1

Fig 1. Restored cognitive function in ADTh17 (APP/PS1
+
CD4
cre
STAT3
fl/fl
) mice Barnes
maze. Barnes maze analysis over 10 days in all 4 genotypes. Training days were performed at
repeated trials=3 (day 1-4), retention day at day 7, and reversal of goal box at day 8 and 9.
*p<.05; **p<.01; ***p<.001 compared to WT littermate controls. All error bars are ± SEM.

117

Fig 2. Restored cognitive function in ADTh17 (APP/PS1
+
CD4
cre
STAT3
fl/fl
) mice in
cognitive tests. Open field analysis was performed for total distance travelled (a) and duration
of time spent in peripheral vs. central area of enclosure (b). Novel object recognition was
performed at repeated trials=3 for all genotypes. Mice were introduced to both objects for 5
min in the first trial and given 1h prior to testing short term memory (c). Y maze analysis was
performed in all 4 genotypes and measured on spontaneous alternation (d) and number of
entries in each arm (e). After 24 h, mice were given a new object to assess for long term
memory (f). *p<.05; **p<.01; compared to WT.

CD45
118

Fig 3. Infiltration of Aβ-Th17 cells in ADTh17 and ADTh17 mice with neutralizing anti-
IL-17. Congenic CD45.1
+
mice were in vivo and in vitro expanded for Aβ-Th17 cells for 10
days. Representative FACS plots of CD45.2
+
animals adoptively transferred with Aβ-Th17 and
given anti-IL-17 or isotype control for 30 days (a). Quantitation of the number of cells (b) and
% of cells (c) extracted from whole brain of each genotype. *p<.05. All error bars are ± SEM.

Chapter 3

Fig 1. Immunotherapeutic treatment of Medulloblastoma via nanoparticles (a) Schematic
diagram of TGF-β production and role of integrins in immune surveillance and tumor growth.
(b) Schematic of nanoparticles encapsulated with SB-505124 and RGD peptide.

119

Figure 2. Survival curve analysis of nanoparticle treatment in SmoA1 mice. (a) Survival Curve analysis
adapted from PNAS paper “T-cell TGF-β signaling abrogation restricts medulloblastoma progression, PNAS,
2014”. (b) Survival curve analysis of current report in SmoA1 mice not treated with nanoparticles. (c) Survival
proportions of nanoparticle treated SmoA1 mouse groups treated weekly for 12 weeks starting at 5 months of age
(*p<.05). NP=nanoparticle; NPR=nanoparticle+RGD peptide;NP+SB=nanoparticle+SB505124;
NPR+SB=nanoparticle+RGD peptide+SB505124.

120

Figure 3. Improved Splaying and Wobbling Ataxia motor behavior in SB-RGD nanoparticle treated SmoA1
mice. (a) Splaying and Wobbling ataxia test for motor impairment scale. (b) Mean splay scores over 12 weeks of
weekly injections in each treatment group (***p<.001, n.s=not significant) (c) Mean wobble scores over 12 weeks
of weekly injections in each treatment group (***p<.001, n.s=not significant).

121

Figure 4. Mobilization of activated T lymphocyte population in SB-RGD nanoparticle treated SmoA1 mice.
(a) FACS plots of drained lymph nodes from nanoparticle blank treatment group in SmoA1 mice (top panel) and
full nanoparticle treatment with RGD peptide and SB-505124 drug (bottom panel). (CD45+=pan-leukocyte marker,
TCRαβ= T-cell specific receptor (b) Graphical representation of the number of immune cells found in each
nanoparticle treatment group (*p<.05, **p<.01, n.s=not significant).

Abstract (if available)

Abstract T helper 17 (Tₕ17) cells are present in Alzheimer’s disease (AD) and in mouse models of the disease. However, the impact of Tₕ17 cells on neuroinflammation and AD pathogenesis is unknown. To address this, we crossed the APP/PS1 mouse model of cerebral amyloidosis with a cre-lox recombinant mouse deficient in Tₕ17 cells (APP/PS1⁺CD4ᶜʳᵉ-Stat3ᶠ/ᶠ). Tₕ17 depletion in APP/PS1 mice decreased cerebral amyloid burden and increased Aβ uptake and clearance by mononuclear phagocytes by quantitative in silico 3D modelling. Further, adoptive transfer of Aβ-specific Tₕ17 cells into APP/PS1 mice decreased Aβ phagocytosis and exacerbated cerebral amyloidosis. Finally, administration of anti-IL-17 neutralizing antibody prior to adoptive transfer reversed these phenotypes. Our results suggest that Tₕ17 cells negatively instruct mononuclear phagocytes, precluding Aβ clearance. Harnessing Tₕ17-mediated innate immunity may be therapeutically relevant for AD.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

TREM2 and C1q signaling regulates immunoproteostasis in Alzheimer's disease

PDF

Macrophage-specific Stat3 deletion in a mouse model of Alzheimer’s disease

PDF

Protein phosphatase 2A and annexin A5: modulators of cellular functions

PDF

Averting dementia: renin-angiotensin system and angiogenic cells in cognitive decline

PDF

Blood-brain barrier pathophysiology in cognitive impairment and injury

PDF

Innate immune response to amyloid-beta: relevance to Alzheirmer’s disease and neuroinflammation

PDF

Sex differences in the TgF344-AD rat model: Investigating the behavior, pathology, and neuroanatomical structures

PDF

Discovery of mitochondria-DNA-encoded microproteins in neurodegeneration

PDF

Role of oxidative stress in age-associated mild cognitive impairment and Alzheimer's disease

PDF

A transgenic mouse model with overexpression of human RAGE in endothelial cells presents enhanced amyloid-beta transport into the brain

PDF

The role of specific gut microbiota in regulating the development of Alzheimer抯 disease

PDF

Investigating the immunomodulatory mechanisms of steroid resistance in acute respiratory distress syndrome and steroid-resistant asthma

PDF

A novel therapeutic approach in asthma: depleting CD52-expressing leukocytes suppresses airway hyperreactivity and ameliorates lung inflammation

Asset Metadata

Creator Im, Kwok (Chris) Wai (author)

Core Title Adaptive immunity in the central nervous system

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Publication Date 07/25/2020

Defense Date 05/15/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Alzheimer's disease,immunology,Inflammation,microglia,Neuroscience,OAI-PMH Harvest,T helper 17 cells

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Finch, Caleb (committee chair), Pike, Christian (committee member), Shao, Ling (committee member)

Creator Email kwok.w.im@gmail.com,kwokim@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-347826

Unique identifier UC11663273

Identifier etd-ImKwokChri-8751.pdf (filename),usctheses-c89-347826 (legacy record id)

Legacy Identifier etd-ImKwokChri-8751.pdf

Dmrecord 347826

Document Type Dissertation

Rights Im, Kwok (Chris) Wai

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA