The role of intestinal mononuclear phagocytes in control ...

1

The role of intestinal mononuclear phagocytes in

control of mucosal T cell homeostasis

Casandra Maria Panea

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy in the Graduate School of Arts and Sciences

COLUMBIA UNIVERSITY

2016

2

©2016 Casandra Maria Panea

All rights reserved

3

Abstract

The role of intestinal mononuclear phagocytes in control of mucosal T

cell homeostasis

Casandra Maria Panea

The intestine is constantly exposed to a wide variety of dietary antigens,

commensal bacteria and pathogens, toward which it has evolved complex immune

responses to protect the host. The intestinal immune system relies on innate immune

cells, such as mononuclear phagocytes (MNPs), that include dendritic cells (DCs),

monocytes (Mo) and macrophages (Mfs), to sense and respond to luminal and mucosal

challenges. MNPs are essential players as they instruct adaptive immune cells, in

particular T cells, to discriminate between innocuous and harmful antigens. Generation of

different CD4 T cell responses to commensal and pathogenic bacteria is crucial for

maintaining a healthy gut environment, but the associated cellular mechanisms are poorly

understood. Lamina propria (LP) T helper 17 (Th17) cells participate in mucosal

protection and are induced by epithelium-associated commensal segmented filamentous

bacteria (SFB). Several reports suggest that the cytokine environment induced by gut

bacteria is sufficient to drive LP Th17 cell differentiation. In this context, intestinal DCs

are proposed to facilitate the conversion of naïve CD4 T cells to Th17 cells within gut-

draining lymph nodes. Whether such mechanisms control commensal-mediated Th17 cell

differentiation has not been examined. In this work, I explore the mechanisms of

induction of Th17 cells by SFB, with a particular focus on the role of antigen-presenting

cells in this process.

4

Initiation of CD4 T cell responses requires both major histocompatibility II

(MHCII)-mediated antigen presentation and cytokine stimulation, which can be provided

by the same or different subsets of intestinal MNPs. To test the requirement for either

function in the induction of Th17 cells by SFB, we analyzed the role of SFB-induced

cytokine environment in driving Th17 cell differentiation of non-SFB transgenic CD4 T

cells. We find that although the cytokine environment is important, it is not sufficient to

promote Th17 cell differentiation of activated CD4 T cells. In fact, we show that MHCII-

dependent antigen presentation of SFB antigens by intestinal MNPs is crucial for Th17

cell induction. Expression of MHCII on CD11c+ cells was necessary and sufficient for

SFB-induced Th17 cell differentiation. We also show that most SFB-induced Th17 cells

respond to SFB antigens, which stressed that they carry T cell receptors that recognize

SFB moieties. SFB primed and induced Th17 cells locally in the LP and Th17 cell

induction occurred normally in mice lacking secondary lymphoid organs.

Our results outline the complex role of MNPs in the regulation of intestinal Th17

cell homeostasis, and we investigated the contribution of individual subsets to SFB-

specific Th17 cell differentiation. Although the role of DCs in initiating T cell responses

is well appreciated, how Mfs contribute to the generation of CD4 T cell responses to

intestinal microbes is unclear. To this end, I examined the role of mucosal DCs and Mfs

in Th17 induction by SFB in vivo. Employing DC and Mf subset-specific depletion and

gain-of-function mouse models, I show that Mfs, and not conventional CD103+ DCs, are

essential for generation of SFB-specific Th17 responses. Thus, Mfs drive mucosal T cell

responses to certain commensal bacteria.

1

TABLE OF CONTENTS

List of Figures iv

List of Abbreviations vii

Acknowledgments x

Chapter one: Introduction 1

I. Architecture of the gut 1

II. Intestinal mononuclear phagocyte (MNP) system 4

III. Intestinal MNP subsets 5

A. Development of intestinal macrophages 10

B. Development of intestinal DCs 12

1. CD103+CD11b- dendritic cells (CD103 SP DCs) 13

2. CD103+CD11b+ dendritic cells (DP DCs) 14

3. CD103- dendritic cells (CD11b SP DCs) 15

IV. Intestinal CD4 T cells 16

A. Overview of effector CD4 T cell differentiation 18

B. Induction of intestinal CD4 T cell responses 18

C. Control of intestinal CD4 T cell differentiation by MNP subsets 21

1. Th1 cells 21

2. Th2 cells 22

3. Th17 cells 23

i. Commitment to Th17 cell lineage 23

ii. Th17 cell-inducing cytokines 24

iii. Microbe-mediated Th17 cell induction 25

i

2

4. Regulatory T cells 27

V. Functions of intestinal mononuclear phagocytes 28

A. Antigen uptake 28

B. Antigen processing and presentation 31

C. Priming of CD4 T cells 33

D. Induction of pTreg differentiation for oral tolerance 35

E. Induction of pTreg differentiation by commensals 37

F. Induction of Th17 cell differentiation 38

VI. SFB-specific modulation of host immunity 41

A. Segmented filamentous bacteria (SFB) 41

B. SFB-mediated regulation of innate and humoral immune responses 43

C. Innate recognition of SFB 44

VII. Commensal-specific effector CD4 T cell responses 46

Chapter two: Segmented filamentous bacteria antigens presented by intestinal

dendritic cells drive mucosal Th17 cell differentiation 50

I. Summary 51

II. Results 52

A. Microbiota-induced intestinal Th17 cells are selected on MHCII 52

B. SFB-induced intestinal Th17 cells recognize SFB antigens 53

C. Most-lamina propria Th17 cells recognize SFB antigens 57

D. MHCII expression on DC is necessary and sufficient for induction of

Th17 cells by SFB 58

E. MHCII expression on ILCs controls intestinal Th17 cells 60

ii

3

F. Th17 cell induction by SFB does not require LN or organized GALT 62

III. Experimental procedures 65

IV. Acknowledgments 69

V. Figures 70

Chapter three: Intestinal monocyte-derived macrophages control commensal-

specific Th17 responses 93

I. Summary 94

II. Results 95

A. DP DCs are dispensable for Th17 cell induction 95

B. CD103 DCs are dispensable for Th17 cell induction by SFB 97

C. Conventional DCs are dispensable for commensal Th17 cell induction at

steady state 98

D. Nongenotoxic depletion of intestinal monocyte-derived cells prevents SFB-

specific Th17 cell responses 99

E. Transfer of exogenous monocytes rescues defects in Th17 cell induction

following Mf depletion 100

F. Specific depletion of CD64 Mfs leads to the loss of SFB-mediated Th17 cell

induction 102

III. Experimental procedures 103

IV. Acknowledgments 107

V. Figures 108

Chapter four: Discussion 133

REFERENCES 146

iii

4

List of Figures

Chapter one figures

Figure 1-1. Architecture of the gut 3

Table 1-1. Phenotype of intestinal MNPs 8

Figure 1-2. Distinct ontogeny and phenotype of intestinal Mfs and DCs 9

Figure 1-3. Mechanisms of antigen sampling by intestinal MNPs 31

Chapter two figures

Figure 2-1. Induction of intestinal Th17 cells by SFB requires MHCII expression in the

periphery 70

Figure 2-2. SFB-induced intestinal Th17 cells preferentially respond to SFB antigens 71

Figure 2-3. Most intestinal SFB-induced Th17 cells recognize SFB 73

Figure 2-4. DC expression of MHCII is necessary and sufficient for SFB-mediated Th17

cell induction 74

Figure 2-5. RORgt+ ILCs inhibit differentiation of SFB-independent intestinal Th17 cells

through MHCII 75

Figure 2-6. Priming and induction of Th17 cells by SFB occurs in the small intestine 76

Figure 2-7. SFB induce Th17 cells in the absence of secondary lymphoid organs 77

Supplemental Table. RT-PCR primers. 78

Supplemental Figures

Figure 2-S1. SFB-induced intestinal Th17 cells require MHCII expression in the

periphery 79

Figure 2-S2. SFB do not induce Th17 cell differentiation of non-SFB Tg T cells 81

Figure 2-S3. SFB do not induce Th17 cell differentiation of non-SFB Tg T cells 83

iv

5

Figure 2-S4. SFB induce Th17 cells with diverse Vβ utilization 86

Figure 2-S5. Effects of SFB colonization in DCΔMHCII mice 87

Figure 2-S6. SFB-mediated Th17 cell responses in IECΔMHCII and ILC3ΔMHCII mice 89

Figure 2-S7. SFB priming of CD4 T cells in gut mucosa 91

Table 2-S1. Diverse TCR repertoire of SFB-recognizing hybridomas 92

Chapter three figures

Figure 3-1. CD103+CD11b+ (DP) DCs are dispensable for commensal Th17 cell

induction 108

Figure 3-2. CD103+ DCs are dispensable for commensal Th17 cell induction 110

Figure 3-3. Conventional DCs are dispensable for commensal Th17 cell induction 112

Figure 3-4. Intestinal Mfs are required for mucosal Th17 cell induction 114

Figure 3-5. Exogenous monocytes recover Th17 cell induction in Mf-depleted mice 116

Figure 3-6. Treatment with CSF1R-blocking antibody impedes Th17

responses to SFB 118

Figure 3-7. Central role of intestinal Mfs in generation of commensal-induced

Th17 cells 120

Supplemental Figures

Figure 3-S1. Phenotype of mononuclear phagocyte subsets 121

Figure 3-S2. DP DCs are not required for SFB-induced Th17 cell responses 123

Figure 3-S3. CD103+CD11b- (CD103 SP) DCs are dispensable for commensal Th17 cell

induction 124

Figure 3-S4. Cell subsets in CCR2-DTR mice following DT treatment 126

Figure 3-S5. Recovery of intestinal Mfs by monocyte transfer 128

v

6

Figure 3-S6. CCR2 expression on lamina propria MNP subsets 130

Table 3-S1. MNP subsets in various mouse models used in this study 132

vi

7

List of Abbreviations

AID – activation-induced cytidine deaminase AHR – aryl hydrocarbon receptor AML1 – acute myeloid leukemia 1 APC – antigen presenting cells BATF3 – basic leucine zipper transcriptional factor, ATF-like Bcl-6 – B-cell lymphoma 6 protein Blimp1 – PR domain zinc finger protein 1 BM – bone marrow CBir – commensal flagellin CCL – C-C chemokine ligand CCR – C-C chemokine receptor CD – cluster of differentiation cDC – classical/conventional dendritic cells CDP – common dendritic cell progenitor CLEC – C-type lectin domain family CLR – C-type lectin receptors CMP – common myeloid progenitor CX3CR1 – C-X3-C chemokine receptor 1, fractalkine receptor CXCR – C-X-C chemokine receptor DC – dendritic cells DP – double positive DSS – dextran sodium sulphate DT/DTR – diphtheria toxin/receptor EAE – experimental autoimmune encephalomyelitis EHEC – enterohemorrhagic Escherichia coli EMP – erythro-myeloid progenitors Esam – endothelial cell adhesion molecule FAE – follicle-associated epithelium FcRn – neonatal Fc receptor Flt3 – fms-like tyrosine kinase 3 receptor Flt3L – fms-like tyrosine kinase 3 ligand FoxP3 – forkhead box P3 GALT – gut associated lymphoid tissues GAP – goblet cell-associated passages GC – goblet cells GF – germ-free GM-CSF (Csf2) – granulocyte-macrophage colony-stimulating factor (colony stimulating factor 2) GMP – granulocyte/macrophage progenitor HSC – hematopoietic stem cells ID2 – inhibitor of DNA binding 2 IEC – intestinal epithelial cells IEL – intraepithelial lymphocytes IFN – interferon

vii

8

IgA – immunoglobulin A IL – interleukin ILC – innate lymphoid cells ILF – isolated lymphoid follicles iNOS – inducible nitric oxide synthase IRF – interferon regulatory factor LI – large intestine LP – lamina propria LTα – lymphotoxin α1β2 LTβR – lymphotoxin β receptor LysM – lysozyme M M-CSF (Csf1) – macrophage-colony stimulating factor (colony stimulating factor 1) M-CSFR (Csf1r) – macrophage-colony stimulating factor receptor (colony stimulating factor 1 receptor) MafB – V-maf musculoaponeurotic fibrosarcoma oncogene homolog B MAMPs – microbe-associated molecular patterns MDP – monocyte and dendritic cell progenitor Mfs – macrophages MHCII – major histocompatibility complex II MLN – mesenteric lymph nodes MMDTR – monocyte/macrophage diphtheria toxin receptor MNP – mononuclear phagocytes Mo – monocytes MyD88 – myeloid differentiation primary response gene 88 NFIL3 – nuclear factor, interleukin 3 regulated NOD – nucleotide oligomerization domain NLR – NOD-like receptors NRP1 – neuropilin 1 OTII – ovalbumin peptide-specific T cell receptor P2X, P2Y – purinergic receptor PP – Peyer’s patches PRR – pattern recognition receptor PSA – polysaccharide A pTregs – peripherally-induced regulatory T cells PU.1 – PU-box binding transcription factor RA – retinoic acid RANK – receptor activator of nuclear factor κ B RIPK – receptor interacting protein kinase RLR – retinoic acid inducible gene I (RIG-1) like receptors RORγ – RAR-related orphan receptor gamma SAA – serum amyloid A SFB – segmented filamentous bacteria SI – small intestine SIRPα – signal regulatory protein alpha SP – single positive SPF – specific pathogen-free

viii

9

STAT – signal transducer and activator of transcription T-bet – T-box 21 transcription factor TCR – T cell receptor TED – transepithelial dendrites TF – transcription factor TGF – transforming growth factor TNF – tumor necrosis factor Th1/2/17 – type 1/2/17 effector T helper cells TLR – toll-like receptor Tr1 – type 1 regulatory cells Trif – TIR-domain-containing adapter-inducing interferon-β TRP1 – tyrosinase related protein 1 tTregs – thymus-derived (natural) regulatory T cells WT – wildtype Zbtb46 – zinc finger and BTB domain-containing protein 46

ix

10

Acknowledgments

My journey as a PhD student has been one of both scientific and personal growth.

On the one hand, it broadened my scientific horizons, it fed my hunger for knowledge,

and it shattered the barriers of what is scientifically possible or impossible to pursue. On

the other hand, it revealed my own limits and it helped form my identity and career goals.

I could not have arrived at this defining stage in my life without the continuous support of

my family, friends, and mentors.

First, I would like to thank my thesis advisor and mentor, Ivaylo (Ivo) Ivanov, for

giving me the opportunity to grow as a scientist in his lab and for guiding me in my quest

to contribute to the field of mucosal immunology. I owe the work described in this thesis

and the knowledge I gained throughout my scientific journey to the support I received

from Ivo. Second, I thank all the past and present members of the Ivanov lab for their

constant scientific, technical, and moral support. I am especially grateful to Carolyn Lee,

Yoshiyuki Goto, Adam Farkas, Shahla Abdollahi, and Marta Galan-Diez, for their steady

patience and help, for their mentorship, and for building a rewarding and fun team

environment. I want to thank my qualifying committee (Virginia Papaioannou, Chozha

Rathinam, and Ben Ohlstein) and thesis committee (Boris Reizis, Steve Reiner, Chozha

Rathinam) for believing in me and providing insightful comments over the years for

advancing and completing my thesis study. I am also honored and thank Tobias Hohl for

accepting to participate in my thesis defense committee. In addition, I want to express my

gratitude to our collaborators (Terri Laufer, Leszek Ignatowicz, Milena Bogunovic, and

Tobias Hohl) and fellow immunology lab members for contributing to the progress of my

x

11

work, for interesting discussions and for their discerning comments and criticism. Many

thanks go to Stacy Warren for help through all the administrative aspects of being a

graduate student, to Amir Figueroa for maintaining the flow cytometry core, which I used

religiously, and to Mauricio for his steady help with our mouse colony.

I am grateful to all my friends, who have enriched my life and have been a

comfort through difficult and joyous times. Thank you Nick, Katrina, Claudia and Cip,

Mahala, Amy, and Gina for bringing sunshine and hope into my life. Thank you to my

adoptive families, Mary-Jo and late husband Richard Warren, and Mariana Ristea and

Rick Farber, who opened their homes and hearts to me and helped me through life and

professional dilemmas.

Last but not least, I am indebted to my family for being supportive, patient and

loving throughout the good, the bad, and the ugly. My mother and father, Mariana and

Ioan, have sacrificed tremendously so that my brother, Razvan, and I could pursue our

dreams, even when these dreams would send us miles and miles away from them. My

dear brother, Razvan, thank you for your contagious enthusiasm and optimism – I am

convinced your own PhD journey will be fruitful, fun and rewarding beyond words. My

love and life partner, Dan, thank you for your unconditional love, patience, sapience, and

encouragement - I am truly blessed to have you in my life. This thesis is for you, my

family.

xi

12

To the four pillars of love in my life

my parents, Mariana and Ioan Panea

my brother, Razvan Ioan Panea

my life partner, Dan Ionut Ristea

xii

1

Chapter one

INTRODUCTION

I. Architecture of the gut

The intestine functions as a port of entry for food antigens and clinically relevant

pathogens and a harbor for trillions of commensal microbes. The functions of the

intestine include digestion and absorption of nutrients, secretion of hormones, and

mounting of appropriate immune responses to food antigens, as well as commensal and

pathogenic microbes. The anatomical layers of the gut in conjunction with the varied

immune and non-hematopoietic cell types support these complex functions1. Intestinal

walls are divided into four main layers consisting of the mucosa, submucosa, muscularis

propria and serosa2-4 (Figure 1-1A).

The mucosa is the innermost layer and is composed of three layers: epithelium,

lamina propria, and muscular mucosae. As the first layer of the mucosa, the epithelium

faces the lumen and exhibits invaginations called crypts of Lieberkuhn and finger-like

projections known as villi. The epithelium contains several types of epithelial cells with

distinct functions. The most abundant cell type is the enterocytes, which are specialized

in nutrient absorption. Goblet cells secrete mucins for generation of a stratified mucus

layer5, while enteroendocrine cells secrete various gastrointestinal hormones6. Microfold

(M) cells contribute to development of immune responses, as they sample and deliver

antigens to gut-associated lymphoid tissue (GALT). Tuft cells initiate type 2 effector T

cell (Th2) responses.7-9. The precise functions of another epithelial cell lineage, cup cells,

2

are still unknown10,11. The crypts of Lieberkuhn are home to intestinal stem cells and

Paneth cells, an epithelial cell lineage that supports growth of stem cells and secretes

antimicrobial peptides into the lumen12. Underneath the epithelium and protected from

luminal contents and constant microbial aggressions is the second layer of the mucosa,

the lamina propria13,14,15. Lamina propria (LP) is a loose connective tissue replete with

immune cells active in tolerogenic and inflammatory responses. Associated with the

mucosa are lymphoid aggregates such as Peyer’s patches (PP) and isolated lymphoid

follicles (ILFs), known collectively as organized gut-associated lymphoid tissue (GALT).

PPs are present in the small intestine and contain M cells, which allow passage of luminal

antigens for sampling by antigen-presenting cells for induction of both immunoglobulin

A (IgA) responses and effector T cell responses. ILFs are scattered throughout the small

intestine and generated in response to microbial stimuli and control enteric flora through

induction of IgA responses16.

Underneath the LP is its supporting submucosa, which is outlined by smooth

muscles that form the muscularis externa2. The muscularis externa consists of two layers

of smooth muscles, longitudinal and circular, alternating with two layers of nerve

supplies, myenteric plexus and deep muscular plexus, respectively2. The enteric nervous

layers of the muscularis regulate gut peristalsis and tolerogenic immune responses

through crosstalk with muscularis macrophages 17,18.

3

Figure. 1-1: Architecture of the small intestine.

A. The small intestine is comprised of stratified layers of specialized tissue. The

small intestine can be subdivided into four major layers: mucosa, submucosa, muscularis

externa, and serosa. The mucosa consists of the epithelium and the lamina propria. The

epithelial sublayer, constantly replenished by stem cells residing in intestinal crypts,

functions as a barrier for the underlying layers from the constant luminal aggressions.

B. Overview of the immune cells of the small intestinal mucosa. Through production

of stratified mucus layers, antimicrobial peptides, and immunoglobulins, the intestinal

epithelial layer separates the immune cells of lamina propria from the contents of the

lumen. Mononuclear phagocytes (MNPs), including CD64 Mf and the different types of

LP DCs, have the capacity to sample the lumen for both oral and microbial antigens

while maintaining the integrity of the epithelial barrier. ILCs integrate stimuli from either

4

IECs or LP MNPs and complement and strengthen the intestinal epithelial barrier.

Various LP DC subsets prime T cells either in the GALT (i.e. PPs and ILFs) or the gut-

draining mesenteric lymph nodes to induce specific effector lineages. The majority of

mucosal effector T cells are activated, regardless of their intraepithelial or lamina propria

compartment, and they commit to their T helper cell fate in response to a specific

cytokine milieu. Depending on the type of GALT, B cells differentiate into IgA-

producing plasma cells in a T cell- dependent (PPs) or T cell-independent (ILFs) manner,

and the secretory IgA molecules are crucial for coating microbes. Abbreviations: IEC,

intestinal epithelial cell; IEL, intraepithelial lymphocyte; ILC, innate lymphoid cell; PP,

Peyer’s Patch; ILF, isolated lymphoid follicle; DC, dendritic cell; Mf, macrophage; IL,

interleukin; RA, retinoic acid; TGF, transforming growth factor; IgA, immunoglobulin A;

Th, effector T helper cell lineage; pTreg, peripherally-induced regulatory T cell.

II. Intestinal mononuclear phagocyte system

The intestine is constantly exposed to a wide variety of dietary antigens,

commensal bacteria and pathogens, toward which it has evolved complex immune

responses to protect the host. The intestinal immune system relies on innate immune

cells, such as dendritic cells (DCs) and macrophages (Mfs) to sense and respond to

luminal and mucosal challenges. MNPs are essential players as they instruct adaptive

immune cells to discriminate between innocuous and harmful antigens. As a result, they

promote tolerance to harmless dietary and commensal antigens and facilitate mounting of

active immunity to pathogens. Both DCs and Mfs can sample antigen from the

5

environment, and DCs can migrate to the gut draining lymph nodes (mesenteric lymph

nodes, MLNs), where they present antigens to naïve T cells for induction of appropriate

effector T cell responses. Mfs are mostly resident in the lamina propria and perform local

functions including engulfing and clearing pathogens, amplifying effector T cell

responses and recruiting other immune cells for tissue repair and wound healing. While

much is known about the functions of intestinal MNPs in induction of effector T cell

responses to various pathogens, their role in commensal-driven T cell homeostasis is less

clear. Elucidating the contribution of the different MNP subsets toward commensal-

mediated T cell responses will help uncover mechanisms of discrimination between

harmless and harmful microbes.

III. Intestinal MNP subsets

Intestinal MNPs arise from a common BM progenitor, but intestinal DC subsets

and Mfs follow distinct developmental pathways due to defined precursors (Figure 1-2).

The common myeloid progenitors (CMP) in the BM have the potential to generate DCs,

Mfs, granulocytes, neutrophils and eosinophils19. CMPs then differentiate into two

lineages, macrophage/dendritic cell progenitors (MDPs) and granulocyte/macrophage

progenitors (GMPs), where the MDP lineage generates only DCs and

monocytes/macrophages20,21. MDPs are identified based on expression of chemokine

receptor CX3CR1, growth factor receptors FLT3, M-CSFR, and stem cell factor, cKIT,

while lacking expression of major histocompatibility complex II (MHCII), lymphoid,

erythrocyte and NK lineage markers. MDPs give rise to two lineages, monocytes and

6

common DC progenitors (CDP)20. CDPs retain expression of CX3CR1, FLT3 and M-

CSFR but lose potential to differentiate into monocytes/macrophages. CDPs generate

plasmacytoid DCs (pDCs) and pre-DCs22,23. Pre-DCs are committed to the conventional

DC (cDC) lineage and express CD11c but are MHCII negative. In addition, expression of

zinc finger transcription factor Zbtb46 identifies a subset that has lost pDC potential as

early as the CDP lineage24. This finding underscores Zbtb46 as a marker of both cDC-

restricted progenitors and of differentiated cDC subsets. Pre-DCs exit the BM and

circulate in blood to tissues, where they mature into cDC1 cells (CD8α+ in lymphoid

organs) and cDC2 cells (CD11b+ or CD4+ in lymphoid organs) that undergo local

proliferation24,25.

Ly6chi monocytes retain expression of CX3CR1 and M-CSFR, gain expression of

CD11b and the chemokine receptor CCR2, but lose expression of FLT3 and cKIT. BM

Ly6chi monocytes have a short half-life and emigrate from the BM in response to

chemokine ligand 2 (CCL2, also known as monocyte chemoattractant protein 1 (MCP1))

and CCL7 (also known as MCP3), which are ligands for CCR226,27. While MDPs and

CDPs reside within the BM, pre-DCs and Ly6chi monocytes circulate in the blood and

can extravasate into tissues to repopulate DC and Mf populations, respectively.

Circulating Ly6chi monocytes give rise to Ly6clow monocytes, which mostly patrol blood

vessels to ensure their integrity at steady state28,29. Ly6chi monocytes are constantly

recruited to tissues such as intestine, skin and heart to replenish resident macrophages30-

32. In other tissues, macrophages originate from embryonic progenitors. Microglia

develop exclusively from yolk sac-derived progenitors identified as erythro-myeloid

progenitors33, EMPs. Most other tissue-resident macrophages originate from EMP-

7

derived fetal liver macrophages and fetal liver monocytes, although the latter origin is

still debated due to differences in fate mapping models and marker identification of

progenitor populations34-38. A small contribution from adult HSC-derived circulating

monocytes is seen with increase in age or in response to tissue-specific signals 32,35,39.

Once the progenitors seed the tissue, they undergo local proliferation to maintain the

resident macrophage niche throughout adulthood40-46. Similarly, during radiation-induced

monocytopenia or inflammation, remaining resident macrophages expand via local

proliferation47-50.

Intestinal MNPs consist of several DC and Mf subsets, which are interspersed

within the GALT, LP, muscularis externa and populate MLNs3,51,52 (Figure 1-1B). MNPs

are identified based on expression of CD45, which marks all descendants of bone

marrow-derived hematopoietic stem cells, and lack of lymphocyte, granulocyte and

erythrocyte lineage markers53. Tissue MNPs express integrin CD11c and upregulate

expression of major histocompatibility complex II (MHCII)54. LP and MLN DCs express

high levels of both MHCII and CD11c, whereas LP Mfs and muscularis Mfs express

intermediate and low levels of CD11c, respectively3,17. Within the intestine, LP and MLN

DC subsets express conventional DC markers CD24 and CD26, and differential levels of

integrins CD103 and CD11b. LP, muscularis and MLN macrophages lack CD103 but

express CD11b, as well as CX3CR1, and the phagocytic receptor, FcγRI or CD64.

Transcriptomic analyses revealed additional transcriptional factors and surface receptors

that define these populations based on their origins, phenotypes and functions (Table 1-1)

17,55-58.

8

Currently, it is unclear whether progenitors seed discrete segments and

compartments of the gut and whether distinct precursors replenish the same MNP subset

during homeostasis and inflammation. While some recent studies propose that precursors

to lymphoid classical DC lineages are primed in the bone marrow57, it is not known

whether the same imprinting occurs for peripheral DC and macrophage progenitors. In

addition, confounding evidence around phenotypic markers used to describe the fate or

function of specific subsets (i.e. CX3CR1 and F4/80 expression on macrophages versus

DCs; MHCII, CD80 and CD86 positivity to describe functional antigen presenting cells)

underscores the need to move from phenotypic to functional classification of MNPs.

Table 1-1. Phenotype of intestinal MNPs. Intestinal MNPs can be classified as DCs or

Mfs based on their origins, their transcription factor requirements and other surface

receptors or ligands that instruct their development and function. MNP-mononuclear

phagocytes; TF-transcription factors; SP-single positive; DP-double positive. References:

17,18,30,55,56,58-61 and Immgen Gene Skyline 2016.

9

Figure 1-2. Distinct ontogeny and phenotype of intestinal Mfs and DCs (Adapted

from 62). In response to a CCL2 gradient, CCR2+Ly6chi blood monocytes enter the gut

mucosa regularly and differentiate through a series of monocyte/macrophage

intermediates into mature, tissue-resident intestinal Mfs. Intestinal Mfs express high

levels of CD64, F4/80, and CX3CR1. Albeit some DCs express intermediate levels of

F4/80 and CX3CR1, all intestinal DCs express high levels of CD24 and Zbtb46.

Committed Flt3L-dependent progenitors enter the gut mucosa constitutively and

reconstitute at least three different intestinal DC subsets. Both CD103-negative subsets

are presumably GALT-derived, but the populations are heterogeneous and their exact

10

origins are unclear. Abbreviations: GALT, gut-associated lymphoid tissues; Mf,

macrophage; cDC, classical dendritic cell; Flt3L, fms-like tyrosine kinase 3 ligand;

Zbtb46, zinc finger and BTB domain containing 46; IEC, intestinal epithelial cell.

A. Development of intestinal macrophages

Macrophages are the most abundant MNPs in both SI and LI lamina propria63

representing over 50% of the CD11c+MHCII+ cells at steady state64. Development of

tissue-resident macrophages has been thoroughly investigated34-38.

Embryonic progenitors seed the intestinal mucosa and proliferate in situ during

the neonatal period but upon weaning, they are fully replaced by newly differentiated

macrophages 65. These adult intestinal macrophages originate from circulating

monocytes, which are recruited to the intestine through a CCL2-CCR2-dependent

mechanism and differentiate locally in the lamina propria in response to bacterial signals

65. In competitive CCR2-/-:WT bone-marrow chimeras, macrophages derive exclusively

from CCR2-sufficient monocytes 30,66. However, monocytopenic CCR2-/- mice show

similar proportions of intestinal macrophages as wildtype controls pointing to a role for

CCR2-independent recruitment of circulating monocytes to the lamina propria, as has

been described during bacterial infection26 (our observations and 66,67). CCR2-/- mice

show similar maturation of intestinal macrophages and priming of CD4 T cells compared

to wildtype controls at steady state 67, but CCR2-/- macrophages are functionally impaired

and fail to license ILC3 cells to clear infection with C. rodentium 68.

11

Intestinal macrophage development occurs through a series of monocyte-

macrophage intermediates (described as “Mo-waterfall” 16), which is characterized by

gradual downregulation of LY6C expression and upregulation of CX3CR1, MHCII,

CD64, and CD11c expression leading to Ly6cnegCCR2low/negCX3CR1hi

CD64hiMHCIIhiCD11cint/lowCD11b+ Mfs 30,36,52. A similar differentiation process has been

noted for dermal monocyte-derived DCs 32. The “Mo-waterfall” is an attribute of recently

extravasated monocytes because LY6C-expressing blood monocytes are exclusively

Ly6chiCCR2hiCX3CR1intCD64lowMHCIInegCD11clow/negCD11b+ 30. Ly6chi blood

monocytes generate Ly6cnegCX3CR1hiCCR2-CD11b+CD115+ blood monocytes by 19hrs

upon exit from BM 28,36,52, however, adoptive transfers of Ly6cneg cells cannot replenish

the lamina propria macrophage niche 52. Therefore, Ly6chi blood monocytes are the sole

progenitors of lamina propria macrophages. Similar to other nonlymphoid tissues, lamina

propria macrophages undergo clonal expansion from extravasated Ly6chi blood

monocytes as competitive monocyte transfers lead to patchy reconstitutions of

macrophages52. Lamina propria macrophage development from blood monocytes occurs

under the control of macrophage-colony stimulating factor (M-CSF) 18,51,52,69,70 and does

not rely on fms-like tyrosine kinase 3 ligand (Flt3L) 51,52. In addition to growth factors,

commitment to and maintenance of the macrophage fate are driven by transcription

factors, e.g. PU.1, AML-1 and MafB 71-73.

MLN macrophages appear to develop directly from blood monocytes and do not

require a lamina propria intermediate as numbers of MLN-resident macrophages are not

reduced in CCR7-/- mice, in which migration of MNP subsets to MLN is impaired 30. In

the case of PPs, both Ly6chi and Ly6clow/neg blood monocytes can replenish macrophages

12

but in distinct subcompartments, i.e. Ly6clow/neg blood monocytes can repopulate the

subepithelial dome of PPs, but not the PP-associated villi 52. Muscularis Mfs are depleted

similarly to mucosal macrophages by monocyte-macrophage ablation models, such as

LysMΔCsf1rDTR (also known as MM-DTR) and anti-CSF1 monoclonal antibody,

suggesting a similar developmental origin to lamina propria macrophages 70. While their

dependency on Ly6chi blood monocytes has not been investigated directly, Seo et al

suggest that CD115+CD11clowMHCII+CD11b+ muscularis macrophages are not depleted

in DT-treated CCR2-DTR, which they attribute to a lack of contribution of circulating

monocytes to this population68. However, these results may indicate a slower turnover

rate of muscularis macrophages compared to mucosal macrophages since the DT-treated

recipients were analyzed 72hrs after the first DT injection. Indeed, when we treat CCR2-

DTR mice with DT for longer periods of time, we could achieve complete depletion of

both mucosal and muscularis macrophages 69.

B. Development of intestinal DCs

Intestinal LP DCs are short-lived and are constantly replenished from pre-

DCs51,52,55,74. The LP dendritic cell network can be divided into three main populations

based on differential expression of CD103 and CD11b and developmental control by

transcription and growth factors. An additional subset of CD103-CD11b- DCs has also

been proposed, however, its origins are unclear as this subset is highly variable and may

contain contaminating MNPs from MLNs62. All DC subsets are represented in different

proportions throughout distinct segments of the small intestine (i.e. duodenum, jejunum,

13

and ileum), but whether their localization contributes to regional immune specialization

remains to be determined64. Future studies will have to address also whether DC

progenitors home to distinct segments of the intestine to replenish local MNP

populations.

CD103+CD11b- dendritic cells (CD103 SP DCs)

Similarly to CD8α conventional DCs, and other nonlymphoid tissue CD103+

DCs, LP CD103+CD11b- dendritic cells develop from Flt3L-dependent pre-DCs via the

MDP-CDP axis and do not rely on GM-CSF or M-CSF 51,52,56,75,76.

CD103 SP DCs represent ~30% of total cDCs in the small intestine, ~75% of total

cDCs in the colon, over 50% in PPs and migratory MLN DCs 51,64,77. Despite their large

representation in the GALT, CD103 SP DCs are present in normal numbers in the LP of

RORγt-/- mice, which lack GALT, suggesting that they do not require GALT for their

differentiation and development 78. CD103 SP DC development is driven by the

transcription factors ID2, IRF8, BATF3, and NFIL3 74,79-82, although ID2 expression is

not restricted to CD103+ and CD8α cDCs 83 and NFIL3 also controls production of

IL12p40 from macrophages84. Autoactivation of IRF8 is required for specification of an

early clonogenic progenitor of CD8α DCs in the BM. BATF3 maintains autoactivation

of IRF8 in this progenitor and ensures final differentiation to CD8α cDCs. In the absence

of BATF3, differentiation into CD8α DC fails due to decay of IRF8 autoactivation and all

pre-DC progenitors generate CD4+ cDCs 85. This study complements an earlier finding

that commitment to cDC1 (CD8α DCs) and cDC2 (CD4+ DCs) lineages occurs in the BM

and not the periphery 57. Moreover, comparative transcriptional profiling of CD103 SP

14

DCs and CD103+CD11b+ DCs (DP DCs) revealed divergent programming in the

specification of each lineage 58. For example, the transcriptional repressor BCL6 was

enriched in CD103 SP DCs and CD103 SP DC development was severely impaired in

BCL6-deficient mice. On the other hand, Blimp1 was preferentially expressed in DP DCs

and DC-specific ablation of Blimp1 resulted in decreased levels of DP DCs and a reduced

ratio of DP DCs to CD103 SP DCs 58.

CD103+CD11b+ dendritic cells (double-positive, DP DCs)

DP DCs represent over 60% of total cDCs in the small intestinal lamina propria,

~5-10% of the total cDCs in PPs and colonic lamina propria and less than 50% of

migratory MLN DCs 51,64,70,77.

DP DCs resemble CD4+CD11b+ conventional DCs through their requirement for

the growth factor Flt3L and transcription factors Irf2, Irf4 and Notch2 for their

development and survival 56,61,86. As mentioned above, the transcriptional repressor

Blimp1 also appears to control in a cell-intrinsic manner the development or maintenance

of DP DCs 58. Based on competitive BM chimeras, DP DC development partially

requires GM-CSF signaling to maintain normal numbers 51,52. In addition, lymphotoxin β

receptor (LTβR) signaling promotes homeostatic expansion of DP DCs, similarly to its

contribution towards splenic Notch2-dependent Esam+CD11b+ DCs 86,87. Signal

regulatory protein alpha (SIRPα; CD172)-CD47 axis may also control the survival of DP

DCs as SIRPα- and CD47-deficient mice have 50% reduction in DP DC levels in SI LP,

LI LP and MLN 88.

15

DP DCs may be a heterogeneous population that includes some monocyte-derived

cells as some expression of F4/80, and intermediate levels of CX3CR1 and CSF1R have

been observed. Moreover, DP DCs are not completely deleted in Flt3L-/- mice and

Zbtb46-DTR BM chimeras treated with DT for a short period of time 56,57,70,74,75. The

heterogeneity of this population was proposed to resemble that of CD8+CD11b+ splenic

DCs, which are subdivided into Notch2-dependent Esam+CD11b+ cells and Notch2-

independent Esam-CD11b+CX3CR1-GFPhi monocyte-like cells 56,86. In our work, we

noticed that a small proportion of DP DCs express F4/80int, CX3CR1int and CCR2int

(CD64 and CD115 are not expressed), but we also obtained complete ablation of the

population in Zbtb46-DTR BM chimeras and no significant ablation in LysMΔCsf1rDTR or

CCR2-DTR mice treated with DT for 7 days. Therefore, it is possible that the

composition of this subset depends on environmental cues, such as inflammatory

microbiota.

It is currently unclear whether the two subsets of CD103 DCs develop from

defined progenitors and whether they are clonally committed prior to entry into the

intestinal lamina propria. In fact, recent findings suggest commitment occurs in the BM57.

CD103− intestinal DCs (also, CD11b SP DCs)

Most intestinal DCs express CD103 (discussed above), but a CD103− DC

population has emerged in recent years. CD103− DCs express CD11b and were grouped

initially with CD11b+ Mfs, but they express the DC-specific marker CD24 and lack

expression of CD64, distinguishing them from CD64+ Mfs 56,61,64,66,69,70,77,78. CD11b SP

DCs represent ~5-10% of total cDCs in small intestine, colon and GALT 56,61,66,77.

16

Development of CD11b SP DCs has yet to be fully understood because of the lack of

known developmental transcription factors. CD11b SP DCs expand in response to Flt3L

and are reduced in RORγt-/- mice suggesting that they are GALT-derived 61,66,78. IRF4

deficiency impairs the development of CD103- migratory MLN DCs, suggesting a

potential role for this transcription factor in the specification and migration to the lymph

of a subset of CD103- SI LP DCs that expresses CCR2 61,66. Earlier BrdU pulse-chase

experiments revealed that CD11b SP DCs accumulate within MLNs at a faster pace than

within SI LP and undergo local homeostatic proliferation suggesting distinct origin from

blood precursors89. In contrast to CD64 Mfs, Ly6clowCD11b+CCR2neg cells, can replenish

PP CD11b SP DCs within the subepithelial dome8. Transfers of Ly6chigh blood

monocytes do not reconstitute CD11b SP DCs, demonstrating that these cells do not

share developmental origin with intestinal macrophages.

CD11b SP DCs express intermediate levels of CX3CR1 and are reduced, albeit

not ablated, in mice treated with high doses of anti-CSF1R monoclonal antibody, as well

as DT-treated LysMCsf1rDTR mice (70, and our own observations). Currently, the origin of

this population is debated and they have been proposed to be a mix of conventional DCs,

monocyte-derived cells and monocyte-macrophage intermediates 56.

IV. Intestinal CD4 T cells

The intestinal LP contains both innate and adaptive immune cells involved in

acquiring, processing and translating food and microbial stimuli into information needed

to maintain intestinal immune homeostasis. As mentioned above, intestinal MNP subsets

17

sample antigens from the environment and deliver it to mucosal sites where adaptive

immune responses are initiated. Within these sites, T and B cells are primed and

instructed to home to the LP, or disseminate systemically, to perform their effector

functions. Beside LP, a unique subset of T cells resides in the epithelial layer. These T

cells, called intra-epithelial lymphocytes (IELs) consist of 60% γδ TCRs and fewer αβ

TCRs90,91. TCRγδ IELs are mostly CD8αα+, whereas TCRαβ IELs are either CD8αα+ or

CD8αβ+. IELs are either natural or induced depending on they type of antigens they

recognize and their mechanism of induction92. Natural IELs are activated in the thymus in

presence of self-antigens, populate the gut early in life and maintain constant numbers

over time. Induced IELs are conventional T cells, which encounter their non-self antigen

in peripheral lymph nodes and subsequently migrate to the epithelium. Recent work has

revealed that IRF8-dependent CD103+CD11b- DCs are responsible for development of

CD8αβ and CD4+CD8αα+ IELs93. The majority of LP CD4 T and IELs present in

conventional mice have effector/memory phenotype, which is explained by constant

antigen exposure, however their specificities remain poorly characterized 94-97. Effector

CD4 T cells dominate the intestine and GALT at steady state. Effector CD4 T cells

consist of diverse subsets with specific cytokine and chemokine profiles. They activate

cytotoxic T cells, instruct MNPs to perform bactericidal functions, and promote B cell

differentiation and antibody production. Th1, Th2 and Th17 cells promote immune

responses and regulatory T cells (Tregs) perform anti-inflammatory functions (Figure 1-

1B).

18

A. Overview of effector CD4 T cell differentiation

The first step of mucosal CD4 T cell activation involves MHCII-mediated

presentation of antigens by innate immune cells. Engagement of the T cell receptor with

cognate antigen-MHCII complex is then followed by crosslinking of costimulatory

molecules from both the APC and the primed CD4 T cell. Cytokines derived from either

the same APCs or other innate immune cells are crucial for initiating the transcriptional

changes required for skewing the differentiation of activated CD4 T cells towards a

specific lineage. Key cytokines produced by each lineage then play a role in maintenance

of the lineage identity through positive feedback or auto-amplification loops as well as

through cross-inhibition of other lineages. Nevertheless, mature effector CD4 T cells can

de-differentiate in response to various signals, i.e. from Tregs to Th1798,99, from Th17 to

Tregs100 and from Th17 to Th1101. Conversion of CD8 T cells into MHCI-restricted

regulatory-like CD4 T cells in the colonic lamina propria has also been reported,

underscoring the unique nature of the gut environment 102,103.

B. Induction of intestinal CD4 T cell responses

Organized secondary lymphoid tissues associate with each organ and are thought

to be the sites where T cell responses are initiated (i.e. inductive sites). Naïve T cells

encounter luminal antigen in the gut-draining MLN or GALT, which includes PPs and

ILFs. Formation of any of these structures requires lymphotoxin-α1β2 (LTα)-LTβR

interactions and RORγt. In addition, ILFs also depend on RANK and TNF-receptor I

signaling for their development 104,105.

19

Antigen-loaded DCs from the LP, but also PPs and ILFs, migrate to MLNs in a

CCR7-dependent manner to prime and activate naïve T cells106,107. MLNs are the main

induction sites for regulatory T cell (Tregs)-facilitated oral tolerance as mice lacking

MLNs or mice unable to deliver antigens to MLNs fail to induce tolerance to orally

administered soluble antigens107-109. Most of migratory MLN DCs can be identified as

MHCIIhiCD11c+ cells compared to resident MLN DCs, which are MHCIIint. Homing of

activated T cells to the gut mucosa relies on expression of homing receptor CCR9 and

integrin α4β7, which are induced on T cells while still present in the MLNs by CD103 SP

DCs 77,89,93. The ligand for integrin α4β7 is mucosal addressin cell-adhesion molecule

(MadCAM)-1110, which is expressed by endothelial cells of mucosal tissues, whereas the

ligand for CCR9 is the chemokine CCL25111, which is released by small intestinal

epithelial cells. CCR9-dependent homing is a characteristic of SI LP, as G protein-

coupled receptor (GPR) 15 controls T cell homing to the colonic LP112.

In the absence of DP DCs 61,113 both induction of gut tropism and peripherally-

induced Tregs (pTregs) are unaffected reinforcing a clear division of labor between

CD103 SP DCs and DP DCs. The ability of CD103 SP DCs to induce gut tropism on

activated T cells relies on their ability to synthesize the vitamin A metabolite, retinoic

acid, whose production is enhanced by intrinsic IL-4 signaling, TLR2 ligation and

GMCSF signaling 114-117. Both in vitro and ex vivo assays blocking retinoic acid receptor

signaling abolishes the ability of CD103 SP DCs to convert activated T cells into pTregs,

highlighting the essential role RA plays in conjunction with TGFβ for induction of

FoxP3+ pTregs 118-121. In fact, CD103+ DCs can activate latent TGFβ through their

20

expression of αvβ8, which endows them with the specialized ability to induce pTregs in

the MLNs 122,123.

In addition to the RA-producing CD103 SP DCs, MLN-resident stromal cells

produce high levels of RA and can both imprint activated T cells, specifically with α4β7,

and enhance the gut-tropism function of migratory CD103 SP DCs 124. This is a feature of

MLN resident stromal cells and of the mesenteric microenvironment as transplantations

of peripheral nonmucosal lymph nodes within the gut mesenteries does not result in

imprinting of gut homing receptors on T cells, even though SI LP DCs can populate the

graft 124,125.

PPs are lymphoid aggregates that develop embryonically. They are covered by

specialized epithelium, called follicle-associated epithelium (FAE). PPs contain germinal

centers, where IgA B cells develop in a T cell-dependent manner. PPs are found

throughout the small intestine but are more pronounced in the terminal ileum, where

bacterial loads are greater. In contrast to MLNs, PPs do not have afferent lymphatics, but

instead acquire antigens directly from the lumen through M cells. M cells deliver the

antigens to DCs localized underneath, in the subepithelial dome 126. PPs also play a role

in oral tolerance and may deliver antigens via efferent lymphatics to MLNs 126,127.

ILFs are small lymphoid aggregates, generally containing a single germinal

center. They develop after birth, progressively, from even smaller lymphoid aggregates

known as cryptopatches, in response to commensal and stromal cell-derived signals 128-

131. Within ILFs, B cells can class switch to IgA in the absence of T cells 128. IgA B cells

generated in ILFs and PPs migrate to MLNs to differentiate into IgA-secreting cells,

which then home to the lamina propria to release secretory IgA 128.

21

C. Control of intestinal CD4 T cells differentiation by MNP subsets

1. Th1 cells

Classical Th1 cells are primed with antigens derived from intracellular bacteria

and viruses and respond to type I and II interferons and IL-12. These cytokines trigger the

sequential activation of STAT1 and STAT4 and finally, the induction of the master

transcriptional regulator T-bet and the signature cytokine, IFNγ 132-134. Th1 cells express

CCR1, CCR5 and CXCR3, which allow them to traffic to inflammation sites 135. In

addition to their ability to control infections with intracellular pathogens such as

Toxoplasma gondii 136, Th1 cells have also been implicated in various autoimmune

inflammations, including inflammatory bowel diseases 94,137,138.

While Th1 cells are represented in the LP of both SI and colon at steady state, the

exact signals that trigger their development, the role of GALT in the priming of these

cells as well as their function at steady state are not so clear 135,139-142. Colonization of GF

mice with Bacteroides fragilis, a human-specific gram-negative commensal, polarizes

CD4 T cells towards a Th1 phenotype, although IL-10 producing Tregs are also induced

in the gut143. B. fragilis-derived Polysaccharide A (PSA) induces Th1 cell differentiation

via stimulation of IL-12 secretion by CD11c+ cells and PSA antigen presentation

underscoring the significance of innate immune pathways in Th1 cell development 143.

Among intestinal MNPs, CD103 SP DCs are proposed to be the main inducers of SI Th1

responses and cytotoxic T lymphocyte activity in vivo144. To induce Th1 responses,

CD103 SP DCs produce IL-6 and IL-12p40 following ligation of TLR3, TLR7, and

TLR9144. Moreover, CD103 SP DCs are crucial for mounting protective Th1 responses to

22

Trichuris muris in MLN93. Similarly, ablation of colonic CD103 SP DCs in DT-treated

Clec9A-DTR mice leads to severe colitis because of deficient IFNγ-dependent activation

of anti-inflammatory genes in IECs145. In contrast, during colitis Ly6chi monocytes

develop into proinflammatory CD103-CX3CR1intCD11b+ DCs, which accumulate in the

LI LP, secrete IL-12, IL-23, and iNOS and drive differentiation of IFNγ-producing Th1

cells60. During intestinal infection, intestinal epithelial tissue damage leads to leakage of

commensal antigens into the lamina propria, which drives both pathogen-specific effector

Th1 cell differentiation and commensal-specific memory Th1 cell development 146.

Therefore, different mechanisms of induction and MNP subsets are required for driving

Th1 cell differentiation at steady state and during inflammation.

2. Th2 cells

In neonatal and germ-free animals, intestinal effector CD4 T cells are biased

towards Th2 lineage. Early life microbial colonization leads to a significant

underrepresentation of this lineage, probably because of lack of infection with intestinal

helminths in specific pathogen-free mice 135,143,147. Classical Th2 cells develop in

response to parasitic helminths and possibly other extracellular microbes that stimulate

innate immune cells to produce IL-4, which trigger activation of STAT6 and induction of

the master transcription regulator GATA3 148,149. Fully differentiated Th2 cells then

induce production of IL-4, IL-5, and IL-13 effector cytokines, as well as expression of

chemokine receptors CCR3 and CCR4 133,150. Th2 cells can also affect systemic immunity

by mediating the pathology of allergic responses and asthma. In this regard, allergic Th2

responses to orally delivered peanut antigens are driven by IL-4 produced by activated

23

CD4 T cells. Although the specific LP MNP subset that promotes T cell-intrinsic IL-4

secretion is unknown, signaling through OX40L expressed on LP DCs appears to control

IL-4 production151. More recent reports propose a role for IRF4-dependent DCs in lung

Th2 differentiation and CD301b+ dermal DCs in Th2 skin immunity. Their intestinal DC

counterparts, however, are unclear152,153.

3. Th17 cells

i. Commitment to Th17 cell lineage

Th17 cells are important for clearance of extracellular pathogens and for

promotion of tissue repair via the signature cytokines IL-17A, IL-17F, and IL-22. Th17

lineage specific cytokines regulate granulopoiesis, recruit neutrophils, induce

antimicrobial peptide production, and promote anatomical containment of microbes 154-

156. In addition, intestinal Th17 cells exacerbate the severity of several organ autoimmune

diseases 157. Th17 cells develop in response to extracellular bacteria and fungi under the

combined action of IL-6 and TGFβ1 and self-maintain via autocrine signaling from IL-21

and IL-1β 147,158-161. Cytokine receptor engagement activates STAT3, which upregulates

the master transcriptional regulator RORγt 154,162 with help from several transcription

factors163 including IRF4 164,165, Batf 166, c-Maf167, Ahr 168,169, and RORα 170. During the

early phase of differentiation, Th17 cells secrete IL-21, while during the last phase they

upregulate IL-23 receptor and secrete the signature effector cytokines, IL-17A, IL-17F

and IL-22 171,172. Th17 cells co-express CCR6 and CCR4 173, which promote recruitment

to inflammatory sites. In experimental autoimmune encephalomyelitis (EAE), lack of

CCR6 on Th17 cells decreases severity of disease 174. Despite their enrichment at

24

intestinal sites, Th17 cells do not express unique gut homing receptors. In fact, exposure

to retinoic acid, the inducer of gut tropism, suppresses their fate and skews activated CD4

T cells towards a FoxP3+ regulatory T cell lineage 120.

ii. Th17 cell-inducing cytokines

Mice deficient in IL-6 or carrying a mutated form of TGFβ1 confirmed the

dominant role that these cytokines play in driving Th17 cell differentiation in vivo

158,159,175,176. However, IL-6 may not be absolutely required, because RORγt+ CD4 T cells

can develop in IL-6-/- mice 177. A definitive requirement for most other cytokines is

unclear. Despite the fact that IL-21 was reported to be necessary for optimal Th17 cell

differentiation ex vivo and in peripheral lymphoid organs 178, IL-21-/- animals display an

increased number of Th17 cells in the SI LP 175. Because IL-21 appears to promote IL-10

production, it is possible that lack of IL-21 results in enhanced Th17 cell levels due to

reduced suppression by IL-10 135. In addition, IL-21 is pathogenic in both acute and

chronic models of colitis 178-180. Although IL-23 is necessary for the late stages of Th17

cell development 154,181,182, absence of IL-23 does not impede Th17 cell differentiation at

steady state in SI LP 175, but it confers protection during infection with Citrobacter

rodentium in LI LP 159. IL-23 may decrease production of IL-22 by Th17 cells in the SI

175. In addition, IL-23 may play a role in Th17 cell development in LI, where its levels are

negatively regulated by IL-25. In this case, IL-25 deficiency leads to increased numbers

of Th17 cells through enhanced transcription of IL-23p19 183. Several reports converge

on a role for IL-23 in promoting development of pathogenic Th17 cells as various

methods of abrogating IL-23 levels or signaling reduce symptoms of colitis 184-187, EAE

25

188, psoriatic arthritis and psoriasis 189,190, and rheumatoid arthritis 191. In addition, IL-23

signaling promotes induction of IL-17A and pro-inflammatory chemokines by newly

differentiated Th17 cells without upregulating IL-10 production, reinforcing its role of

supporting pro-inflammatory responses 192. IL-23 also stimulates production of TGFβ3

from developing Th17 cells, which together with IL-6 subverts the differentiation

program towards pathogenic Th17 cells193. IL-23 is thought to promote GM-CSF

production by Th17 cells to confer pathogenicity during EAE 194,195 and colitis 196. In

these contexts, GM-CSF may act in a paracrine manner since GM-CSF-deficient T cells

can still induce disease 197.

iii. Microbe-mediated Th17 cell induction

In the steady state, Th17 cells preferentially reside in the SI LP, and to a lesser

extent in the colon162,198. Their development relies on the presence of intestinal

microbiota and they are specifically induced in the terminal ileum of mice colonized with

segmented filamentous bacteria (SFB) 147,175. Specific-pathogen free mice (SPF) mice

raised at The Jackson Laboratory carry no SFB and have very low levels of Th17 cells,

whereas certain SPF mice raised at Taconic farms are enriched for SFB and have high

Th17 cell levels 175. Germ-free mice and antibiotic-treated conventional mice have

decreased levels of Th17 cells in the SI LP 175,199. Colonization of germ-free or Jackson

SPF mice from with SFB recovers the levels of IL-17A and IL-22-secreting Th17 cells in

SI 147. We find that the cognate antigens for the majority of SFB-induced Th17 cells are

SFB-derived and are presented by LP CD11c+MHCII+ cells. In addition, SFB-mediated

Th17 cell differentiation can only be driven by host-specific SFB through adhesion to

26

host IECs200. Adhesion to intact, but not disrupted, IECs appears to be a prerequisite for

Th17 cell induction in both SI and LI LP by any microbe, including Citrobacter

rodentium, EHEC O157:H7, and Candida albicans200. It will be interesting to identify the

molecular signature of TH17 cells induced by non-mucosa associated commensals.

Microbiota-mediated TLR9 signaling has also been shown to promote

inflammatory Th17 cell induction in the small intestine 201. Specifically, stimulation of

TLR9-expressing LP DCs with CpG-rich motifs derived from commensal DNA leads to

increased generation of IFNγ and IL-17 producing effector T cells and concomitantly, to

decreased levels of pTregs within the SI GALT and LP 201. Hence, commensal-derived

DNA motifs trigger IL-6 production from TLR9-expressing LP DCs, and tip the balance

in favor of effector T cell differentiation. We and others identified a negative regulatory

loop between MHCII-expressing ILC3 cells and Th17 cells, which occurs independently

of SFB status. Specifically, we found that lack of MHCII on ILC3 in RORγtΔMHCII leads

to higher levels of commensal-dependent Th17 cells 202. Further studies revealed that

ILC3 cells limit commensal-specific CD4 T cell responses via MHCII interactions that

resemble thymic negative selection 203,204.

In the colon, Th17 cell levels increase with age and occur at normal levels in PP

and colonic patch-null mice 199. Furthermore, colonic Th17 cells are preferentially

induced in response to bacterial-derived ATP. This effect is mediated by colonic

CD70+CD11c+ cells, which express P2X and P2Y ATP receptors 199.

27

4. Regulatory T cells

Within the antigen-rich intestinal environment, effector Th cells run the risk of

mounting responses to self antigens and triggering inflammation and autoimmune

diseases. Thymus-derived natural regulatory T cells (tTregs) and peripherally-induced

Tregs (pTregs) represent populations of CD4 T cells that suppress overt effector Th

responses by inhibiting T cell proliferation. At steady state, tTregs and pTregs can be

distinguished through the expression of Helios and neuropilin1 (Nrp1) with pTregs

displaying low to undetectable levels of these markers, though, during inflammation such

as EAE, pTregs may also upregulate expression of Nrp1 205,206.

The master regulator of Tregs is the transcription factor FoxP3, which is induced

in naive CD4 T cells in the periphery by high levels of TGF-β207,208. Although both Th17

cells and FoxP3+ Tregs require TGFβ and retinoic acid during the initial steps of their

development, IL-6 and IL-21 inhibit TGFβ-induced upregulation of FoxP3 and skew

CD4 T cell differentiation towards Th17 cell fate. In addition, SI LP hosts FoxP3-

regulatory cells, Tr1-like cells 209. Both FoxP3+ pTregs and FoxP3- Tr1-like cells produce

large amounts of the anti-inflammatory cytokine IL-10 210,211. Although FoxP3+ pTregs

and IL-10-expressing CD4 T cells may have non-redundant functions, FoxP3ΔIL-10 and

FoxP3ΔIL-10Ra mice develop spontaneous colitis underlining an important role for IL-10-

IL-10Ra axis in regulating Foxp3 Treg function 212,213.

GF mice have reduced numbers of colonic pTregs, but not SI pTregs 214. Treg-

inducing potential is restricted to certain members of the commensal microbiota.

Exposure of GF mice specifically to 46 Clostridium strains is sufficient to drive

development of colonic Helios-FoxP3+ pTregs and attenuate the development of DSS-

28

induced colitis 214. Clostridium-driven pTreg induction does not require classical innate

immune signaling 214. Instead, it is mediated by short-chain fatty acids produced by

digestion of dietary fibers by Clostridium strains 215-217. In contrast, B. fragilis induces IL-

10 + Tregs via signaling through TLR2 expressed on FoxP3+ Tregs, independently of LP

DCs218.

V. Functions of intestinal mononuclear phagocytes

A. Antigen uptake

LP MNPs can pick up antigen transported across the intestinal epithelium or

directly from the lumen (Figure 1-3). One prevalent route of antigen sampling is

phagocytosis and transcytosis through specialized cells of the FAE of PPs known as M

cells 219. M cells uptake a wide variety of macromolecules including food, environmental

and microbial antigens through receptor mediated endocytosis or yet undefined

mechanisms that involve immunosurveillance receptors expressed on their apical surface,

such as glycoprotein 2, uromodulin, cellular prion protein and others 220. Following

transcytosis through the FAE, bulk antigens exit into the subepithelial dome, where

MNPs pick them up to present to nearby lymphocytes to induce mucosal immune

responses. In Spi-B-/- mice, which lack the ability to induce maturation of M cells,

antigen sampling is deficient and results in reduced expansion of antigen-specific T cells

in response to infection with Salmonella typhimurium 221,222.

A second route delivers luminal antigens directly to the LP via neonatal Fc

receptor (FcRn)-mediated bidirectional transport of IgG-antigen complexes 223. In this

case, FcRn binds IgG in the LP, transports it through the IEC basolateral surface to the

29

apical surface, allows it to form complexes with its cognate antigen in the lumen, and

transports the IgG-antigen complexes in the converse direction to CD11c+ LP cells for

antigen processing and presentation 223.

A third route focuses on the constant immunosurveillance of the lumen by

CD11c+ cells via extensions named transepithelial dendrites (TED), which establish tight

junction-like structures with neighboring IECs and preserve the villus brush border 224-226.

Rescigno et al also proposed that CD11c+ cells could employ TEDs to engulf exhausted

and apoptotic epithelial cells to facilitate the renewal of the epithelial barrier without

disrupting its integrity 224. Using CD11c and MHCII reporter mice, several reports

confirm the formation of epithelium-penetrating dendrites in the small intestinal LP,

however, they do not agree on the number and distribution of TEDs 225-227. Initial studies

suggested that CX3CR1 expression conferred CD11c+ cells the ability to form TEDs and

sample commensal and pathogenic bacterial antigens for mounting proper immune

responses 228. In addition, while TEDs form at steady state, their number increases in the

terminal ileum in the presence of invasive or non-invasive Salmonella typhimurium in a

MyD88-dependent manner suggesting that microbial signals drive increased TED

formation and luminal sampling 226. In CX3CR1gfp/gfp mice TED formation by CD11c+

cells and uptake of DsRed2-labeled non-pathogenic E.coli or non-invasive GFP-

expressing Salmonella typhimurium were impaired, while acquisition of antigens via M

cells was unaltered 228. However, the extent to which TED-mediated sampling contributes

to sensing of non-invasive microbes is not clear 225. CX3CR1 expression also does not

seem required for TED formation and TED formation has also been observed in the

absence of CX3CR1 in the proximal ileum suggesting that inflammatory stimuli or

30

antigen overload can bypass the requirement for CX3CR1 expression for initiation of

antigen sampling 226,229.

A fourth route relies on capture of low molecular weight soluble antigen by goblet

cells (GC) and delivery through goblet cell-associated passages (GAPs) to

CD103+CX3CR1- small intestinal DCs for induction of tolerogenic responses 227. Either

by making stable contact with or by actively probing GAPs, CD103+ LP DCs, but not

CD103- LP DCs, acquire soluble ovalbumin and cross-present it efficiently to OTI T cells

in vitro. GAP-mediated delivery of antigen is preferred over direct sampling of luminal

antigen as few transepithelial dendrites are observed after intraluminal injection of FITC-

dextran or ovalbumin 227. In addition, transfer of antigens between two different intestinal

MNP subsets to induce oral tolerance has been reported 230. CD11b+CX3CR1+ cells were

shown to extend dendrites to acquire luminal antigen, process it and deliver it to CD103+

DCs via connexin-mediated gap junctions 230. In the absence of gap junctions in

CD11cΔCx43 mice, orally-fed ovalbumin protein accumulates in SI CD11b+CX3CR1+

phagocytes, which process it and successfully present it to OTII CD4 T cells ex vivo.

Antigenic material from MHCII-expressing epithelial cells could be transferred to

LP DCs in the form of exosome-like vesicles named “tolerosomes” 231. These structures

form by fusion of MHCII-carrying vesicles with endosomes containing luminal antigenic

peptides and can reduce ear swelling in a delayed type hypersensitivity reaction 231.

31

Figure 1-3. Mechanisms of antigen sampling by intestinal MNPs.

A. CD64 Mfs take up luminal antigens, process them and transfer them to conventional

DCs via gap junctions for presentation. B. Both CD64 Mfs and CD24 DCs sample

antigens from the lumen through transepithelial dendrites. Any MNP population can

present the antigens and generate the appropriate cytokine milieu. C. IECs take up

antigens by transcytosis. Unprocessed antigens are released at the basolateral surface via

exocytosis or IEC-associated passages to MNP subsets. D. CD64 Mfs engulf dying IECs

and process the microbial antigens that had been endocytosed by these IECs.

B. Antigen processing and presentation

Most studies focus on antigen uptake and presentation but little is known about how and

where luminal antigens are processed for induction of mucosal immune responses.

Orally-delivered antigens are presumably heavily processed by different physiological

32

factors they encounter en route to the intestinal lumen by processes such as low pH,

digestive proteases, bile. This early processing can expose antigenic epitopes, which may

facilitate their recognition, endocytosis or transcytosis, and further cleavage by

specialized IECs and LP MNPs 232. With direct luminal antigen sampling, a reasonable

assumption is that LP CD11c+ cells are equipped to both acquire and process antigens

into peptides ready for display on MHCII molecules. In the case of antigen transfer from

CD11b+ LP Mfs to CD11b+ LP DCs, Mfs engulf ovalbumin protein, process it into

smaller peptides and deliver these peptides to LP DCs responsible for MHCII-dependent

presentation 230. In support, although CX3CR1gfp/+ LP cells take up more ovalbumin

protein in vivo than CD103+ DCs, they are less efficient at priming OTII CD4 T cells ex

vivo 229. In cases of antigen overload, insufficient communication with or absence of

CD103+ DCs, CX3CR1+ phagocytes can accumulate sufficient antigenic peptides, present

and prime CD4 T cells 230. Whether similar antigen pre-processing occurs at all in

specialized IECs prior to transfer to DCs it is not known. Of note, CD103+ SI DCs appear

to receive full AlexaFluor647-labeled ovalbumin protein or FITC-dextran via GAP-

mediated transfer from GCs, which suggests that GCs do not process the soluble antigens

they retrieve from the lumen 227. This finding confirms earlier studies of the ability of

CD8+DEC-205+ and CD8-33-D1+ splenic DC to process and present antigens, which

revealed their intrinsic bias towards MHCI-mediated cross-presentation and MHCII-

mediated presentation, respectively 233.

Intestinal MNP subsets differ in their ability to present antigens and prime CD4 T

and CD8 T cells, although similarities to CD8+DEC-205+ and CD8-33-D1+ splenic DCs

exist. Isolated CD103+CD11b- DCs have an enhanced capacity to cross-present antigens

33

to CD8 T cells, in vitro, similarly to their splenic counterpart of CD8+DEC-205+ DCs

78,227. In addition, both ex vivo and in vivo studies show that all CD11c+MHCII+ LP MNP

subsets can present antigen to CD4 T cells, albeit CD103+ DC subsets can do so more

efficiently 61,64,78,230. Abundant ex vivo work revealed preferential presentation of dietary

antigens by CD103+CD11b-, CD103+CD11b+, and some CD103- LP DCs 56,61,77,227, while

bacterial antigen presentation is performed more extensively by CD103+CD11b+, CD103-

CD11b+ LP DCs and CX3CR1+ Mfs 51,69,78,234-237. In addition, intestinal inflammation can

dictate a switch from tolerogenic to immunogenic antigen presentation for CD103+ DCs

238.

C. Priming of CD4 T cells

As highlighted above, intestinal MNP subsets play a crucial role in initiation of

CD4 T cell responses to luminal antigen. In the conventional response, MNPs become

loaded with antigen within the LP or PPs and migrate to MLNs, where they prime newly

arrived naive CD4 T cells for regulation of mucosal immune responses. A vast majority

of studies reinforces the role of conventional DCs in priming CD4 T cells, which is

related to their ability to migrate and bring antigens to MLNs. All LP DC subsets express

CCR7, which is a chemokine critically required for migration of LP APCs to MLNs 51,106.

Defective migration of DCs to MLNs in CCR7-/- mice results in impaired induction of

pTregs and oral tolerance 107. Of note, antigen presentation by DCs is more important

than innate immune activation mediated by DCs in mounting adaptive immune responses

239. ZBTB46-mediated deletion of MHCII on all DCs led to reduced numbers of pTregs,

lower levels of serum and fecal IgA and bacterial-mediated intestinal inflammation 239.

34

Among DCs, CD103+CD11b- LP DCs are most likely the main migratory subset that

drives FoxP3+ pTreg differentiation and tolerance to oral antigens within the MLNs 77.

As described above, a lot is known about the role of LP DCs in priming of CD4 T

cells, but the role of CD64 Mfs is less clear. Despite the low levels of CCR7 transcripts at

steady state, some CD64 Mfs have been retrieved from the MLNs and FoxP3+ pTreg

levels were reduced in MLNs of mice lacking monocytes and macrophages, MM-DTR 77.

While priming of pTregs was not impaired in MM-DTR BM chimeras, the decreased

pTreg levels suggest that antigen-loaded LP macrophages can either deliver cargo to LP

DCs or migrate to MLNs and deliver their cargo to MLN DCs77. In fact, evidence for

transfer of cargo between Mfs and DCs comes from gain-of-function experiments

performed in Mycobacterium tuberculosis (Mtb) infected animals 240. In this case, transfer

of MHCII-expressing CCR2+ inflammatory monocytes into MHCII-/- animals rescued the

transport of Mtb antigens to the MLN without recovering priming of antigen-specific

CD4 T cells 240. Interestingly, a similar mechanism may occur in the intestine in

antibiotic-treated animals infected with non-invasive pathogens 236. In this microbial

context, CX3CR1hi MNPs rather than CD103+DCs upregulate CCR7 expression and

deliver heat-killed, non-invasive Salmonella to the MLNs 236. These results support a role

for LP Mf migration to MLNs in cases of dysbiosis, where they may facilitate priming of

CD4 T cells by classical DCs.

Where does CD4 T cell priming occur? Colonic Th17 cells numbers are normal in

PP- and colonic patch-null, but MLN-sufficient mice, suggesting that colonic Th17 cell

priming does not require PPs199. We and others also showed that CD11c+MHCII+ cells

can prime SI Th17 cells in response to SFB in LTα-/- mice in the absence of GALT and

35

MLNs, suggesting that priming of mucosal Th17 cells can occur locally in the lamina

propria202,241. In addition, colonic FoxP3+ pTregs are represented in normal proportions

and are induced in normal numbers in LTα-/- mice, suggesting that colonic pTreg priming

does not require GALT and MLN242. Interestingly, priming of SI FoxP3+ pTreg is

defective in LTα-/- mice, hinting at different mechanisms of induction of SI and LI

pTregs242. Because both SI Th17 cells and LI FoxP3+ pTregs are induced by commensal

species (i.e. SFB and Clostridiales, respectively), an intriguing possibility is that

commensal-driven mucosal T cells can bypass conventional priming in GALT and MLNs

and instead, undergo priming locally in the lamina propria. As such, induction of certain

commensal-specific T cell responses would occur in a distinct location from priming of T

cells to oral antigens or pathogenic bacteria. It will be important to investigate whether

such differential priming (i.e. in the MLN versus the LP) is employed in responses to

different types of intestinal microbes (e.g., commensals vs pathogens or mucosa-adapted

commensals such as murine SFB, murine Clostridiales vs other luminal commensal

species).

D. Induction of pTreg differentiation for oral tolerance

Oral tolerance is a characteristic of intestinal immunity and refers to the local and

systemic immune unresponsiveness to innocuous antigens administered orally.

MLNs are sites of robust pTreg differentiation mediated by small intestinal CD103+ DCs

118,121, and particularly CD103 SP DCs 77. Intriguingly, MLNs seem to be required

differentially for the induction of FoxP3+Helios- Tregs in the SI compared to the LI.

36

Specifically, FoxP3+Helios- Tregs show impaired accumulation within the SI LP in LTα-/-

mice compared to normal accumulation in LI LP 242. Similarly, ova-fed LTα-/- mice fail to

induce FoxP3 pTreg differentiation of transferred OTII CD4+CD25- T cells in the SI but

not LI, suggesting that priming of LI pTregs occurs locally within the LI LP 242. Key to SI

pTreg differentiation and achievement of oral tolerance are migration of oral antigen-

loaded CD103+ DCs to the MLNs and imprinting of gut-homing receptors on activated T

cells by these DCs 243. Subsequently, pTregs home to the SI LP where they undergo local

proliferation in response to IL-10 production by CX3CR1+ macrophages and establish

intestinal tolerance 244. Interestingly, CCR7-dependent migration of intestinal DCs to the

MLNs may not be required for accumulation of normal numbers of pTregs within the SI

or LI LP 242. In fact, CCR7-deficiency appears to nearly double the proportions of pTregs

in both SI and LI LP. Because priming of SI pTregs is defective in the absence of MLNs,

it is possible that another chemokine receptor-ligand axis acts redundantly with CCR7-

CCL19/21.

Most studies investigate oral tolerance using ovalbumin as a model oral antigen,

however, other antigens such as tryptophan245 and curcumin246 have also been shown to

induce tolerance and anti-inflammatory FoxP3+ pTregs.

As mentioned previously, GCs can transfer Ova protein to CD103+ DCs via GAPs

for cross-presentation to OTI cells. In the process, CD103+ DCs also capture GC-derived

proteins, which may be used to prime Tregs and induce suppression of T cell responses to

self antigens 227. Thus, depending on the acquisition mode, CD103+ DCs may promote

both oral tolerance and anergy of self-antigen-specific T cells.

37

E. Induction of pTreg differentiation by commensals

Intestinal MNPs induce pTreg differentiation in response to bacterial stimuli, in

addition to diet-derived antigens. Work with human monocyte-derived dendritic cells

showed that direct contact of DCs with individual members of commensal flora (i.e.

Bacteroides and Bifidobacteria) induced DC maturation and a switch to a tolerogenic

fate, which the monocyte-derived DCs employed to instruct differentiation of IL-10

producing Tr1 cells 247. Similarly, MLN CD11c+ DCs loaded with B. fragilis-derived

PSA can drive the differentiation of a mixture of Th1 and IL-10 producing Tregs that

home to the colonic LP 143. Since commensals tend to adapt and evolve with the host, the

above examples expose pTreg development pathways that are induced in response to

non-host specific commensals. In this regard, 46 indigenous strains of Clostridiales can

consistently induce robust Helios-FoxP3+ IL-10-producing CD4 T cells within the colonic

LP that ameliorate colitis and reduce systemic IgE responses to OVA 214. Subsequently,

bacterial metabolites that lead to colonic pTreg conversion were described. Short-chain

fatty acids (SCFA), e.g. propionate and butyrate, produced during bacterial-mediated

fermentation of dietary fibers or starches induced extrathymic FoxP3+ IL-10-producing

Tregs 215-217. Administration of SCFA to GF mice was sufficient to induce upregulation of

IL-10 and FoxP3 expression by activated CD4 T cells and to ameliorate T cell-driven

colitis in lymphopenic mice 216,217. The role of intestinal MNP subsets was not

specifically investigated in these studies 215.

Muscularis macrophages (MM) have been recently shown to express a

transcriptional profile that resembles anti-inflammatory macrophages, suggesting they

may function in tissue repair and wound healing 17. Specifically, infection with non-

38

invasive Salmonella activates gut extrinsic sympathetic neurons, which trigger release of

norepinephrine and stimulate β2 adrenergic receptors expressed on MMs to support

transcription of Retnla, IL-10, Arg1 and Chi3l3 17. Whether the conversion of MMs into

anti-inflammatory macrophages leads to pTreg-mediated suppression of overt immune

responses and wound healing remains to be investigated. However, this finding provides

an example of communication between enteric neurons and resident macrophages, in

which neurons directly prime macrophages to perform anti-inflammatory functions in

response to certain microbes. Neurotransmitters may act on dendritic cells to confer

immunoregulatory potential to peripheral T cells. Vasoactive intestinal peptide (VIP)

produced by intestinal enteroendocrine and immune cells suppress LPS-induced

maturation of bone-marrow derived DCs and instead stimulate IL-10 secretion by

BMDCs 248. VIP-imbued DCs promote differentiation of IL-10 and TGFβ-secreting Tr1-

like cells in vitro 248. Administration of VIP to a mouse model of relapse-remitting EAE

leads to emergence of IL-10-producing FoxP3+CD4+CD25+ T cells from CD4+CD25- T

that can accumulate in the perivascular cuffs and suppress proinflammatory T cell

proliferation 249. Along the same lines, intravenous administration of BMDCs pre-treated

with VIP during chemically-induced colitis ameliorates diseases by suppressing Th1-

driven responses, stimulating IL-10 and TGFβ production and inducing Tr1-like cells 250.

F. Induction of Th17 cell differentiation

Which specific LP MNP subsets induce Th17 cell differentiation at steady state or

inflammation is a topic of intense research. Both LP DC and LP Mfs have been reported

to induce Th17 cell differentiation, however, these observations were largely based on

39

preferential induction of Th17 cells by sorted MNP subsets in vitro 64,66,199. Within the SI

LP, DP DCs have been highlighted as the population responsible for induction of Th17

cells both in vitro and in vivo 56,61,86,113. Endogenous Th17 cell levels were reduced in

CD11cΔNotch2 animals, in which DP DC development is impaired, as well as in

CD11cIRF4flox/- animals, in which DP DC survival and migration to MLNs are

defective56,61,86. Similarly, SI and MLN Th17 cells were reduced in an ablation model of

DP DCs (Langerin-DTA, 113). However, the role of DP DCs in initiation of microbe-

specific Th17 cell responses was not investigated. In addition, it was unclear which MNP

subsets generate Th17 cell polarizing cytokines and present antigens for induction of

microbe-specific Th17 cell differentiation.

DP DCs are necessary for development of Th17 cell differentiation in response to

pathogens or luminal commensals234,237. DP DCs promote Th17 cell differentiation

through stimulation of pattern recognition receptors (PRR) such as toll-like receptors

(TLR) 234,237. TLR5 ligation on SI DP DC (also defined as CD11chiCD11bhi cells) induces

Th17 cell responses to Salmonella typhimurium and protects against lethal infection with

the microbe234. In addition, Salmonella typhimurium flagellin stimulates IL-23 production

from TLR5-expressing DP DCs, which enhances innate immune production of defenders

IL-22 and RegIIIγ 235. TLR5 ligation on DP DCs by a commensal-derived flagellin

(CBir1 flagellin) leads to CBir1-specific Th17 cell differentiation237. Of note, in

conventional mice, CBir1 flagellin is heavily coated with IgA and does not translocate

through the intestinal epithelial barrier. However, CBir1 flagellin delivered

intragastrically to TCRβxδ-/- animals can penetrate into SI LP. CBir1 flagellin-treated

TCRβxδ-/- animals supported production of both IL-17 and IFNγ from adoptively

40

transferred CBir1 CD4 T cells, suggesting that this commensal induces a mixture of Th1

and Th17 cell differentiation237. During colitis, CD103+ MLN DCs lose their tolerogenic

properties and induce development of IFNγ and IL-17 producing CD4 T cells238.

Similarly, DP DCs induce IL-17-producing CD4 T cells in LI LP of colitic Mucin-2-

deficient mice through generating high levels of IL-6 251. In addition, during infection

with C. rodentium, CD11cΔNotch2 are susceptible to the pathogen and succumb to death

during the adaptive immune response phase 87. Similarly, mice lacking all DCs, Flt3l-/-

mice and Zbtb46-DTR BM chimeras, succumb to death during C. rodentium infection 87.

Thus, CD103+ DCs and particularly, DP DCs, may induce Th17 cell differentiation in

response to inflammatory microbe-driven gut environment.

Although present in small numbers in MLN and LP, CD103- DCs have been

reported to induce IFNγ and IL-17-producing CD4 T cells in the absence of overt

stimulation78. One subset of CD103- DCs, CCR2+CD103-CD11b+ DCs, was shown to

preferentially induce Th17 cell differentiation through production of high levels of IL-

12/IL-23p40 but not IL-6 66. Whether CD103- DCs are involved in microbial antigen

presentation for Th17 cell induction is not known. Furthermore, because they are GALT-

derived 78 it is unclear whether they preferentially function within the GALT or LP.

CD64 Mfs may also be involved in Th17 cell differentiation both during

infections and at steady state. While CD11cΔNotch2 mice are susceptible to infection with

C. rodentium, Langerin-DTA mice are resistant arguing against a role for DP DCs in

inducing Th17 cell immunity to the pathogen113. Cell-intrinsic Notch2 signaling is

necessary for upregulation of CD11c expression on CD11c+CX3CR1-GFPhi LP cells, and

LP Mfs promote secretion of IFNγ from Th1 and Th17 cells during C.rodentium

41

infection75,252. At steady state, commensals stimulate PRRs on SI CX3CR1+ Mfs, which

instruct the Mfs to secrete IL-1β but not IL-6 69,177,237,253. Secreted IL-1β appears

sufficient to drive SI Th17 cell development177. In addition, epithelial cell-derived IL-18

antagonizes IL-1R signaling on CD4 T cells and limits colonic Th17 cell

differentiation254. It is unclear whether a similar mechanism functions in SFB-driven

Th17 cell differentiation. SFB can induce Th17 cell differentiation in the absence of TLR

and IL-1R signaling147,199. In addition, colonic Mfs, defined as CD70+CD11c+ cells, can

induce Th17 cell differentiation in response to commensal-derived, but not pathogen-

derived, ATP199. In fact, intraperitoneal or rectal administration of an ATP analogue into

GF mice can rescue colonic Th17 cell levels to those observed in specific-pathogen free

mice (SPF)199. Overall, little is known about the role of MNP subsets in microbe-specific

Th17 cell differentiation.

VI. SFB-specific modulation of host immunity

A. Segmented filamentous bacteria (SFB)

SFB are the first example of a commensal species capable of modulating host

adaptive immune response at steady state. SFB are spore-forming gram-positive

anaerobic commensals that have segmented filamentous morphology and attach tightly to

IECs exclusively in the terminal ileum 255. SFB are currently unculturable, although

methods for their temporary propagation have been described256. Of note, colonization of

germ-free mice with in vitro-propagated SFB resulted in high levels of luminal SFB and

preferential attachment to caecal IECs rather than the terminal ileal IECs, which

42

underscores that additional optimization of the in vitro culture system is required. SFB

are widespread among vertebrates 257,258 and show host species-specific adaptations 259,260.

Mouse and rat SFB genomes contain up to 10% strain-specific genes and orthologous

proteins display 80% identity 261. Interestingly, single-cell sequencing revealed that even

host-specific SFB genomes display some polymorphisms suggesting that genetic

heterogeneity and potentially, distinct lineages/strains co-exist within a host 262. SFB

possess a reduced genome of only 1.57 Megabases. Comparative metabolic pathway

genome analysis reveals that SFB rely heavily on the host for basic metabolic functions,

evidence for a mutualistic relationship with the host 261,263. The SFB genome is enriched

for pathways that mediate communication with the environment and the host such as

processes involved in interacting, acquiring and transporting metabolites from the lumen

222.

SFB colonization of the terminal ileum appears to reach a maximum during the

time of weaning of mice and rats, after which their levels decrease rapidly264-266. The

fluctuation in SFB levels was presumed to be due to dynamic changes in the levels of

secretory IgA coated-SFB originating from mother’s milk and later, from self-produced

secretory IgA in the small intestine 264-266. The changes in SFB levels pre- and post-

weaning in addition to their sensitivity to general antibiotic cocktails hampered the

discovery of human SFB, especially since most human metagenome collections

employed for comparison contain samples of low age and geographical location diversity

222. More recent efforts aim to discover SFB in feces obtained from young children and

young adults and reinforce the significance of timing, diet and health status of the human

cohort 267,268.

43

B. SFB-mediated regulation of innate and humoral immune responses

SFB have multiple immunomodulatory effects. SFB promote accumulation of

intraepithelial lymphocytes and induce upregulation of MHCII molecules and

fucosylation of glycolipids on SI IECs 269-271. SFB also induce production of

antimicrobial peptides and stress response genes such as serum amyloid A proteins

(SAA) 1-3 from IECs147,202. In addition, SFB have been shown to stimulate production of

secretory IgA and enhanced generation of IgA+ plasma cells following interaction with

IECs 198,200. IgA is the most abundant immunoglobulin isotype at mucosal surfaces and it

mediates host intestinal immunity by coating pathogenic bacteria with high affinity 243.

In line with these observations, SFB stimulate robust development of PP germinal

centers. SFB can also drive development of tertiary lymphoid structures in the SI. Both

lymphoid structures are employed by SFB as inductive sites for specific and non-specific

gut IgA and Th17 cell responses272. SFB expand in the proximal third of the SI in the

absence of secretory IgA in mice deficient for activation-induced cytidine deaminase

(AID), which suggests that secretory IgA coating inhibits SFB attachment to the

duodenum273. SFB-induced IgA+ plasma cells display different specificity and

diversification profiles compared to E.coli -induced IgA plasma cells 272, which may

contribute to the ability of SFB to modulate host response to infections with pathogenic

bacteria. Secretory IgA coating of SFB is similar to that of bacterial drivers of colitis (i.e.

Prevotellaceae, Helicobacter species). Specifically, it is the result of high affinity,

antigen-specific, T-cell dependent antibody responses. SFB can exacerbate the

development of rheumatoid arthritis and EAE 274,275. High secretory IgA coating of SFB

44

may contribute to regulating their levels and preventing overt inflammation. In contrast,

most other commensal species are coated with low affinity IgA induced in a T-cell

independent process 276.

C. Innate recognition of SFB

Microbes engage the innate immune system through pattern-recognition receptors

(PRRs). PRRs recognize microbe-associated molecular patterns (MAMPs), such as

lipopolysaccharide, flagellin, bacterial DNA, viral RNA, and fungal β-glucan. PRRs have

crucial roles in protecting against pathogens, without inducing inflammation to gut-

resident commensals. PRRs fall into four categories: toll-like receptors (TLRs), NOD-

like receptors (NLRs), C-type lectin receptors (CLRs), and retinoic acid-inducible gene I

(RIG-1)-like receptors (RLRs). TLRs are expressed either on the cell surface or in

endosomes. TLRs activate adaptor molecules such as myeloid differentiation primary

response 88 protein (MYD88) and TIR domain containing adaptor protein inducing IFNβ

(TRIF). NLRs and RLRs are cytoplasmic proteins. NLRs signal through adaptor protein

receptor-interacting protein kinase (RIPK), which activates similar downstream

molecules to MyD88. CLRs are activated in response to fungi, whereas RLRs are

involved in intracellular virus recognition. The PRR involved in SFB recognition remains

unknown.

Disruption of TLR or NLR signaling in Myd88-/-Trif-/-, MyD88-/-Tlr3-/-, Trif-/-, and

Ripk2-/- mice does not impede Th17 cell development suggesting that SFB mediate their

effects independent of TLR and NLR signaling 175,199. Commensal-derived ATP can

induce colonic Th17 cell differentiation199. However, SFB-monoassociated mice have

45

lower levels of ATP in ileal and colonic contents than specific-pathogen free (SPF) mice,

suggesting ATP is not required for SFB-mediated Th17 cell differentiation. It is possible

that different PRR pathways act redundantly to facilitate Th17 cell differentiation by

SFB. Thus, in the absence of all MyD88-dependent pathways, Ripk2-dependent

mechanisms may take their place. However, absence of one PRR may not be sufficient to

signal activation of redundant mechanisms and thus, Th17 cell differentiation would be

impaired. In this regard, MyD88-dependent TLR9 sensing of commensal DNA alters the

ratio of effector T cells to pTregs in the SI. It is not clear, however, whether this innate

recognition affects SFB-mediated Th17 cell differentiation201. Overall, these results

would suggest no requirement for IL-1β-IL-1R axis, which also signals through MyD88,

but reduced levels of Th17 cells have been reported in MyD88-/- mice 177. In this case, IL-

1β was necessary and sufficient for Th17 cell differentiation as total LP cells from IL-1β-

/- or IL-1R-/- BM chimeras secrete low levels of IL-17A and IL-22 ex vivo 177. In addition,

microbiota drive constitutive and high production of IL-1β by SI LP MNPs in

conventionalized B6 mice, which correlates with normal levels of RORγt+IL-17A+ CD4

T cells 200,277. Wildtype BALB/c mice, which produce less IL-1β than wildtype B6 mice,

show reduced accumulation of Th17 cells in the presence of attached SFB 121.

SFB attach tightly to IECs in the terminal ileum and trigger cytoskeletal

rearrangements within the host epithelial cells278,279. Whether these rearrangements

played a role in innate sensing of SFB was unclear. In addition, it was not known which

innate immune cells sense SFB moieties for generation of cytokine environment and/or

antigen presentation for Th17 cell induction.

46

VII. Commensal-specific effector T cell responses

Commensals drive development of distinct effector CD4 T cell lineages as GF

mice have low numbers of Th17 cells in the SI LP and FoxP3+ pTregs in the LI LP,

respectively147,214,280. Development of effector CD4 T cell responses requires both

MHCII-mediated presentation of cognate antigen and an effector lineage-skewing

cytokine environment. Prior to my thesis study, much was known about the role of

cytokine environment in induction of effector T cell responses, but it was unclear whether

microbial antigen presentation was necessary. In addition, the geography in which

priming and differentiation of commensal-specific effector CD4 T cells occurs had not

been examined.

Antibiotic treatment depletes even the low levels of Th17 cells observed in SFB-

negative mice, suggesting that most if not all Th17 cells in the SI LP are induced by

microbiota175,281. Most of the ensuing reports proposed that the commensal community or

specific species induced cytokine environments conducive for development of Th17 or

pTreg cells. Microbiota-induced cytokine environment was sufficient to activate naïve

OTII CD4 T cells and drive Th17 cell differentiation in the presence of cognate antigen

in vitro and ex vivo 56,61,64,66. In fact, LP Th17 cells developed even in the absence of

cognate antigen, in Marilyn transgenic (Tg) mice, which recognize an MHCII-restricted

self-male antigen, and in TCR7 Tg mice, which recognize an epitope from hen egg

lysozyme 282. Commensal-derived DNA motifs preferentially induced Th17 cells via

TLR9 stimulation201. In the colon, commensal-derived ATP drove colonic Th17 cell

differentiation199. Direct ligation of TLR2 on effector CD4 T cells by B. fragilis PSA led

47

to induction of IL-10-producing FoxP3+ pTregs283. In addition, Clostridial species-

derived butyrate and propionate promoted differentiation of FoxP3+ pTregs215-217. In the

case of SFB, SFB-induced SAA proteins facilitated Th17 cell differentiation147.

Specifically, co-culture of splenic CD4 T cells with LP CD11c+ in the presence of

recombinant SAA protein induced Th17 cell-specific transcriptional changes in activated

CD4 T cells. These changes phenocopied the cytokine profile of in vivo activated Th17

cells through production of IL-17A, IL-17F, IL-22, IL-21 but not IFNγ 147. These

findings suggested that cytokine environment was necessary and potentially sufficient for

induction of Th17 and pTreg differentiation.

Whether microbiota induces microbe-specific T cell responses was not fully

understood. In this regard, it was demonstrated that most of the TCRs carried by colonic

pTregs had a distinct repertoire than those of effector FoxP3- T cells or of FoxP3+ Tregs

isolated from spleen or MLNs284. These TCRs were proven to be specific for bacterial

antigens. In addition, pTreg TCRs were associated with suppressive immune responses as

transfer of these TCRs onto effector CD4 T cells induced colitis. Whether these

commensal-specific TCRs induce the development of tTregs is unclear284,285.

Nevertheless, this finding raised the possibility that in addition to exposure to non-self

antigens in the thymus, effector CD4 T cells were instructed in the periphery with non-

self antigens derived from commensals. In support of peripheral education of T cells,

FoxP3+RORγt+ CD4 T cells did not develop in SI LP of Marilyn and TCR7 Tg mice in

the absence of cognate antigen282.

Despite the fact that SI Th17 cells developed to normal levels in Marilyn and

TCR7 Tg animals, it was unclear whether SFB-induced Th17 cells recognize SFB

48

moieties in conventional mice at steady state. In general, it was unclear whether

commensal-specific T cells developed and were maintained in the mucosa. The mucosal

epithelium acts like a firewall preventing translocation of luminal commensals into the

lamina propria. However, gastrointestinal infections with invasive pathogens such as

Toxoplasma gondii damage the epithelial barrier and lead to translocation of commensal

bacteria into the LP146,286. In fact, infection with this pathogen led to development of

effector Th1 cells specific for commensal Cbir1 flagellin along with those specific for

pathogen-derived antigens142. During infection, the CBir1-specific Th1 cells were

protective by facilitating clearance of the pathogen. However, CBir1-specific Th1 cells

persisted long-term in the SI and showed rapid re-activation upon secondary encounter

with CBir1 antigen leading to development of immune responses against the commensal.

These findings proposed the existence of naïve commensal-specific CD4 T cells at steady

state, which can become activated during acute infections. These results also raised the

question whether commensal-specific responses developed against all commensal

antigens or only towards the most abundant or accessible antigens. In addition, the impact

that commensal-derived antigens could have on local and systemic responses over time

was unclear.

In the studies mentioned above, LP MNP cells play a role in generating a Th1,

Th17 or pTreg-conducive cytokine environment or in delivering antigens for priming of

commensal-specific CD4 T cells. However, it was not explored whether specific LP

MNP subsets are required for commensal-specific mucosal T cell responses. The precise

location of commensal antigen presentation and priming of CD4 T cells was also not

49

investigated. Furthermore, it is unclear whether commensal-specific effector T cells differ

from pathogen-specific T cells at the molecular level.

50

CHAPTER TWO

Segmented filamentous bacteria antigens presented by intestinal dendritic cells drive mucosal Th17 cell differentiation

Yoshiyuki Goto1,#, Casandra Panea1,#, Gaku Nakato1, Anna Cebula2, Carolyn Lee1, Marta Galan Diez1, Terri M. Laufer3, Leszek Ignatowicz2, & Ivaylo I. Ivanov1 1Department of Microbiology and Immunology, Columbia University Medical Center, New York, NY 10032, USA

2Center for Biotechnology and Genomic Medicine, Georgia Regents University, Augusta, GA 30912, USA

3Department of Medicine, Perelman School of Medicine, University of Pennsylvania and Philadelphia Veterans Affairs Medical Center, Philadelphia, PA 19104, USA #These authors contributed equally to this work

Corresponding author: Ivaylo I. Ivanov ([email protected])

Citation: Goto, Y., C. Panea, et al. (2014). "Segmented filamentous bacteria antigens presented by intestinal dendritic cells drive mucosal Th17 cell differentiation." Immunity 40(4): 594-607.

51

SUMMARY

How commensal microbiota contributes to immune cell homeostasis at barrier surfaces is

poorly understood. Lamina propria (LP) T helper 17 (Th17) cells participate in mucosal

protection and are induced by commensal segmented filamentous bacteria (SFB). Here

we show that MHCII-dependent antigen presentation of SFB antigens by intestinal

dendritic cells (DCs) is crucial for Th17 cell induction. Expression of MHCII on CD11c+

cells was necessary and sufficient for SFB-induced Th17 cell differentiation. Most SFB-

induced Th17 cells recognized SFB in an MHCII-dependent manner. SFB primed and

induced Th17 cells locally in the LP and Th17 cell induction occurred normally in mice

lacking secondary lymphoid organs. The importance of other innate cells was unveiled by

the finding that RORgt+ innate lymphoid cells (ILCs) strongly inhibited SFB-independent

Th17 cell differentiation in an MHCII-dependent manner. Our results outline the complex

role of DCs and ILCs in the regulation of intestinal Th17 cell homeostasis.

52

RESULTS

A. Microbiota-induced intestinal Th17 cells are selected on MHCII

SFB preferentially increase Th17 cell proportions in the SI LP (147 and Figure 2-S1A). To

examine the dependence of intestinal microbiota-induced Th17 cells on MHCII we

analyzed intestinal CD4 T cell homeostasis in MHCII-deficient mice (IAb-/-). IAb-/-

animals have greatly decreased numbers of CD4 T cells due to lack of selection on

MHCII. However, a small subset of CD4 T cells is still present in spleen and lymph

nodes 287,288. The small and large intestinal lamina propria of IAb-/- animals contained a

substantial CD4 T cells population (Figure 2-1A). The CD4 T cell population in IAb-/-

animals contained an expanded proportion of Foxp3+ Treg cells in all tissues examined

(Figure 2-1A and 2-S1B). In contrast, SI LP Th17 cells were decreased in SFB-positive

IAb-/- animals compared to WT controls (Figure 2-1A). To examine whether Th17 cells in

IAb-/- mice are controlled by SFB we compared T cell subsets in the presence and absence

of SFB. Th17 cell numbers varied greatly in IAb-/- mice (Figure 2-1B). However, in

contrast to WT mice, SFB-positive IAb-/- mice did not have significantly increased Th17

cell percentages in the gut (Figure 2-1B). Therefore, even though Th17 cells can be

generated in the absence of MHCII, SFB do not induce further Th17 cell differentiation

in MHCII-deficient animals.

IAb-/- mice possess diverse, but quite distinct CD4 T cell repertoire 289. In order to

examine if MHCII expression is required for SFB-mediated induction of WT Th17 cells,

we performed adoptive transfer experiments. We first established that adoptively

53

transferred WT CD4 T cells develop into Th17 cells in the SI LP in an SFB-dependent

manner. Indeed, donor WT CD4 T cells were easily detected in the SI LP of SFB-

negative and SFB-positive recipients, but acquired IL-17 expression only in SFB-positive

recipients, similarly to endogenous CD4 T cells (Figure 2-1C,D). Purified naïve CD4 T

cells also generated Th17 cells only in the SI LP of SFB-positive hosts (Figure 2-S1C,D).

We next transferred WT CD4 T cells into SFB-positive WT and IAb-/- recipients. WT

CD4 T cells were detected in large numbers in the SI LP of IAb-/- recipients but did not

generate any Th17 cells even in the presence of SFB (Figure 2-1E,F). Identical results

were obtained after transfer of purified naïve CD4 T cells (Figure 2-S1C,D). Collectively,

these data demonstrate that the induction of LP Th17 cells by SFB requires MHCII

expression in the periphery.

B. SFB-induced intestinal Th17 cells recognize SFB antigens

SFB presence does not promote Th17 cell differentiation in MHCII-deficient CD4 T

cells, suggesting that SFB provide more than just specific cytokine environment. We,

therefore, examined SFB effects on Th17 cell induction in two TCR Tg models – OTII

and TRP-1, which recognize peptides from chicken ovalbumin (OVA) and mouse

tyrosinase related protein 1 (TRP-1) respectively 290. We first examined OTII mice on a

RAG-sufficient background. These mice contained large number of CD4 T cells in the

LP and a high proportion of these cells expressed IL-17 in SFB-positive mice, and even

in the absence of OVA (Figure 2-S2A). In contrast to spleen and MLNs, CD4 T cells in

the gut contained a large fraction of non-Tg TCRs as demonstrated by the low proportion

54

of Va2hiVb5hi Tg CD4 T cells (Figure 2-S2B). We therefore used this selection against

transgenic TCRs in the LP to compare the re-programming of these cells to Th17 cells in

comparison to cells expressing alternative TCRs. As shown on Figure 2-S2A, Va2hiVb5hi

Tg CD4 T cells expressed very little IL-17, compared to the remaining CD4 T cells, even

after activation with the cognate antigen (OVA). In agreement with this result, virtually

all IL-17 expressing cells in the LP, expressed alternative TCRs, demonstrating an

exclusion of Tg TCRs at the expense of endogenously formed TCRs with broad antigenic

specificities in the Th17 cell subset (Figure 2-S2C). As a result, purified intestinal IL-17+

CD4 T cells from OTII.B6 Tg mice responded equally well to OVA and to SFB antigens,

in sharp contrast to lymph node CD4 T cells, which responded to OVA only (Figure 2-

S2D). These results demonstrate that the intestinal Th17 cell population in OTII.B6 Tg

mice is enriched for non-Tg specificities, due to favorable co-expression of non-

transgenic TCRs. They also suggest that non-SFB Tg T cells, e.g. OTII cells, are not

efficiently induced into the Th17 cell lineage by SFB.

To more directly examine the effects of SFB on non-SFB Tg T cells, we

examined TCR Tg animals on a RAG-deficient background, which lack alternative

endogenous TCRs. In both, OTII.RAG and TRP-1.RAG Tg mice small numbers of Tg T

cells were present in the SI LP in the absence of the cognate antigen, but none of these

cells expressed IL-17 even after SFB colonization (Figure 2-2A and 2-S2E,F). LP Tg T

cells were activated and expanded following administration of cognate antigen, which

also led to induction of effector Th1 and Th17 cells (Figure 2-2A and 2-S2E,F).

However, even in the presence of the cognate antigen, SFB colonization did not induce

further conversion of Tg T cells into Th17 cells (Figure 2-2A). Moreover, SFB

55

colonization did not induce Th17 cell differentiation of TRP-1.RAG Tg CD4 T cells

transferred into WT mice separately or together with WT CD4 T cells, despite

considerable expansion and activation of the Tg T cells and despite the presence of

endogenous SFB-induced Th17 cells and induction of Th17 cell differentiation in co-

transferred WT cells (Figure 2-S3A,B). Combined, these experiments demonstrate that

SFB-conditioned intestinal environment is not sufficient to induce IL-17 expression in all

activated CD4 T cells.

We next examined whether SFB-induced Th17 cells preferentially respond to

SFB. We isolated CD4 T cells from SI LP of SFB-positive and SFB-negative WT mice

and compared their response to SFB antigens ex vivo. Purified SI LP CD4 T cells from

SFB-positive WT mice responded to SFB antigens, while SI LP CD4 T cells isolated

from SFB-negative mice, did not (Figure 2-2B). In contrast, SI LP CD4 T cells from

SFB-positive and SFB-negative mice did not respond significantly to a number of non-

SFB bacteria, including Gram-negative E. coli, Gram-positive Clostridium perfringens,

and cultured murine intestinal isolates (Figure 2-2C), demonstrating that LP CD4 T cells

from SFB-positive animals are specifically enriched for SFB reactivities. The SFB-

specific response required antigen presenting cells and MHCII expression, because

purified WT SI LP CD4 T cells from SFB-positive mice did not respond to SFB antigens

in the absence of DCs or when co-cultured with MHCII-deficient DCs as antigen

presenting cells (Figure 2-2D and data not shown).

To investigate directly the response of gut Th17 cells, we purified GFP+ (Th17)

and GFP- (non-Th17) CD4 T cells from the SI LP of SFB-colonized Il17GFP reporter mice

(Figure 2-S3C) and stimulated them in vitro with various bacterial lysates. Th17 cells

56

responded strongly to SFB, while non-Th17 cells did not respond to SFB above

background (Figure 2-2E,F). The response of purified Th17 cells was specific to SFB,

because the same cells did not respond significantly to cultured bacteria (Figure 2-2F). In

response to SFB, all proliferated GFP+ cells continued to express IL-17 (GFP) (Figure 2-

S3G), in contrast to the small proportion of proliferated GFP- cells, which remained

mostly IL-17- (Figure 2-S3G). Furthermore, LP Th17 cells did not respond to lysates

prepared from feces of germ-free (GF) or SFB-negative conventionally raised mice (SPF)

that contained similar numbers of total bacteria (Figure 2-2F) confirming that the

response is SFB-specific. To examine whether the response is directed towards an

antigen from SFB, as opposed to an SFB-induced host protein, we prepared lysates from

SFB filaments purified by cell sorting (Figure 2-S3D). LP Th17 cells responded to sorted

SFB filaments, while non-Th17 cells did not (Figure 2-S3E). We conclude that LP Th17

cells respond to SFB-derived protein antigens. To examine whether the SFB-specific

response in intestinal Th17 cells is directed by the presence of SFB, we purified GFP+

(Th17) cells from SI LP of Il17GFP reporter mice before and after SFB colonization and

examined their response to SFB. In contrast to Th17 cells isolated from SFB-positive

mice, Th17 cells from SFB-negative mice did not proliferate in response to the same SFB

antigen preparation (Figure 2-2G). Similar results were obtained when Th17 cells were

isolated on the basis of RORγt expression from SFB-positive and SFB-negative RorcGFP

reporter mice (Figure 2-S3F). These results demonstrate that Th17 cells from SFB-

positive, but not from SFB-negative, mice preferentially recognize SFB antigens and are,

therefore, enriched for SFB specificities.

57

C. Most lamina propria Th17 cells recognize SFB antigens

The in vitro co-culture experiments showed that LP Th17 cells from SFB-positive mice

respond to SFB antigens. However, the strong proliferative response in these experiments

can be due to expansion of a small subset of clones within the starting Th17 population.

To more directly quantify the proportion of LP Th17 cells that recognize SFB antigens

we decided to query the TCR specificities of individual cells in the total LP Th17

population. To this goal we generated a collection of T cell hybridomas from GFP+

(Th17) and GFP- (non-Th17) SI LP CD4 T cells, isolated from SFB-positive mice, and

examined the response of individual clones to SFB and control bacteria. As shown on

Figure 2-3A, 43 out of 94 hybridomas from LP Th17 cells, or 46%, responded to SFB. In

contrast, only 3% of the non-Th17 cell hybridomas (3 out of 96) responded to SFB

(Figure 2-3A). Most Th17 cell hybridomas responded strongly to SFB (50-100% of the

maximum anti-CD3/anti-CD28 stimulation) and did not respond to E. coli or C.

perfringens antigens (Figure 2-3B and data not shown). In comparison, the three positive

hybridomas from non-Th17 cells responded weakly (5-35%), though specifically, to SFB

antigens (Figure 2-3B). Taking into account that Th17 cells from SFB-negative mice do

not respond to SFB antigens and that there is 5-7 fold Th17 cell increase upon SFB

colonization, the hybridoma results show the presence of at least 55-57% SFB-reactive

cells in the SFB-induced Th17 cell population. This response was diverse and polyclonal,

as demonstrated by the sequences of SFB-responsive hybridomas (Table 2-S1). 14 out of

15 sequenced SFB-responsive hybridomas had unique TCRb CDR3 junctions with

diverse lengths (Table 2-S1). SFB-induced Th17 cells in vivo, as well as Th17 cell

58

hybridomas used a wide range of Vb gene segments, though we did note a relative

abundance of Vb14 TCRs (Figure 2-S4 and Table 2-S1). We, therefore, conclude that

most SFB-induced Th17 cells recognize SFB antigens and this SFB-specific Th17 cell

response is diverse and polyclonal.

D. MHCII expression on DC is necessary and sufficient for induction of Th17 cells

by SFB

MHCII expression in the periphery is required for SFB-mediated Th17 cell induction.

We, therefore, next investigated the nature of the participating MHCII-expressing cells

that present SFB antigens. Several types of MHCII-expressing cells exist in the LP.

These include professional APCs, DCs and B cells, as well as other cell types, such as

intestinal epithelial cells (IECs) and ILCs. To investigate the role of professional APCs,

we first examined Th17 cell induction by SFB in the absence of B cells. Th17 cell

induction after SFB colonization occurred normally in Cm-deficient mice 291, which lack

mature B cells (Figure 2-S5H), demonstrating that B cells are not required for SFB-

mediated Th17 cell induction.

To investigate whether MHCII-dependent antigen presentation by intestinal DCs

is required, we generated mice with DC-specific ablation of MHCII (DCΔMHCII mice), by

inter-crossing IAbflox mice 292 with CD11c-Cre mice 293 (Figure 2-4). All major LP

CD11c+ subsets were present in DCΔMHCII mice in similar numbers, however they

completely lacked MHCII expression (Figure 2-4A and Figure 2-S5A). In contrast,

MHCII expression was present on CD11c-negative cells (Figure 2-4A), including IgA+

59

plasma cells and B220+ B cells, though it was reduced on the latter (Figure 2-S5B).

In the absence of SFB, DCΔMHCII mice had slightly increased Th17 cell percentage

over control littermates (Figure 2-4B,C). SFB colonization induced robust Th17 cell

differentiation in control animals. In contrast, SFB colonization did not lead to a

statistical increase in Th17 cells in DCΔMHCII mice, even though SFB numbers and

attachment were similar to the littermate controls (Figure 2-4B,C and Figure 2-S5F,G).

Ablation of MHCII on DCs did not affect the expression of major Th17-inducing

cytokines [e.g. IL-6, IL-23, transforming growth factor-b (TGF-b), IL-1b], or the

induction of a number of other cytokines and enzymes, such as SAA1, SAA3, and iNOS,

by SFB (Figure 2-4D and 2-S5C), demonstrating lack of major changes in the cytokine

milieu. We conclude that MHCII expression on intestinal DCs is required for induction of

Th17 cells by SFB. Consistent with our previous studies 147, SFB colonization did not

affect Th1 or Treg cell proportions in WT or DCΔMHCII mice (Figure 2-S5D,E). Treg cell

proportions were increased in the intestinal LP of DCΔMHCII mice, though, as previously

reported 294, they were decreased in spleen and MLN, suggesting that MHCII expression

on DCs may have different roles in Treg cell control in gut versus secondary lymphoid

organs (Figure 2-S5E).

To examine whether MHCII expression on DCs is sufficient for induction of

Th17 cells by SFB, we examined Th17 cell induction in mice in which MHCII

expression is restricted to CD11c+ cells (MHCIICD11c mice) 295. Because MHCIICD11c

mice lack MHCII expression on thymic epithelium and are deficient in CD4 thymic

positive selection, we performed adoptive transfers. As in previous experiments, transfer

of WT CD4 T cells led to an SFB-dependent Th17 cell induction in the SI LP, which was

60

abrogated in the absence of peripheral MHCII expression in MHCII-deficient recipients

(Figure 2-4E,F). Transfer of WT CD4 T cells into littermate MHCIICD11c mice led to

considerable induction of Th17 cells in the transferred cells, demonstrating that recovery

of MHCII expression only on CD11c+ cells is sufficient to promote Th17 cell induction

(Figure 2-4E,F).

Combined, these results show that MHCII expression on intestinal DCs is

necessary and sufficient for SFB-mediated induction of Th17 cells.

E. MHCII expression on ILCs controls intestinal Th17 cells

Several other non-conventional antigen-presenting cell subsets express MHCII in the

intestine. These include IECs and ILCs 203,296. MHCII expression on IECs has an

unknown function and is controlled by commensal bacteria 270. Notably, MHCII

expression on IECs was induced very specifically by SFB only in the terminal ileum

(Figure 2-5A and data not shown). To investigate the role of IEC MHCII expression in

SFB-mediated Th17 cell induction we generated IECΔMHCII mice, in which MHCII was

deleted only on IECs by crossing IAbflox mice with Villin-Cre mice 297 (Figure 2-S6A,C).

Colonization of SFB-free IECΔMHCII mice with SFB (Figure 2-S6A,B) led to induction of

Th17 cells, similar to that in littermate controls (Figure 2-5B,C), demonstrating that

MHCII expression on IECs is not required for this process.

We also discovered that a subset of intestinal ILCs express MHCII (Figure 2-5D),

as also reported recently 203. A large proportion of Lin-RORgt+ ILC3 cells express

MHCII (Figure 2-5D). MHCII expression was most prevalent in c-kit+NKp46-RORgt+

61

ILCs, and on only a small fraction of NKp46+ or c-kit- ILCs (Figure 2-5D and 2-S6D). To

examine the role of MHCII expression on ILCs in SFB-mediated Th17 cell induction we

generated ILC3ΔMHCII mice, in which MHCII was deleted only on RORgt+ ILCs by inter-

crossing MHCII-floxed mice with RORgt-Cre mice 298 (Figure 2-5D and 2-S6D,E). All

three ILC3 subsets were present in ILC3ΔMHCII mice (data not shown). MHCII expression

was completely ablated on Lin-NKp46-c-kit+RORgt+ LTi-like cells and significantly

decreased in the small subset of MHCII+ cells within the remaining two ILC3 subsets

(Figure 2-5D and 2-S6D). ILC3ΔMHCII mice did not demonstrate any signs of rectal

prolapse or intestinal inflammation in our colony and MHCII deletion on ILC3s did not

affect the percentage of SI LP Tregs (Figure 2-S6F,G). Surprisingly, in contrast to SFB-

free control littermates, which had low numbers of Th17 cells, ILC3ΔMHCII animals

contained high percentage and numbers of SI LP Th17 cells even in the absence of SFB

(Figure 2-5E,F). SFB-free ILC3ΔMHCII mice contained as many Th17 cells as SFB-

colonized WT littermates (Figure 2-5F). Colonization of WT mice with fecal bacteria

from SFB-negative ILC3ΔMHCII animals did not induce Th17 cells, arguing against an

outgrowth of other Th17 cell-inducing bacteria (Figure 2-S6H). In contrast to Th17 cells

in SFB-positive WT animals, Th17 cells in SFB-negative ILC3ΔMHCII mice did not

respond to SFB antigens in vitro (Figure 2-S6I). Colonization with SFB induced further

increase in both percentages and total numbers of Th17 cells in 9-week old ILC3ΔMHCII

mice (Figure 2-5F and Figure 2-S6J,K). In agreement with our observation that SFB

induce SFB-specific Th17 cells, SFB colonization induced Th17 TCR repertoire changes

in ILC3ΔMHCII mice, such as the induction of Vb14+IL-17+ CD4 T cells (Figure 2-5E and

2-S6L), and a response to SFB antigens in vitro (Figure S6I). Collectively, these results

62

demonstrate that Th17 cells are increased in SFB-negative ILC3ΔMHCII mice, but SFB are

still capable of inducing Th17 cells in these animals.

F. Th17 cell induction by SFB does not require LN or organized GALT

Our results show that SFB antigens are presented by iDCs in the context of MHCII to

induce SFB-specific Th17 cells. To examine the site of Th17 cell priming in WT mice we

analyzed the kinetics of SFB-mediated CD4 T cell proliferation and Th17 cell

differentiation in different tissues following adoptive transfer (Figure 2-6). CD4 T cells

were purified from spleens and LNs of Il17GFP reporter mice, labeled with proliferation

dye and transferred into congenic WT recipients before or after SFB colonization.

Proliferation was scored by dye dilution and Th17 cell differentiation by induction of

GFP (IL-17) expression at different time points post transfer. A small number of

proliferating transfer cells were first detected in the SI LP at Day 3 after transfer (Figure

2-6A). The number of proliferating cells increased by Day 5 and some of those produced

IL-17, again only in the SI LP, but not in spleen or MLN. By Day 7, transferred cells in

SFB-colonized animals proliferated robustly and differentiated into Th17 cells in the SI

LP. T cell proliferation and Th17 cell induction was dependent on the presence of SFB

and was very low in SFB- animals (Figure 2-6A). In contrast, despite being present in

larger numbers, very few transferred cells proliferated in the MLN at Day 7 and none

expressed IL-17 (Figure 2-6A). Albeit lower than the SI LP, proliferation and Th17 cell

induction was also observed in Peyer’s Patches (PPs) starting at Day 6 (Figure 2-S7).

Further increase in proliferation and Th17 cell differentiation of transferred cells in the SI

63

LP was observed at 2 weeks post transfer (Figure 2-6B and data not shown). However,

we did not detect significant proliferation or IL-17 expression in MLNs, iLNs, or spleen

of SFB-colonized animals at any time-point, suggesting that SFB priming and induction

of Th17 cells occurs in the small intestine itself (Figure 2-6B and data not shown).

To investigate directly whether organized GALT is required, we examined the

induction of Th17 cells by SFB in lymphotoxin-a (LTa)-deficient mice. Lta-/- mice

possess a defect in generation of secondary lymphoid organs and lack PPs and isolated

lymphoid follicles (ILFs) in the intestine, as well as peripheral lymph nodes, including

MLNs. Lta-/- animals also lack LP B cells 299. Despite these defects, induction of Th17

cells by SFB, including induction of Vb14+IL-17+ cells, was unimpeded in Lta-/- mice

(Figure 2-7), demonstrating that organized GALT is not required for this process and

confirming that LP B cells are also dispensable. Therefore, sampling of SFB antigens can

occur outside Peyer’s Patches or MLNs and Th17 cell induction does not require priming

in peripheral lymph nodes, suggesting that iDCs acquire SFB antigens and prime CD4 T

cells locally in the LP.

Our results shed further light into the complex interactions involved in controlling

intestinal Th17 cell homeostasis. Antigen presentation by MHCII plays central role in the

induction of Th17 cells by commensal microbiota. Intestinal DCs acquire antigens from

Th17 cell-inducing bacteria to promote antigen-specific Th17 responses locally in the

lamina propria. At the same time, RORgt+ ILCs control excessive Th17 cell responses by

inhibiting Th17 cell differentiation also in an MHCII-dependent manner. Further

characterization of this process, such as identification of the participating SFB antigens or

the involvement of specific subsets of DCs or ILCs, will help in better understanding the

64

molecular and cellular mechanisms involved in the host-commensal crosstalk regulating

T cell homeostasis in the gut.

65

EXPERIMENTAL PROCEDURES

Mice

MHCII-floxed (IAbF/F), Cd11c-Cre, Vil-Cre, Cm-deficient, Il17GFP, TRP-1.RAG1 and

Lta-/- mice were obtained from Jackson Laboratory. Rorc-Cre mice 298 were a gift from

Dan Littman (NYU). MHCII-deficient (IAb-/-), OTII.B6 and OTII.RAG1 mice were

obtained from Taconic Farms, the latter through the NIAID Exchange Program 300.

MHCIICD11c mice (also known as CD11c-Abb mice) were previously described 295. All

mice were bred and housed under specific pathogen-free conditions at Columbia

University Medical Center. To control for microbiota and caging effects all experiments

were performed with littermate control and gene-deleted animals housed in the same

cage.

SFB colonization and Th17 cell induction

Bacterial genomic DNA isolation from fecal pellets and quantitative PCR for the SFB

16S rRNA gene were performed as previously described 147. SFB colonization was

performed by oral gavage with fecal pellets from SFB-monocolonized mice or with fecal

pellets from SFB-negative Jackson B6 mice colonized with feces from SFB-

monocolonized mice unless otherwise noted. Control mice were gavaged with fecal

pellets from SFB-negative littermates. SI LP Th17 cell induction was assessed 2-3 weeks

after colonization.

Activation of TCR Tg T cells

To activate TCR Tg T cells, OTII.B6 and OTII.RAG Tg mice were fed 1% OVA protein

66

in the drinking water for 12 days. In addition the mice were orally gavaged with 50 mg of

OVA protein (OVA, grade V; Sigma) on Day 1,3, and 5. TRP-1.RAG Tg mice were

immunized i.p. with 50µg TRP1 peptide (CGTCRPGWRGAACNQKILTVR, 92%

purity, Biomatik) in DPBS and 10µg LPS (Sigma-Aldrich) on Day 1 and 7. Control

animals were immunized only with 10µg LPS in DPBS.

Lamina propria cell isolation and adoptive transfers

Lamina propria lymphocytes, intracellular cytokine staining, and Foxp3 staining were

performed as previously described 175. For adoptive transfers, 5-10 x 106 MACS-purified

CD4 T cells (Miltenyi Biotec; 95-98% purity) or 5 x 106 FACS sorted

TCRb+CD4+CD62LhiCD44lo naïve T cells (BD Influx cell sorter) were transferred

intravenously into Ly5.1 WT recipients before or 10-14 days after SFB colonization.

Th17 cell induction in transferred cells in different tissues was assessed 2 weeks after

transfer unless otherwise noted. In some experiments cells were labeled with CellTrace

Violet proliferation dye (Life Technologies).

In vitro co-culture experiments

LP TCRb+CD4+ T cell subsets were purified by cell sorting and labeled with CellTrace

Violet proliferation dye. 5 x 104 CD4 T cells were co-cultured in 96-well U-bottom plates

with either 5 x 104 MACS purified splenic CD11c+ cells or 2 x 105 total TCRa-KO

splenocytes as APCs in the presence or absence of autoclaved bacterial lysates. T cell

proliferation was assessed 72 hours later by dye dilution. For SFB antigens, SFB

filaments were purified from feces of SFB-monocolonized mice. Briefly, individual fecal

67

pellets from SFB-monocolonized mice were homogenized in PBS. The supernatant was

cleared from debris by several low-speed centrifugations and bacteria were pelleted by

centrifugation at 4,000g and washed twice with PBS. After the final wash the pellet was

resuspended in PBS and layered onto 60% w/v Nycodenz density gradient. SFB filaments

were collected at the interphase and the procedure repeated. Finally, the SFB filaments

were washed twice in PBS. SFB and other bacterial antigens were prepared by

autoclaving bacterial suspensions and used at 1:200 dilution.

Hybridoma generation and screening

FACS purified SI LP TCRb+CD4+GFP+ and TCRb+CD4+GFP- T cells were stimulated in

vitro for 3 days in tissue culture plates coated with 5ug/ml each of anti-CD3 and anti-

CD28 mAbs, fused with BW5147 thymoma 301 and plated in limited dilutions in selective

media. Individual clones were picked 10 days later and expanded in 24-well plates. The

response of cloned hybridomas towards autoclaved bacterial lysates was measured using

the HT-2 assay 302. In brief, 105 hybridoma cells were stimulated with plate-bound

aCD3/aCD28 (positive control) or incubated with 105 bone-marrow-derived dendritic

cells (or splenocytes from TCRα-KO mice) alone (no antigen control), or with SFB, E.

coli, or Clostridium perfringens lysates. After 24  h, the amount of secreted IL-2 was

measured with the detector HT-2 cell line. The proliferation of HT-2 cells in response to

IL-2 was measured with the MTT (Sigma) assay 302, and the response for each

hybridoma was plotted as percentage from the IL-2 response in the aCD3/aCD28-

stimulated positive control.

68

Statistics

Significance was scored by using unpaired two-tailed t test unless otherwise noted. P

values were represented on figures as follows: ns, p ≥ 0.05, * p < 0.05, ** p < 0.01, *** p

< 0.005, **** p < 0.001, ***** p < 0.0005, ****** p < 0.0001. Error bars on all figures

represent standard deviation of the mean.

69

ACKNOWLEDGEMENTS

We thank members of the I.I. laboratory for their help in all aspects of the project. We

thank Dr. Yoshinori Umesaki at the Yakult Central Institute for providing feces from

SFB-monocolonized mice. We thank Kenya Honda and Seiko Narushima for developing

and sharing the SFB sorting protocol. We thank Steve Reiner and Boris Reizis for

reading the manuscript and for invaluable scientific discussions. We thank Siu-Hong Ho

in the Columbia Center for Translational Immunology Flow Cytometry Core and Amir

Figueroa at the Department of Microbiology and Immunology Flow Cytometry Core for

help with sorting. This work was supported by the National Institute of Health

1R01DK098378 to I.I. and by the Crohn’s and Colitis Foundation of America

SRA#259540 to I.I. G.N. was supported by a long-term research grant from Toyobo

Biofoundation. I.I. is a Pew Scholar in the Biomedical Sciences, supported by the Pew

Charitable Trust.

70

FIGURES

Figure 2-1. -Induction of intestinal Th17 cells by SFB requires MHCII expression in

the periphery

A. Th17 and Treg cell proportions in SI LP of SFB-positive WT and MHCII-deficient

(IAb-/-) mice. Foxp3 and cytokine staining plots are gated on TCRb+CD4+ cells

B. Th17 and Treg cell proportions in SI LP of SFB-negative (Jackson microbiota) and

SFB-positive (Taconic microbiota) WT and IAb-/- mice. Plots gated on TCRb+CD4+ cells

C-D. WT CD45.1+ CD4 T cells were adoptively transferred into WT CD45.2 mice before

or 12 days after SFB colonization. Cytokine expression in host (CD45.2+) and donor

(CD45.1+) SI LP TCRb+CD4+ cells 2 weeks after transfer. Data from one of multiple

experiments

71

E-F. Th17 cell induction in WT CD4 T cells two weeks after transfer into SFB-positive WT and

MHCII-deficient recipients. Plots gated on TCRb+CD4+ cells. Data from one of multiple

experiments.

Figure 2-2. SFB-induced intestinal Th17 cells preferentially respond to SFB antigens

A. Th17 cell proportions in the SI LP of OTII.RAG and TRP-1.RAG TCR Tg mice

before and after SFB colonization in the absence or presence of cognate antigen.

Representative data from 5 independent experiments

B-C. Proliferation response of sorted SI LP TCRb+CD4+ cells from SFB-negative (Jax)

and SFB-positive (Tac) WT B6 mice to SFB (B,C) or other bacterial antigens (C). T cell

proliferation was scored by dye dilution on Day 3. Ec, E. coli, Cp, Clostridium

72

perfringens; MIB, mouse intestinal bacteria (cultured isolates from feces of SFB-negative

(Jackson) mice); “-“ – no antigen. Representative data from 5 independent experiments

D. SI LP TCRb+CD4+ cells were purified from SFB-negative (No SFB) and SFB-positive

(SFB+) WT mice and co-cultured with SFB antigens as in (B) and WT or IAb-/- DCs. Data

from 2 independent experiments

E-F. SI LP GFP+ (Th17) and GFP- (non-Th17) TCRb+CD4+ cells from SFB-positive

Il17GFP mice were stimulated in vitro with SFB (E,F) or various bacterial antigens (F) as

in (B) or with lysates from germ-free (GF) or SFB-negative SPF (SPF) animals.

Representative data from multiple experiments

G. SI LP GFP+ (Th17) and GFP- (non-Th17) TCRb+CD4+ cells from SFB-positive

(SFB+) or SFB-negative (No SFB) Il17GFP mice were stimulated in vitro with SFB

antigens as in (B). Representative data from 2 independent experiments

73

Figure 2-3. Most intestinal SFB-induced Th17 cells recognize SFB

T cell hybridomas were generated from SI LP GFP+ (Th17) and GFP- (non-Th17) CD4 T

cells from SFB-positive Il17GFP mice. Data combined from 2 independent experiments

A. Number of hybridomas responding to SFB

B. Response of individual hybridomas (percentage of maximum anti-CD3 and anti-CD28

stimulation) to SFB or E. coli antigens as assessed by IL-2 production. Clones were

ordered in decreasing amounts of IL-2 production.

74

Figure 2-4. DC expression of MHCII is necessary and sufficient for SFB-mediated

Th17 cell induction

A. SI LP lymphocytes from DCΔMHCII and control littermates. Left panels, gated on

TCRb-CD4- cells. Right panels, gated on CD11c+ cells

B-C. Th17 cell induction in DCΔMHCII mice and control littermates 2 weeks after SFB

colonization. Plots gated on TCRb+CD4+ cells. Representative data from 4 independent

experiments

D. Relative cytokine expression (RT-PCR) in terminal ileum of DCΔMHCII and control

littermates (WT) 2 weeks after colonization with SFB. nd – below threshold of detection

E-F. Th17 cell differentiation of WT CD45.1+ CD4 T cells in the SI LP 2 weeks after

transfer into SFB-positive IAb+/-, IAb -/-, and IAbCD11c CD45.2+ recipient littermates. Data

75

from 2 independent experiments

Figure 2-5. RORgt+ ILCs inhibit differentiation of SFB-independent intestinal Th17

cells through MHCII

A. MHCII expression on IECs in Jackson B6 mice before and 2 weeks after SFB

colonization. Arrows point to SFB filaments attaching to IECs

B-C. Th17 cell induction in IECΔMHCII mice and control littermates 2 weeks after SFB

colonization. Plots gated on TCRb+CD4+ cells. Data from one of 2 independent

experiments

D. Expression of MHCII on SI LP c-kit+NKp46-RORgt+ group 3 ILCs in ILC3ΔMHCII

mice and control littermates

76

E-F. Th17 cell induction in ILC3ΔMHCII and control littermates 2 weeks after SFB

colonization. Plots gated on TCRb+CD4+ cells. Data from 2 independent experiments

Figure 2-6. Priming and induction of Th17 cells by SFB occurs in the small intestine

CellTrace Violet labeled CD45.2+ CD4 T cells from Il17GFP mice were transferred into

WT CD45.1+ recipients before (No SFB) or after (+SFB) SFB colonization

A-B. Proliferation (A,B) and Th17 cell induction (A) at indicated time points. Plots are

gated on CD45.2+TCRb+CD4+ transferred cells. SI LP, small intestinal LP; MLN,

mesenteric lymph nodes; SPL, spleen. Combined data from 3 independent experiments.

77

Figure 2-7. SFB induce Th17 cells in the absence of secondary lymphoid organs

SI LP lymphocytes were isolated from Lta-/- and control littermates 2 weeks after

colonization with SFB

A-C. Th17 and Vb14+IL-17+ cells induction in TCRb+CD4+ cells. Representative data

from one of 2 independent experiments

D. Foxp3+ Treg cell proportions in TCRb+CD4+ cells. Combined data from 2 independent

experiments

78

SUPPLEMENTAL INFORMATION

RT-PCR primers:

IL-1b Fw: 5'-CAACCAACAAGTGATATTCTCCATG-3' Rev: 5'-GATCCACACTCTCCAGCTGCA-3'

IL-6 Fw: 5'-CCACTTCACAAGTCGGAGGC-3' Rev: 5'-TGCAAGTGCATCATCGTTGTTC-3'

IL-17A Fw: 5'-GGACTCTCCACCGCAATGA-3' Rev: 5'-GGCACTGAGCTTCCCAGATC-3'

IL-21 Fw: 5'-AAGATTCCTGAGGATCCGAGAAG-3' Rev: 5'-TGCATTCGTGAGCGTCTATAGTG-3'

IL-23 p19 Fw: 5'-CTGGAACGCACATGCACCAG-3' Rev: 5'-TGTTGTCCTTGAGTCCTTGTGG-3'

TGF-b Fw: 5'-CACTGATACGCCTGAGTGGC-3' Rev: 5'-TGCTGTCACAAGAGCAGTGAG-3'

SAA1 Fw: 5'-CATTTGTTCACGAGGCTTTCC-3' Rev: 5'-GTTTTTCCAGTTAGCTTCCTTCATGT-3'

SAA3 Fw: 5'-CGCAGCACGAGCAGGAT-3' Rev: 5'-CCAGGATCAAGATGCAAAGAATG-3'

NOS2 Fw: 5'-ATGCTGCCACCTTGGAGTTCAC-3' Rev: 5'-GGCCACCCACCTCCAGTAGC-3'

Reg3g Fw: 5'-CCTGATGCTCCTTTCTCAGG-3' Rev: 5'-ATGTCCTGAGGGCCTCTTTT-3'

79

SUPPLEMENTAL FIGURES

Figure 2-S1. SFB-induced intestinal Th17 cells require MHCII expression in the

periphery

A. Cytokine expression in SI LP TCRβ+CD4+ cells in 10-week old SFB-monocolonized

mice (SFB-mono) and germ-free (GF) controls or in SFB-negative conventionally-raised

mice (SPF) before and 2 weeks after colonization with SFB

B. Foxp3 expression in CD4 T cells isolated from the corresponding tissues of SFB-

positive WT and IAb-/- mice. Plots gated on TCRβ+CD4+ cells

C-D. 5x106 sorted naïve CD4 T cells were transferred into WT and IAb-/- mice before or

after colonization with SFB. IL-17 expression was examined in transferred cells in the SI

80

LP 2 weeks after transfer. Plots gated on donor TCRβ+CD4+ cells

81

82

Figure 2-S2. SFB do not induce Th17 cell differentiation of non-SFB Tg T cells

A. IL-17 expression in SI LP CD4 T cells from SFB-positive OTII.B6 TCR Tg mice

before or after stimulation with cognate antigen. OVA stimulation was performed by

supplying 1% OVA protein in the drinking water ad libitum for 10-14 days. Left,

Vα2hiVβ5hi plots are gated through the gate shown in (B). Right, IL-17 expression in

CD4 T cells outside the Vα2hiVβ5hi gate

B. Expression of the Tg Vα2 and Vβ5 in CD4 T cells isolated from the corresponding

tissues of SFB-positive OTII.B6 TCR Tg mice in the absence of OVA stimulation. Plots

are gated on TCRβ+CD4+ cells. Vα2hiVβ5hi CD4 T cells include cells expressing the Tg

TCR

C. SI LP Th17 cells express alternative TCRs. Vα2 and Vβ5 expression in IL-17+ vs IL-

17- CD4 T cells in the SI LP of SFB-positive OVA.B6 Tg mice. All plots initially gated

on TCRβ+CD4+ cells

D. SI LP Th17 cells and lymph node CD4 T cells from OTII.B6 Il17GFP mice were

labeled with CellTrace Violet (SI) or CFSE (LN) and stimulated in vitro with TCRα-

deficient splenocytes and OVA peptide or SFB lysates for 3 days. Plots gated on

TCRβ+CD4+ cells. Top, GFP+ (IL-17+) CD4 T cells purified by FACS from SI LP.

Middle, CD4 T cells were purified from MLNs by MACS using anti-CD4 magnetic

beads. Bottom, Splenic CD4 T cells from TRP-1.RAG Tg mice were purified by MACS

using anti-CD4 magnetic beads, labeled with CFSE, and stimulated in vitro with TRP-1

peptide or SFB antigens for 3 days in the presence TCRα-deficient splenocytes

E-F. SFB-free OTII.RAG or TRP-1.RAG TCR Tg mice were colonized with SFB-

containing microbiota from Taconic B6 mice. Colonization levels were confirmed by Q-

83

PCR. 2-4 weeks later SFB-negative and SFB-positive mice were stimulated in vivo with

the corresponding antigen

E. OTII.RAG mice were stimulated with OVA by oral gavage at Day 1, 3 and 5 and by

providing 1% OVA in the drinking water ad libitum. SI LP cells were isolated on Day

12-14. RAG-sufficient SFB-positive control mouse is shown for comparison

F. TRP-1.RAG mice were immunized i.p. with 50 µg TRP-1 peptide and 10 µg LPS on

Day 1 and 7. Control mice received 10 µg LPS only. SI LP cells were isolated on Day

12-14.

Figure 2-S3. SFB do not induce Th17 cell differentiation of non-SFB Tg T cells

A. 5x106 CD45.2+ CD4 T cells were purified from spleens and peripheral LNs of TRP-

84

1.RAG mice by MACS using anti-CD4 magnetic beads and transferred into CD45.1+

recipients before or 12-14 days after colonization with SFB (Day 0). Recipient animas

were immunized i.p. with 50 µg TRP-1 peptide and 10 µg LPS on Day 1 and 7. Control

mice received transferred CD4 T cells and 10 µg LPS only. Th17 cell induction was

examined in transferred Tg (CD45.2+) and endogenous WT (CD45.1+) cells 12-14 days

after transfer

B. 5x106 CD45.2+CD90.1+ WT CD4 T cells and 5x106 CD45.2+CD90.2+ TRP-1 Tg CD4

T cells were purified from spleens and peripheral LNs of WT and TRP-1.RAG mice

respectively, combined and co-transferred into WT CD45.1+CD90.2+ recipients before or

12-14 days after colonization with SFB (Day 0). Recipient animas were immunized i.p.

with 50 µg TRP-1 peptide and 10 µg LPS on Day 1 and 7. Th17 cell induction was

examined in three different types of cells in the same animal 12 days after transfer. H -

endogenous host WT cells (CD45.1+CD90.2+); W – transferred WT cells

(CD45.2+CD90.1+); T - transferred Tg cells (CD45.2+CD90.2+)

C. GFP and IL-17 expression in total SI LP cells (left and middle panel) and in CD4 T

cells (right panel) isolated from the SI LP of heterozygous Il17GFP reporter mice

D. Sorting scheme for isolation of purified SFB filaments. Feces from SFB-

monocolonized mice were processed as described in Methods and stained with the

SYTO9 component of the Live/DeadRBacLightTM Bacterial Viability Kit (Life

Technologies). SFB filaments were identified by SYTO9 staining and large size (SSC-W)

and sorted on FACSAriaII (BD) to high purity. Representative FACS plots and photos

show the presence of mostly large SFB filaments and absence of small bacterial cells or

host/fecal debris after sorting. PBS only, negative control of sterile PBS used to

85

resuspend the SFB sample after sorting. Arrowheads, SFB filaments of variable size in

the sample used for sorting

E. GFP+ (Th17) and GFP- (non-Th17) TCRb+CD4+ cells were purified by FACS from SI

LP of SFB-positive Il17GFP reporter mice and stimulated in vitro with SFB lysates

prepared after density gradient (SFB Gradient) or from SFB filaments purified by cell

sorting on panel C (SFB Sorted)

F. GFP+ (Th17) and GFP- (non-Th17) TCRb+CD4+ cells were purified by FACS from SI

LP of SFB-positive (SFB+) or SFB-negative (No SFB) RorcGFP reporter mice and

stimulated in vitro with SFB antigens

G. IL-17 expression in proliferated (CellTracelo) GFP+ and GFP- SI LP CD4 T cells from

Figure 2F stimulated in vitro with SFB antigens for 3 days. Plots are gated on live,

proliferated T cells

86

Figure 2-S4. SFB induce Th17 cells with diverse Vβ utilization

A. SFB-negative Jackson B6 mice (Jax) were colonized with SFB by oral gavage with

SFB-containing Taconic feces (Jax + SFB) and IL-17 induction in different Vb families

was examined by flow cytometry 14 days after colonization. Plots represent percentage

of IL-17+ cells in the TCRβ+CD4+Vb+ subset for the corresponding Vb.

B. Vb and IL-17 expression in SI LP CD4 T cells of germ-free and SFB-monocolonized

C57BL/6 mice. Plots represent percentage of IL-17+ cells in the TCRβ+CD4+Vb+ subset

for the corresponding Vb. ND, not determined

87

Figure 2-S5. Effects of SFB colonization in DCΔMHCII mice

A. CD11c+ cell subsets in the SI LP of DCΔMHCII and control littermates. Bar plots

represent total numbers or percentage of the corresponding subset in the CD11c+ gate

B. MHCII expression in APC subsets from SI LP of WT and DCΔMHCII mice

C. Relative expression of SFB-induced genes in terminal ileum of DCΔMHCII and control

littermates (WT) before and after colonization with SFB determined by RT-PCR. ns, not

significant

D. Th1 cells in the SI LP of DCΔMHCII and control littermates before and after SFB

colonization

E. Foxp3 expression in CD4 T cells from the indicated organs. SFB-negative DCΔMHCII

88

mice and control littermates were colonized with SFB by oral gavage and Foxp3+ Tregs

analyzed 12-14 days later. Plots represent the percentage of Foxp3+ cells in the

TCRb+CD4+ population. p values compare DCΔMHCII mice to corresponding WT

littermate group

F. SFB levels in fecal pellets from DCΔMHCII and control littermates 3 weeks after SFB

colonization

G. SFB attachment and MHCII expression in terminal ileum epithelial cells in WT and

DCΔMHCII mice colonized with SFB. Note lack of MHCII expression in LP of the

DCΔMHCII littermate. Blue, DNA; Green, MHCII; Red, Actin

H. SFB-negative Cµ-deficient and WT control mice were obtained from Jackson

Laboratory. The mice were co-housed for a week and colonized with SFB by gavage with

fecal homogenates from SFB-monocolonized mice. Control animals were gavaged with

fecal homogenates from SFB-negative mice. SFB absence or colonization was confirmed

by RT-PCR. Th17 cell induction was examined 12 days after gavage.

89

Figure 2-S6. SFB-mediated Th17 cell responses in IECΔMHCII and ILC3ΔMHCII mice

A. MHCII expression on cell subsets from SI LP of WT and IECΔMHCII mice

B. SFB levels in feces of IECΔMHCII and control littermates assessed by 16S rRNA gene

RT-PCR

C. MHCII expression in terminal ileum of IECΔMHCII mice and WT littermates 2 weeks

after colonization with SFB. MHCII expression is observed only in the LP in IECΔMHCII

mice. Arrows point to SFB filaments, attaching to IECs

D. MHCII expression on RORgt+c-kit+NKp46+ (R2) and RORgt+c-kit-NKp46- (R3) ILCs

in WT and ILC3ΔMHCII mice. R2 and R3 gates as shown on Figure 5D

E. MHCII expression on cell subsets from SI LP of WT and ILC3ΔMHCII mice

F. Normal colonic histology and lack of intestinal inflammation in ILC3ΔMHCII mice

90

G. Foxp3+ Tregs in SI LP of ILC3ΔMHCII mice and control littermates before and after

colonization with SFB

H. WT SFB-negative mice were gavaged twice with fecal homogenates from SFB-

negative ILC3ΔMHCII mice (with high levels of SI LP Th17 cells). IFNg and IL-17

expression in SI LP CD4 T cells was examined 3 weeks after gavage

I. CD4 T cells were purified by cell sorting from SI LP of SFB-positive and SFB-

negative WT and ILC3ΔMHCII mice and incubated in vitro with SFB or other bacterial

lysates as in Figure 2. T cell proliferation was examined on Day 3 of culture

J-K. SFB colonization in feces and attachment to terminal ileum villi in ILC3ΔMHCII mice

and control littermates

L. Total numbers of Vb14+IL-17+ Th17 cells in ILC3ΔMHCII mice and control littermates

before and after colonization with SFB

91

Figure 2-S7. SFB priming of CD4 T cells in gut mucosa

107 MACS-purified CD4 T cells from spleens and LNs of CD45.2+ Il17GFP mice were

labeled with CellTrace Violet proliferation dye and adoptively transferred into WT

CD45.1+ mice before or 12 days after SFB colonization. T cell proliferation (dye

dilution) and Th17 cell induction (GFP expression) was examined at different time points

in small intestinal lamina propria (SI LP) and Peyer’s Patches (PP). Very low

proliferation and no Th17 cell induction was detected in SFB-negative animals (not

shown)

92

Table 2-S1. Diverse TCR repertoire of SFB-recognizing hybridomas

CDR3 regions and Vb utilization in TCRb chains, sequenced from 15 SFB-recognizing

hybridomas. Clone numbers correspond to clone numbers on Figure 2-3B. The sequences

are arranged by Vb usage (right-most column). The two identical sequences are

highlighted in red

93

CHAPTER THREE

Intestinal monocyte-derived macrophages control commensal-specific Th17

responses Casandra Panea1, Adam M. Farkas1, Yoshiyuki Goto1, Shahla Abdollahi-Roodsaz1,#, Carolyn Lee1, Balázs Koscsó2, Kavitha Gowda2, Tobias M. Hohl3, Milena Bogunovic2 & Ivaylo I. Ivanov1,� 1Department of Microbiology and Immunology, Columbia University Medical Center, New York, NY 10032 2Department of Microbiology and Immunology, College of Medicine, Pennsylvania State University, Hershey, PA 17033 3Infectious Diseases Service, Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065

#Present Address: Department of Medicine, New York University, New York, NY 10003

�Corresponding author: Ivaylo I. Ivanov ([email protected])

Citation: Panea, C., et al. (2015). "Intestinal Monocyte-Derived Macrophages Control Commensal-Specific Th17 Responses." Cell Rep 12(8): 1314-1324.

94

Summary

Generation of different CD4 T cell responses to commensal and pathogenic bacteria is

crucial for maintaining healthy gut environment, but the associated cellular mechanisms

are poorly understood. Dendritic cells (DCs) and macrophages (Mfs) integrate microbial

signals and direct adaptive immunity. Although the role of DCs in initiating T cell

responses is well appreciated, how Mfs contribute to the generation of CD4 T cell

responses to intestinal microbes is unclear. Th17 cells are critical for mucosal immune

protection and at steady state are induced by commensal bacteria, such as segmented

filamentous bacteria (SFB). Here, we examined the roles of mucosal DCs and Mfs in

Th17 induction by SFB in vivo. We show that Mfs, and not conventional CD103+ DCs,

are essential for generation of SFB-specific Th17 responses. Thus, Mfs drive mucosal T

cell responses to certain commensal bacteria.

95

RESULTS

We recently showed that commensal Th17 cell induction is mediated by the antigen-

presenting function of CD11c+MHCII+ MNPs in the small intestinal (SI) LP 202. To

characterize the role of different MNP subsets in this process we examined Th17 cell

induction by SFB following genetic ablation. Four major CD11c+MHCII+ MNP subsets

were followed throughout this study using the gating strategies in Figure 3-1A and 3-

S1A. Conventional CD103+ LP DCs consist of gut-specific CD103+CD11b+ DCs (DP

DCs) controlled by the transcription factors Notch2 and IRF4 56,61,86 and CD103+CD11b-

DCs (CD103 SP DCs) that require BATF3 and IRF8 for their development 303. The

remaining CD103-CD11b+ MNPs express the chemokine receptor CX3CR1 and consist

predominantly of intestinal Mfs, which were identified based on expression of CD64 and

F4/80 304, and a smaller population of CD64-F4/80- MNPs, that express intermediate

levels of the monocyte/Mf marker CX3CR1, but also express the DC markers CD24 and

CD26 (Figure 3-S1B). Although CD103-CD11b+CD64-CD24+ cells may represent a

phenotypically and developmentally heterogeneous population 305-307, we refer to them

here as CD11b single positive DCs (CD11b SP DCs).

A. DP DCs are dispensable for Th17 cell induction

DP DCs have been shown to promote Th17 cell differentiation in vitro 307. In addition,

we and others have shown a decrease in LP Th17 cell numbers in mice with genetic

deficiency of DP DCs, suggesting a role for this MNP subset in vivo 56,61,86,308. However,

the specific role of DP DCs in microbiota-mediated induction of Th17 cells has not been

96

examined. To this end, we colonized DP DC-deficient mice and wildtype (WT)

littermates, with SFB and examined Th17 cell induction and induction of SFB-specific

CD4 T cells in the SI LP.

Langerin-DTA mice 309 express diphtheria toxin (DT) under transcriptional

control of the human Langerin promoter resulting in selective ablation of epidermal

Langerhans cells as well as DP DCs in the SI LP (Figure 3-1A,B, Table 3-S1 and 308).

Migratory DP DCs were also absent in MLN of Langerin-DTA mice (Figure 3-1C,D).

Colonization of WT littermates with SFB led to induction of RORgt+ and IL-17+ (Th17)

CD4 T cells in the SI LP (Figure 3-1E-J). In addition, SFB colonization resulted in

induction of SFB-specific CD4 T cells as demonstrated by the enrichment of Vb14+ Th17

cells 202,281 (Figure 3-1G,J) and by the response of purified SI LP CD4 T cells to SFB

antigens in vitro (Figure 3-1K,L). When Langerin-DTA mice were colonized with SFB,

Th17 cells in the LP expanded similarly to those in WT littermates (Figure 3-1E-J).

Moreover, significant induction of SFB-specific Vb14+ Th17 cells and response of LP

CD4 T cells to SFB antigens in vitro were evident (Figure 3-1J-L). These results

demonstrate that DP DCs are dispensable for both T cell priming and Th17 cell

differentiation following SFB colonization.

We obtained similar results using another model of DP DC depletion. DP DC

development depends on Notch2 and conditional deletion of Notch2 in CD11c+ cells

leads to significant loss of DP DCs 86. Similarly to Langerin-DTA mice, loss of DP DCs

in CD11c-Cre/Notch2-flox mice did not affect Th17 cell induction by SFB (Figure 3-S2).

97

B. CD103 DCs are dispensable for Th17 cell induction by SFB

CD103 SP DCs are capable of migrating to the MLN, share a developmental pathway

with CD8a+ splenic DC, and are proficient in cross-presentation 303,305,310,311. Whether

they play a non-redundant role in commensal CD4 Th17 cell responses is not known. To

address their role in SFB-induced Th17 cell differentiation, we colonized SFB-negative

BATF3-deficient mice and heterozygous littermates with SFB and compared Th17 cell

induction and induction of SFB-specific CD4 T cells (Figure 3-S3). As previously

reported 303, BATF3-deficient mice lacked CD103 SP DCs in LP and MLN (Figure 3-

S3A-D). Nevertheless, Th17 cell induction after SFB colonization was unaffected in

these animals. Similarly, induction of SFB-specific CD4 T cells and response to SFB

antigens were similar to littermate controls (Figure 3-S3E-M). Therefore, CD103 SP DCs

are not required for commensal-induced Th17 cell priming and differentiation.

The two subsets of CD103+ DCs represent the main conventional DC subsets in

the LP and have both been shown to migrate to MLN and prime CD4 T cell responses

51,52,305,312. To account for potential redundant functions of these subsets in Th17

responses to SFB, we crossed Langerin-DTA mice and BATF3-deficient mice (Figure 3-

2). The resulting double-knockout (DKO) mice lacked all CD103 DC subsets in both SI

LP and MLN (Figure 3-2A-D and Table 3-S1). Despite the absence of virtually all

CD103 DCs, colonization of DKO and littermate control mice with SFB led to a similar

induction of RORgt+ and IL-17+ CD4 T cells in the SI LP (Figure 3-2E-J). In addition,

there was a significant induction of Vb14+RORgt+ and Vb14+IL-17+ SFB-specific CD4 T

cells in the DKO small intestine (Figure 3-2E,G,H,J), and isolated SI LP CD4 T cells

from DKO mice responded to SFB antigens in in vitro proliferation assays similarly to

98

WT CD4 T cells (Figure 3-2K,L). We generated another model of CD103 DC deficiency

by crossing BATF3-deficient mice and CD11c-Cre/Notch2-flox mice. These animals also

lacked both CD103+ DC subsets and showed normal Th17 cell responses to SFB (data

not shown). These results demonstrate that LP CD103 DCs are dispensable for priming of

SFB-specific CD4 T cells and Th17 cell induction in response to SFB.

C. Conventional DCs are dispensable for commensal Th17 cell induction at steady

state

Conventional intestinal DCs depend on the DC-specific growth factor Flt3L 51,306,313. To

determine if conventional DCs in general play a role in generation of SFB-induced Th17

cells, we examined Th17 cell induction in Flt3L-deficient mice. Similarly to CD103+

DCs, CD11b+CD103- DCs have been shown to derive from pre-DC precursors, be

dependent on Flt3L and are, significantly decreased in Flt3L-deficient mice 306. We

established SFB-negative Flt3L-deficient mice and compared Th17 cell induction

following SFB colonization. CD103+ DC were almost absent from the SI LP in these

animals (>90% reduction, compared to heterozygous littermates). Flt3L-deficient mice

also had significantly diminished CD11b SP DCs, in agreement with previous studies 306

(Figure 3-3A,B and Table 3-S1). All subsets of migratory DCs were also severely

reduced in MLN (Figure 3-3C). In contrast, the total number of CD64+ Mfs in the SI LP

was similar between control littermates and Flt3L-deficient mice (Figure 3-3A,B).

Surprisingly, despite the severe defect in DC numbers, as well as possible defects in

lymphocyte development in Flt3L-deficient animals, SFB still induced normal levels of

Th17 cells (Figure 3-3D-I). In addition, priming and generation of SFB-specific CD4 T

99

cells was virtually unperturbed, as was the generation of antigen-specific Th17 cells

(Figure 3-3I-K).

Based on the combined data in Figures 1-3, we conclude that conventional gut

CD103+ DCs and Flt3L-dependent CD103- DCs are not required for the acquisition and

presentation of SFB antigens, priming of SFB-specific T cells and induction of Th17 cells

in the SI LP.

D. Nongenotoxic depletion of intestinal monocyte-derived cells prevents SFB-

specific Th17 cell responses

To directly examine the role of intestinal Mfs we utilized a transient depletion system.

Although only a fraction of LP Mfs express high levels of CCR2 (Figure 3-S6A), steady

state intestinal Mfs are derived from CCR2+ blood monocytes 51,52,314 and can be

depleted in CCR2-DTR mice following diphtheria toxin (DT) treatment 235,315. A single

DT injection led to a near complete ablation of intestinal Mfs beginning at 24h and

lasting until at least 72 hours post treatment (Figure 3-S4D). Depletion of Mfs could be

maintained with DT injections every 2 days for at least 12 days (Figure 3-4A). DT

treatment did not affect CD103 DP DCs, which were still present in the LP and in the

migratory DC population in MLN in treated CCR2-DTR mice (Figure 3-4B-E). In

addition, few LP CD4 T cells and SFB-induced Th17 cells expressed CCR2 and DT

treatment did not affect Th17 cell numbers or the presence of SFB-specific Th17 cells in

SFB-positive CCR2-DTR mice (Figure 3-S4A-C). To assess the role of monocyte-

derived Mfs, SFB-negative CCR2-DTR mice and littermate controls were treated with

DT every 48 hours for 10 days. The mice were colonized with SFB on Day 2 and Th17

100

induction was analyzed 8 days later (Figure 3-4A). SFB colonization was similar between

the two groups (Figure 3-S4E). SFB induced high levels of Th17 cells in control animals

with induction of Vb14+ SFB-specific Th17 cells and proliferation of SI LP CD4 T cells

in response to SFB antigens (Figure 3-4F-J and Figure 3-S4G-I). In contrast, SFB

colonization did not lead to Th17 cell induction in DT treated CCR2-DTR mice (Figure

3-4F,G and Figure 3-S4G,H). Moreover, SI LP CD4 T cells from CCR2-DTR mice

depleted of Mfs did not respond to SFB antigens in vitro and did not contain Vb14+ SFB-

specific Th17 cells (Figure 3-4F,H,I,J and Figure 3-S4G,I). These results suggest that

monocyte-derived cells are required for induction of SFB-specific Th17 cell responses.

E. Transfer of exogenous monocytes rescues defects in Th17 cell induction following

Mf depletion

DT treatment in CCR2-DTR mice resulted in depletion of all CCR2 monocyte-derived

subsets. However, we found that prolonged treatment also affected certain DC subsets.

Prolonged DT treatment led to a loss of CD103 SP DCs (Figure 3-4B,C). In addition, DT

treatment led to depletion of a subset of CD11b SP DCs that express CCR2 306 (Figure 3-

S6A). However, as shown earlier, CD103 SP DCs and Flt3L-dependent CCR2+ CD11b

SP DCs 306 are dispensable for SFB-mediated Th17 cell induction (Figure 3-3 and 3-S3).

Prolonged DT treatment also resulted in a decrease in total migratory DCs in the MLN,

although the numbers of MLN CD103+CD11b+ DCs (DP DCs) were normal (Figure 3-

4D,E and 3-S4F).

To better investigate whether the defect in Th17 cell induction is due to the lack

of monocyte-derived cells, and to further exclude the possibility that DT treatment affects

101

CD4 T cells or other non-monocyte derived populations, we performed gain-of-function

experiments. We isolated Lin-Ly6C+CCR2+ monocytes to high purity from bone marrow

(BM) of CD45.1 C57BL/6 congenic mice. Lineage markers included CD3, B220, NK1.1,

CD11c, and c-Kit, to eliminate dendritic cell progenitors and hematopoietic stem cells

240. CD45.2 CCR2-DTR mice were treated with DT every 60 hours to maintain depletion

of endogenous monocytes. After the initial DT injection, one group of CCR2-DTR mice

received 5-10 x 106 CD45.1+ BM monocytes. Control mice received DT, but did not

receive any recipient cells. Following the monocyte graft, mice were colonized with SFB

and Th17 cell induction was assessed 10 days later (Figure 3-5A). In agreement with

previous studies 52, transferred monocytes exclusively reconstituted the CD64 Mf

compartment and donor-derived CD45.1 cells were not detectable in any of the other

MNP subsets, neither in SI LP nor in MLN (Figure 3-5B,C and Figure 3-S5C,D).

Similarly to previous experiments, SFB colonization did not induce SFB-specific Th17

cells in control CCR2-DTR mice without transfer. In contrast, transfer of monocytes and

recovery of the LP Mf population, resulted in recovery of SFB-specific Th17 cell

responses, including the presence of CD4+RORgt+ cells, CD4+IL-17+ cells in the LP, and

response of LP CD4 T cells to SFB antigens in vitro (Figure 3-5E-L). Interestingly,

monocyte transfer also led to partial increase in endogenous CD45.2+ (host-derived) DCs,

especially in the migratory DC fraction of MLN (Figure 3-5D and 3-S5A,B), although it

did not significantly increase the number of CD103 SP DCs, underscoring the fact that

this subset is dispensable for Th17 cell induction. These results demonstrate that

monocyte-derived LP Mfs are essential for initiation and maintenance of SFB-specific

Th17 cell responses, possibly with help from migratory DCs.

102

F. Specific depletion of CD64 Mfs leads to loss of SFB-mediated Th17 cell induction

To further confirm the role of CD64 Mfs, we sought to implement an independent

depletion model. In contrast to DCs, intestinal Mf development and maintenance depends

on CSF1R (also known as M-CSFR) 51. Injection of a CSF1R-blocking antibody (clone

AFS98) can specifically deplete intestinal Mfs in a dose-dependent manner without

affecting resident DC subsets 18,277,313. We therefore treated WT C57BL/6 mice with a

high dose of AFS98, or control IgG, prior to SFB colonization. As shown in Figure 3-6,

AFS98 treatment led to a significant and specific depletion of intestinal Mfs. The average

depletion was ~95% in the CD64 Mf fraction. In contrast, LP DC subsets, including

CD103 SP DCs, DP DCs, and CD11b SP DCs were not affected by this treatment (Figure

3-6A,B and 3-S6B and Table 3-S1). Moreover, we did not detect any noticeable defects

in the number and phenotype of migratory DC subsets in the MLN (Figure 3-6D,E). SFB

colonization led to Th17 cell induction in mice treated with control IgG 8 days after

introduction of the bacteria, which included induction of CD4+RORgt+ and CD4+IL-17+

cells and induction of SFB-specific Th17 cells as demonstrated by the induction of

Vb14+RORgt+ and Vb14+IL-17+ cells (Figure 3-6F-K). In contrast, in mice treated with

AFS98, RORgt+ and IL-17+ Th17 cells, as well as SFB-specific Th17 cells were

significantly reduced and were similar to the levels in SFB-negative controls (Figure 3-

6F-K). These results demonstrate that intestinal CD64 Mfs are essential for initiation of

antigen-specific Th17 cell responses to an intestinal commensal.

Our data demonstrate a crucial in vivo function of intestinal Mfs in controlling

effector T cell-homeostasis to luminal bacteria. Identification of the exact mechanisms of

103

antigen acquisition and the location of T cell priming will be important future questions

to address. Regardless of the details, this mechanism must be distinct from conventional

sampling of luminal antigens by DCs at steady state or by the DC/Mf-mediated acute

immune response to invasive pathogens. Because of the specific nature of the interaction

of SFB with the host, we propose that this pathway may represent a more general

mechanism for inducing localized effector Th17 cell responses to mucosa-associated non-

invasive bacteria.

104

Experimental Procedures

Mice

Langerin-DTA, BATF3-/-, Notch2F/F, CX3CR1-GFP and CD11c-Cre mice were obtained

from the Jackson Laboratory. Flt3l-/- mice were obtained from Taconic farms and derived

SFB-free by antibiotic treatment of a founder breeding pair, followed by fecal

transplantation of Jackson (SFB-negative) microbiota. CCR2-DTR and CCR2-GFP mice

have been previously described 315. CCR2-DTR mice were re-derived by embryo transfer

and kept SFB-negative in our colony. All mice were bred and housed under specific

pathogen-free conditions at Columbia University Medical Center under IACUC approved

guidelines. To control for microbiota and cage effects, all experiments were performed

with littermate control animals housed in the same cage.

SFB colonization and Th17 cell induction

All mice, regardless of origin, were screened at multiple points for the presence and

levels of SFB by quantitative PCR 316. Bacterial genomic DNA isolation from fecal

pellets and quantitative PCR for the SFB 16S rRNA gene were performed as previously

described 147,316. SFB colonization was performed by oral gavage with SFB-containing

fecal pellets. To control for SFB levels in the feces used for gavage, as well as for other

constituents of the microbiota between experiments, all gavages were performed with

frozen stocks from a single batch of feces obtained from 10 SFB-positive Taconic B6

mice. Control mice were gavaged with fecal pellets from SFB-negative littermates in our

colony or with PBS. SFB colonization levels were confirmed by quantitative PCR and

normalized to levels of total bacteria (UNI). SI LP Th17 cell induction was assessed 8-10

105

days after gavage unless otherwise noted.

Lamina propria cell isolation and in vitro co-culture experiments

Lamina propria (LP) lymphocytes, intracellular cytokine staining, and RORgt staining

were performed as previously described 147. LP CD4+ T cells were purified by positive

selection using anti-CD4 magnetic microbeads and MACS columns (Miltenyi Biotec). 3-

5 x 104 CD4 T cells were co-cultured in 96-well U-bottom plates with 5 x 104 MACS

purified splenic CD11c+ cells as APCs in the presence or absence of autoclaved bacterial

lysates prepared from feces of SFB-monocolonized mice (SFB) or SFB-negative Jackson

C57BL/6 mice (Jax) as previously described 202,316. T cell proliferation was assessed 72

hours later by counting the number of live proliferated CD4 T cells.

DT treatment for ablation of intestinal Mfs

SFB-negative CCR2-DTR mice and littermate controls were treated with 20 ng/g

diphtheria toxin (DT) i.p. on Day 0 and every 48 or 60 hours after that for the duration of

the experiment (a total of 6 or 5 injections respectively). On Day 2 some mice were

gavaged twice with SFB-containing fecal homogenates. Th17 cell induction was

examined on Day 10.

Adoptive transfers

SFB-negative CD45.2 CCR2-DTR mice were treated with DT on Day 0 and every 60

hours after that (total of 5 injections). On Day 1.5 some of the mice received 5-10 x 106

Lin-GFP+ bone marrow monocytes, purified from congenic CD45.1 CCR2-GFP mice 315

106

or Lin-Ly6Chi bone marrow monocytes from CD45.1 C57BL/6 mice by cell sorting.

Transfer of a large number of BM monocytes was required to reconstitute the LP Mf

compartment in monocyte-depleted CCR2-DTR mice to significant levels. Recipient

mice were gavaged with SFB on Day 2 and DC subsets and Th17 cell induction were

examined on Day 12. The Lin(eage) cocktail included B220, CD3, NK1.1, CD11c, and

CD117 (c-Kit). Sorting was performed on a FACS Aria II (BD).

Macrophage depletion

For macrophage depletion, four days prior to SFB colonization, SFB-negative C57BL/6

mice were injected intra-peritoneally with 150 ug/g of body weight of CSF1R blocking

antibody (clone AFS98 317), purified from a hybridoma as described earlier 318

Cell numbers and statistics

To compensate for differences in yield between experiments, in some figures numbers of

lamina propria and mesenteric lymph node mononuclear cell subsets are represented as

percentage of total live single cells (gate R1 in Figure S1B). Significance was determined

by the Student’s unpaired two-tailed t test unless otherwise noted. P values were

represented on figures as follows: ns, p ≥ 0.05, * p < 0.05, ** p < 0.01, *** p < 0.005,

**** p < 0.001, ***** p < 0.0005, ****** p < 0.0001. Error bars on all figures represent

standard deviation of the mean.

107

ACKNOWLEDGMENTS

We thank Ingrid Leiner and Eric Pamer at the Memorial Sloan-Kettering Cancer Center

for providing CCR2-GFP and CCR2-DTR mice. We thank Lei Ding for expert advice in

BM experiments. We thank Darya Esterhazy and Daniel Mucida at the Rockefeller

University for reagents. We thank Amir Figueroa, Kristie Gordon and Siu-Hong Ho at

the Columbia Microbiology, Cancer Center, and Center for Translational Immunology

Flow Cytometry Cores for cell sorting. We thank Boris Reizis and Steve Reiner for

invaluable advice and scientific discussions. This work was supported by the National

Institutes of Health R01-DK098378 to I.I.I., R01-AI093808 to T.M.H., and by the

Crohn’s and Colitis Foundation of America SRA#259540 to I.I.I. I.I.I. is a Pew Scholar

in the Biomedical Sciences, supported by the Pew Charitable Trust.

108

FIGURES

109

Figure 3-1. CD103+CD11b+ (DP) DCs are dispensable for commensal Th17 cell

induction. (A,B) CD11c+MHCII+ MNP subsets in SI LP of Langerin-DTA mice and

control wildtype littermates (WT LM). (A left) FACS plots gated on CD11c+MHCII+

cells. A (right) Distribution of CD64 Mfs and CD11b SP DCs within the CD103-CD11b+

gate. Numbers represent percentage of cells in the corresponding gate. (B) Total number

of cells in individual MNP subsets as defined in (A). (C,D) Total cell numbers and cell

numbers in MNP subsets in the migratory DC fraction of mesenteric lymph nodes

(MLN). Plots in (C) are gated on Lin-CD11cloMHCIIhi migratory DC. (E-G) Induction of

RORgt+ Th17 cells by SFB in small intestinal (SI) LP. Plots gated on TCRb+CD4+ cells.

(H-J) Induction of IL-17+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(K,L) Response of purified SI LP CD4 T cells to SFB antigens or control bacterial

antigens (Jax antigens) prepared as described in Methods. Open circles in bar graphs in

all panels represent data from individual animals. Data from three experiments with

similar results.

110

111

Figure 3-2. CD103+ DCs are dispensable for commensal Th17 cell induction. (A,B)

CD11c+MHCII+ MNP subsets in SI LP of Langerin-DTA X BATF3-/- (DKO) mice and

control littermates. (A left) FACS plots gated on CD11c+MHCII+ cells. Numbers

represent percentage of cells in the corresponding gate. (B) total number of cells in

individual MNP subsets. (C,D) Total cell numbers and cell numbers in MNP subsets in

the migratory DC fraction of mesenteric lympn nodes (MLN). (E-G) Induction of

RORgt+ Th17 cells by SFB in small intestinal (SI) LP. Plots gated on TCRb+CD4+ cells.

(H-J) Induction of IL-17+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(K) Response to SFB antigens of purified SI LP CD4 T cells from DKO and control

littermates. SI LP CD4 T cells were isolated before (No SFB) and after (+SFB)

colonization with SFB and incubated with CD11c+ splenic DCs for 72 hours in the

presence of SFB antigens. (L) Proliferation of purified SI LP CD4 T cells in response to

SFB antigens. LP CD4 T cells were isolated and cultured as in K in the presence of SFB

antigens as described in Methods. Open circles in bar graphs in all panels represent data

from individual animals. Data from three experiments with similar results.

112

113

Figure 3-3. Conventional DCs are dispensable for commensal Th17 cell induction.

(A,B) CD11c+MHCII+ MNP subsets in SI LP of Flt3l-/- mice and control heterozygous

littermates (Flt3l+/-). (A left) FACS plots gated on CD11c+MHCII+ cells. Numbers

represent percentage of cells in the corresponding gate. (B) total number of cells in

individual MNP subsets. (C) Total cell numbers and cell numbers in MNP subsets in the

migratory DC fraction of mesenteric lymph nodes (MLN). (D-F) Induction of RORgt+

Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells. (G-I) Induction of IL-17+

Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells. (J,K) Response of purified

SI LP CD4 T cells to SFB antigens and antigens isolated from SFB-negative feces (Jax

antigens). CD4 T cells were isolated from the SI LP of SFB-colonized Flt3l+/- or Flt3l-/-

mice and assessed for antigen reactivity as described in Methods. Open circles in bar

graphs represent data from individual animals. Data combined from two experiments

with similar results.

114

115

Figure 3-4. Intestinal Mfs are required for mucosal Th17 cell induction. (A)

Experimental design. CCR2-DTR mice and WT littermates were treated with DT every

48h as described in Methods. SFB colonization occurred on Day 3 and Th17 cells were

examined on Day 12. (B,C) CD11c+MHCII+ MNP subsets in SI LP of CCR2-DTR mice

and control littermates (WT LM) treated with DT for 12 days. (B left) FACS plots gated

on CD11c+MHCII+ cells. Numbers represent percentage of cells in the corresponding

gate. (C) Number of cells in individual MNP subsets represented as percentage of total

live SI LP cells (gate R1 in Figure S1B). (D,E) Cell numbers represented as percentage of

total live single cells (D) and MNP subsets in the migratory DC fraction of mesenteric

lymph nodes (MLN). Plots in (E) are gated on Lin-CD11cloMHCIIhi migratory DCs. (F-

H) Induction of RORgt+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(I) Response to SFB antigens of purified SI LP CD4 T cells from DT-treated CCR2-DTR

and control littermates. (J) Proliferation of purified SI LP CD4 T cells. LP CD4 T cells

were isolated and cultured as in (I) in the presence of SFB antigens or control bacterial

antigens (Jax antigens) as described in Methods. Open circles in bar graphs represent data

from individual animals. Data combined from two out of three experiments with similar

results.

116

117

Figure 3-5. Exogenous monocytes recover Th17 cell induction in Mf-depleted mice (A) Experimental design. DT-treated CCR2-DTR mice were reconstituted on Day 1.5

with 5-10 X 106 Lin-Ly6Chi monocytes purified from bone marrow of CD45.1 congenic

mice. (B) Reconstitution of CD64 Mfs in SI LP by transfer of BM monocytes (+Mono).

(C) SI LP MNP subsets represented as percentage of total live single cells (D) Cell

number of migratory DCs in MLN represented as percentage of total live single cells. (E-

G) Induction of RORgt+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(H-J) Induction of IL-17+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(K,L) Response of purified SI LP CD4 T cells to SFB antigens. Open circles represent

data from individual animals. Data combined from four independent experiments.

118

119

Figure 3-6. Treatment with CSF1R-blocking antibody impedes Th17 responses to

SFB

(A,B) MNP subsets in the SI LP of C57BL/6 mice treated with high dose of anti-CSF1R

monoclonal antibody (AFS98) or control IgG four days before SFB colonization. (C)

SFB levels in feces of AFS98 and control IgG treated mice normalized to total bacterial

DNA (UNI). (D,E) Total cell numbers and cell numbers in MNP subsets in the migratory

DC fraction of mesenteric lymph nodes (MLN). (F-H) RORgt+ Th17 cells in SI LP 8

days after SFB gavage. Plots gated on TCRb+CD4+ cells. (I-K) IL-17+ Th17 cells in SI

LP on Day 8 post SFB gavage. Plots gated on TCRb+CD4+ cells.

120

Figure 3-7. Central role of intestinal Mfs in generation of commensal-induced Th17

cells

Intestinal Mfs acquire antigens from epithelium-associated SFB and initiate SFB-specific

Th17 cell responses. CD103+ DCs are dispensable for the induction of Th17 cells.

Intestinal Mfs may participate in the Th17 cell differentiation stage locally in the LP or

collaborate with CX3CR1+ DCs for antigen transfer into MLN or Th17 cell

priming/maintenance in the LP.

121

SUPPLEMENTAL FIGURES

122

Figure 3-S1. Phenotype of mononuclear phagocyte subsets

(A) Gating scheme used to identify intestinal MNP subsets. Lamina propria mononuclear

cells were isolated from small intestines of WT and CX3CR1-GFP mice. Numbers reflect

percentage of cells within the corresponding gate. The gates identifying the four

mononuclear phagocyte subsets are color-coded.

(B) Expression of surface markers by small intestinal lamina propria mononuclear

phagocyte subsets (black line) identified as in (A), compared to a negative cell subset in

the same sample (shaded histogram)

123

Figure 3-S2. DP DCs are not required for SFB-induced Th17 cell responses

(A) Defect in DP DC development in CD11c-Cre/Notch2-flox mice (CD11cΔNotch2 mice)

(B) SFB colonization induces normal levels of Th17 cells in CD11cΔNotch2 mice. Plots

gated on TCRb+CD4+ cells. Data from one of two independent experiments with similar

results.

124

125

Figure 3-S3. CD103+CD11b- (CD103 SP) DCs are dispensable for commensal Th17

cell induction

(A,B) CD11c+MHCII+ MNP subsets in SI LP of BATF3-/- mice and control heterozygous

littermates. (A) FACS plots gated on CD11c+MHCII+ cells. Numbers represent

percentage of cells in the corresponding gate. (B) total number of cells in individual MNP

subsets represented as percentage of total live single cells

(C,D) Total number and individual DC subsets in the migratory fraction from MLN of

BATF3-/- mice and control heterozygous littermates. Plots gated as in Figure S2. Cell

numbers in (D) are presented as percentage of total live single cells

(E-G) Induction of RORgt+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells

(H-K) Induction of IL-17+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells.

(L) Response to SFB antigens of purified SI LP CD4 T cells from BATF3-/- and

BATF3+/- littermates. SI LP CD4 T cells were isolated before (No SFB) and after

(+SFB) colonization with SFB and incubated with CD11c+ splenic DCs for 72 hours in

the presence of SFB antigens.

(M) Proliferation of purified SI LP CD4 T cells in response to various antigens. LP CD4

T cells were isolated and cultured as in F in the presence of SFB or control SFB-negative

bacterial antigens (Jax antigens) as described in Methods. Open circles in bar graphs

represent data from individual animals. Data from one of three experiments with similar

results.

126

127

Figure 3-S4. Cell subsets in CCR2-DTR mice following DT treatment

(A) CCR2 expression on total CD4 T cells in WT and CCR2-DTR mice treated with DT

for 12 days. Plots gated on TCRb+CD4+ cells

(B) CCR2 expression on Th17 cells in WT and CCR2-DTR mice treated with DT for 12

days. Plots gated on TCRb+CD4+IL-17+ cells

(C) Effects of DT treatment on Th17 cells in CCR2-DTR mice. WT and CCR2-DTR

mice were colonized with SFB for several weeks to induce Th17 cell differentiation and

then treated with DT every 48 hours for 12 days. DT treatment led to ablation of CD64

Mfs in CCR2-DTR mice (not shown), but did not lead to decrease in the numbers of

SFB-induced Th17 cells compared to WT mice

(D) Kinetics of Mf depletion in CCR2-DTR mice. CCR2-DTR and WT mice received a

single injection of DT and SI LP MNP subsets were examined 24 or 72 hours later. Plots

gated on CD11c+MHCII+ cells.

(E-I) CCR2-DTR and WT littermates (LM) were treated with DT and colonized with

SFB as described in Figure 4A

(E) SFB levels on Day 12 assessed by Q-PCR for SFB 16S rRNA gene normalized to

total 16S rRNA (UNI)

(F) DC subsets in the migratory fraction of MLN from WT and CCR2-DTR mice. Cell

numbers are presented as percentage of total live single cells

(G-I) Induction of IL-17+ Th17 cells by SFB in SI LP. Plots gated on TCRb+CD4+ cells

128

129

Figure 3-S5. Recovery of intestinal Mfs by monocyte transfer

CCR2-DTR and WT littermates (LM) were treated with DT and colonized with SFB as

described in Figure 5A. Shortly before SFB colonization some mice received an adoptive

transfer of 5-10 X 106 bone marrow monocytes (+ Mono) from CD45.1+ congenic mice

(Figure 5A)

(A,B) DC subsets in the migratory DC fraction of MLN from DT-treated SFB-colonized

WT and CCR2-DTR mice with and without transfer of 5-10 X 106 bone marrow

monocytes (+ Mono)

(C) MNP subsets in small intestinal LP of CCR2-DTR mice with and without monocyte

transfer. MNP plots (far left) are gated on all CD11c+MHCII+ cells. Donor-derived cells

were identified as CD45.1+ and were found only in the CD64+CD24- Mf fraction, but not

in the CD64-CD24+ DC fraction (right plots)

(D) Donor-derived cells are absent in the migratory DC fraction of MLN. Plots gated as

in Figure S2

(E) SFB levels on Day 12 assessed by Q-PCR for SFB 16S rRNA gene normalized to

total 16S rRNA (UNI)

130

131

Figure 3-S6. CCR2 expression on lamina propria MNP subsets

(A) CCR2+ cells in small intestinal LP MNP subsets from WT and CCR2-DTR mice

treated with DT for 12 days. Plots gated on MNP subsets as outlined in Figure S1.

(B) WT C57BL/6 mice were treated with a single injection of blocking anti-CSF1R

antibody (AFS98) or isotype control (IgG). Small intestinal LP MNP subsets were

examined a week later.

132

Table 3-S1. MNP subsets in various mouse models used in this study

Comparison of the size of individual MNP subsets in different mouse models. Total cell

numbers in each fraction in the small intestinal lamina propria (x104). Main numbers

represent mean (x104) and secondary numbers – standard deviation of the mean.

Statistically significant differences from WT littermates are marked in red. Significance

values are reported on the corresponding figures.

133

CHAPTER FOUR

DISCUSSION

Intestinal T cell homeostasis is essential for achieving a balance between chronic

intestinal inflammation and protective mucosal immune responses. How commensal

microbes regulate T cell homeostasis and how this modulation affects immune responses

to innocuous and pathogenic antigens remain poorly understood. It is well appreciated

that distinct commensal species control specific effector T cell fates and modulate innate

immune pathways. However, the nature of the innate immune cells that relay microbial

information and the mechanisms by which they guide mucosal immune responses to

commensals and pathogens are unclear.

Here, we examined in detail the mechanisms of Th17 cell induction by a

commensal bacterium. We made significant progress towards understanding how SFB

interact with innate immune cells to regulate local immune responses. We showed that

cytokine environment is not sufficient to drive SFB-specific Th17 cell differentiation.

However, the response to this microbe is antigen-specific and we find that most SFB-

induced Th17 cells recognize and respond to SFB antigens. We also investigated the role

of various intestinal APCs in this response. We queried all known intestinal lineages that

possess antigen presentation capabilities and identified CD11c+ cells as necessary and

sufficient players in this process. We also identified intestinal Mfs as essential for

induction of the SFB-specific Th17 cell response. During the course of this work, we

came to appreciate the diverse ontogeny of LP MNPs. Accumulating evidence proposes a

division of labor among LP MNPs to explain how immune subsets decode innocuous and

134

pathogenic stimuli, mount proper immune responses and prevent overt

inflammation61,64,67,87,199,229,236,240,319,320. However, studies supporting such roles in vivo

are lacking. To understand how SFB modulate the host adaptive immune system, I set on

to examine the contribution of the various LP MNPs to SFB-specific Th17 cell

differentiation. This work represents one of the first systematic investigations of the role

of individual intestinal MNP subsets in control of intestinal immune homeostasis in vivo.

One of the main findings from this work is that intestinal macrophages are

essential for induction of SFB-specific Th17 cell differentiation. This is in contrast to

previous reports that define DP DCs as the DC subset responsible for Th17 cell

differentiation61,64,234,237. Instead, we find that SI DP DCs are dispensable for generation

of SFB-specific CD4 T cells. Indeed, in the absence of these cells from LP and MLNs in

Langerin-DTA mice, SFB-specific Th17 cell differentiation progresses similarly to

control mice. To exclude the possibility of redundancy among pre-DC derived DCs, we

also examined SFB-specific Th17 cell induction in Langerin-DTA/BATF3-/- and Flt3L-/-

mice, which lack all CD103+ DCs and over 95% of classical DCs, respectively.

Strikingly, SFB can still mediate Th17 cell differentiation in these models, reinforcing

our observation that priming of SFB-specific T cells and Th17 cell induction is not reliant

on conventional migratory LP DC subsets. Furthermore, our results show that DP DCs

are not sufficient for Th17 cell induction by SFB because SFB-specific CD4 T cells are

not generated in DT-treated CCR2-DTR mice, in which DP DCs are the only remaining

DC subset.

What is the actual function of classical DCs and CD64 Mfs in SFB-specific

Th17 cell differentiation? CD64 Mfs have been shown to acquire and process antigens

135

for presentation228,230,236, release Th17-cell inducing cytokines177,200,277, and support

immunoprotective functions during infection with extracellular bacteria75. They can also

migrate to MLNs under certain conditions236. It is therefore possible that CD64 Mfs are

the sole drivers of SFB-specific Th17 cell differentiation. Tissue-resident CD64 Mfs are

insensitive to TLR stimulation and are classically viewed as anti-inflammatory due to

high secretion of cytokines such as IL-10. Indeed we confirmed Il-10 expression from

CD64 Mfs and not LP DCs (unpublished data). However, in the context of infection or

acute inflammation in the colon, Ly6chi monocytes can give rise to proinflammatory

CX3CR1int Mfs with enhanced migratory and antigen presentation abilities without

altering the fate of the residual anti-inflammatory Mfs31,321. Thus, CD64 Mfs derived

from the same progenitors are capable of assuming different fates depending on the

context and location of macrophage maturation and function. However, the CD64 Mfs

examined in our study do not resemble pro-inflammatory Mfs. CD64 Mfs express MHCII

and similar levels of costimulatory molecules CD40, CD80, and CD86 to those of

conventional LP DCs52 suggesting that at least phenotypically they are functional antigen

presenters. In addition, ex vivo assays with sorted SI LP CD64 Mfs and transgenic CD4 T

cells demonstrate that they can induce T cell proliferation, albeit less efficiently than

classical CD103 SP DCs or DP DCs (51,64, and our own data). Previous studies show that

addition of GM-CSF alone to bone-marrow derived monocytes in vitro can drive

expansion and differentiation of monocytes into cells with antigen presentation capability

322. While GM-CSF deficiency does not impair development of LP CD64 Mfs51, these

cells are responsive to GM-CSF released by ILC3 cells in response to microbial

signals277. Thus, microbiota-mediated GM-CSF production may facilitate LP macrophage

136

maturation and antigen presentation. Whether SFB stimulate GM-CSF production from

ILC3 cells or other innate immune cells has not been examined. Despite the lack of this

knowledge, SFB have been recently proposed to facilitate a crosstalk between CD64 Mfs

and ILC3 cells to promote local differentiation of SFB-specific Th17 cells 253.

CD64 Mfs express low levels of CCR7 and are present at low numbers in the

MLNs at steady state 51,64,77, however, they can upregulate CCR7 and migrate to the

lymph when exposed to non-invasive Salmonella typhimurium in antibiotic-treated

animals 236. At steady state, SFB are the sole epithelium-associated commensal species in

laboratory SPF mice (unpublished data from our lab). Therefore, we hypothesize that

innate immune sensing of this abundant species in conjunction with their direct contact

with host IECs encourage SI CD64 Mfs to function as APCs 200,323. As mentioned above,

several reports suggest that innate immune recognition of SFB by SI CD64 Mfs must take

place as these cells produce high IL-1β in mice colonized with SFB, although the exact

signaling pathways are still debated 177,200. We, however, notice that CD64 Mfs produce

similar levels of IL-1β before and after five days of SFB colonization, although we

cannot exclude the possibility that more IL-1β is produced earlier on (unpublished data).

Of note, during C. rodentium infection de novo differentiated intestinal macrophages

release IL-1β by activating the non-canonical caspase-11 inflammasome, thus providing

evidence for IL-1β secretion through a MyD88-independent innate pathway 68. Whether

the same inflammasome pathway recognizes SFB remains to be explored. As both C.

rodentium and SFB attach strongly to IECs and induce Th17 cell differentiation200, it is

plausible that both species undergo similar innate sensing, albeit with different final

outcome (i.e. homeostasis versus immunoprotection).

137

CD64 Mfs may also be required for SFB-specific Th17 cell differentiation by

communicating with and instructing local conventional DCs to prime CD4 T cells. In

light of this possibility, all of the models we employed still carry CD11b SP DCs at

varying levels. While these cells have been shown to migrate to lymph and prime effector

T cells 61,66,78, we do not know if they perform similar functions in SFB colonized mice.

Furthermore, as mentioned above, the origin of these cells is unclear as they seem to be

highly heterogeneous 55,34,56,70. CD11b SP DCs are thought to derive from Flt3L-

dependent precursors66, but are not outcompeted in WT:Flt3L-/- competitive BM

chimeras56. IRF4 deficiency in CD11c+ cells impairs migration of CCR2+CD103- DCs to

the MLNs, but CD11b SP DCs are only reduced by 50% in these mice61,66. These CCR2-

expressing CD11b SP DCs seem to rely on CCR2 for entry into LP, but paradoxically, do

not seem to develop from CCR2-expressing monocytes66. We find that CD11b SP DCs,

including the CCR2+ subset, are significantly reduced in Flt3L-/- LP and MLNs, but not

completely ablated, suggesting that some subpopulations of these DCs rely on Flt3

signaling. As mentioned earlier, Flt3L-/- mice show normal induction and numbers of

SFB-specific Th17 cell cells so we can exclude a role for Flt3L-dependent CD11b SP

DCs in this process. Similarly to Scott et al, we find that transfer of exogenous CCR2-

expressing monocytes does not recover the CCR2+ CD11b SP DCs, or any CD11b SP

DCs for that matter, in either LP or GALT, which underscores that their origin is distinct

from that of CD64 Mfs. However, we have noticed an increase in endogenous CD11b SP

DCs in both SI LP and MLNs in some of the mice with transferred WT monocytes. These

results suggest that the remaining LP DCs in Flt3L-/- may be a mixture of non-classical

APCs such as monocyte-derived DCs, monocyte-macrophage intermediates 324 and even

138

atypical MHCII-expressing eosinophils 325, which may rely on various extrinsic factors

for development and survival (i.e. GM-CSF, M-CSF, IL-34, IL-3, IL-4). Our results

cannot completely exclude a role for these cells in generation of SFB responses.

Despite that nearly all CD11b SP DCs are CX3CR1-GFPint, they also express

classical DC markers such as CD24, CD26, Zbtb46 and are reduced in Flt3L-/- LP

stressing that CX3CR1 expression cannot be used on its own to indicate monocyte origin

or commitment to macrophage fate. Of note, CD11bhi population that is Notch2-

independent and expresses high levels of CX3CR1 and Flt3 and variable levels of CD4

has been reported86, which was suggested to represent nonlymphoid F4/80-expressing DP

DCs56. However, CD11b SP DCs may also easily be the equivalent of this lymphoid

CD11bhi population. Intriguingly, DC-specific ablation of IRF8 leads to increased

proportions of CD11b SP DCs in the LP and significant increase in levels of CD11b SP

DCs within the migratory MLN DCs, suggesting that IRF8 negatively regulates

development or migration of CD11b SP DCs93. Supporting this, we observed increased

levels of CD11b SP DCs in BATF3-/- mice (unpublished data). It will be important to

identify transcriptional signatures of CD11b SP DCs to better define this subset

developmentally, develop strategies for targeting them genetically and identify SFB-

driven transcriptional changes.

Depletion of intestinal Mfs in DT-treated CCR2-DTR led not only to lack of Th17

cell induction by SFB, but also to lack of SFB-specific CD4 T cells in LP. We,

therefore, hypothesize that CD64 Mfs are required for acquisition of SFB antigens.

The exact mechanism(s) by which CD64 Mfs take up SFB antigens (i.e. transepithelial

dendrite-mediated luminal sampling, transfer from IECs, engulfment of apoptotic IECs)

139

will need to be characterized in the future. SFB adhesion to IECs is required for Th17 cell

induction and possibly for induction of SFB-specific T cells200. Thus, SFB are uniquely

positioned for sampling by CD64 Mfs. This mechanism may be common for epithelium-

associated bacteria.

A major unanswered question is the role of CD64 Mfs beyond antigen

acquisition. They may serve as actual antigen-presenting cells in priming SFB-specific

CD4 T cells or transfer the antigens to LP DCs. CD64 Mfs may also participate in the

Th17 cell differentiation stage by producing Th17 cell-inducing cytokines. Our results

strongly suggest role of Mfs as APCs, however they were insufficient to prove such a

role. We have planed the following experiments to address this possibility. To understand

whether CD64 Mfs are necessary for antigen presentation, we propose to reconstitute the

macrophage niche in DT-treated CCR2-DTR mice with MHCII-deficient macrophages.

To this end, we would transfer Ly6chiCCR2+ monocytes from MHCII-/- mice into Mf-

deficient mice and ask whether the newly differentiated macrophages can induce SFB-

specific Th17 cell differentiation. Our preliminary transfers of MHCII-deficient

monocytes were unsuccessful at fully reconstituting the macrophage niche. As an

alternative, we are generating mice that lack MHCII on macrophages and monocytes

(LysMΔMHCII) and we plan to assess SFB-specific Th17 cell differentiation before and

after colonization with SFB. To address whether Mfs are sufficient for SFB-specific

Th17 cell differentiation, we are also generating mice, which carry MHCII specifically on

Mfs (MHCIILysM). Because these gain-of-function mice will not contain endogenous CD4

T cells due to lack of selection on MHCII, we will transfer SFB Tg T cells and assess

their proliferation and differentiation into Th17 cells. Another experiment that would test

140

the requirement of antigen presentation by Mfs is mixed CCR2-DTR:MHCII-/- BM

chimera. Upon DT treatment these BM chimeras would carry only MHCII-deficient Mfs,

whereas there would be an equal representation of MHCII-sufficient and deficient DCs.

Our preliminary results revealed a caveat with this experiment because DT-treatment led

to loss of both MHCII-sufficient Mfs and DCs. We presume that competition at the

Mf/DC precursor stage after DT administration led to over-representation of MHCII-

deficient progenitors. More experiments are necessary to clarify the role of MHCII in

development of macrophages.

In order to gain information about which DC subset contains SFB antigens in vivo

we isolated DP DCs and Mfs from SI LP and examined their ability to activate SFB Tg T

cells without addition of exogenous antigen. The presumption was that only cells

presenting endogenous SFB antigens will activate the Tg T cells. We observed that DP

DCs, but not CD64 Mfs, can induce proliferation of SFB Tg T cells in the absence of

exogenous antigens. We confirmed that isolation of these cells did not impair their ability

to present exogenous antigens. Indeed, both isolated DP DCs and CD64 Mfs could

activate SFB Tg T cells in the presence of exogenous antigen, albeit CD64 Mfs were half

as efficient as DP DCs. These results may suggest that CD64 Mfs do not carry SFB

antigens, however, alternate explanations are also likely. As CD64 Mfs are more

phagocytic than DP DCs, it is possible that SFB antigens are either lost during the

isolation process or degraded to levels below the detection threshold of this assay. In

addition, the proportion of CD64 Mfs that carry SFB antigens may be low such that more

Mfs are necessary in this assay. Of note, Th17 cell levels were observed at normal levels

in mice lacking MHCII on DP DCs (Langerin-DTAΔMHCII, 113), suggesting that antigen

141

presentation by DP DCs is not essential. Using our assay we cannot exclude the

possibility that CD103 SP DCs or CD11b SP DCs carry SFB antigens. We have not been

able to isolate equivalent numbers of these populations compared to DP DCs. We also

have yet to test whether these DC subsets carry SFB antigens in Langerin-DTA or

Langerin-DTA, BATF3-/- mice. In addition, we do not know whether in the absence of

DP DCs or all DCs, CD64 Mfs now carry SFB antigens for presentation to SFB Tg T

cells.

Thus, it is possible that CD64 Mfs transfer SFB antigens to DP DCs (or any DC

subset) for presentation to SFB-specific CD4 T cells. DT-treated CCR2-DTR mice

contain normal levels of DP DCs in LP and MLN but lack CD64 Mfs and lack SFB-

specific Th17 cells. In the antigen transfer scenario, DP DCs would not carry SFB

antigens in DT-treated CCR2-DTR mice and therefore, not be able to induce activation of

SFB-specific CD4 T cells. Conversely, in the absence of all DCs, in Flt3L-/- mice, CD64

Mfs would be loaded with SFB antigens and may present the antigens themselves. As we

have shown above, SFB can induce Th17 cell differentiation in the absence of GALT and

MLN. Therefore, CD64 Mfs may induce activation and differentiation of SFB-specific

Th17 cells within the LP.

In addition, Mfs may participate as cytokine producing cells for induction of SFB-

specific Th17 cell differentiation. In this regard, we examined the production of Th17 cell

cytokines by DP DCs and CD64 Mfs in the ileum of SFB colonized mice. We observed

that CD64 Mfs generated higher levels of IL-6, IL-1β, and IL-23p19 than DP DCs, which

suggests that CD64 Mfs preferentially induce Th17 cell lineage-skewing cytokines.

Because of the contributions of IL-23, IL-1β and IL-6 to distinct stages of Th17 cell

142

differentiation, CD64 Mfs may be required for both initiating the Th17 lineage

transcriptional program and for driving poised SFB-specific RORγt+ CD4 T cells to

secrete Th17 cell-specific cytokines, and thus finally commit to the Th17 cell lineage. In

this regard, SAA proteins were shown to condition the neighboring CD11c+ cells to

produce IL-1β, which together with IL-6 and TGF-β enhance Th17 cell differentiation.

Myeloid cell-derived IL-1β appears to act in a paracrine manner and enhance production

of epithelial SAA, thus ensuring amplification of Th17 cell differentiation200.

In most of our Mo transfer experiments we noticed recovery of endogenous DCs

in the MLNs and higher levels of SFB-specific Th17 cell differentiation. Therefore,

another possibility is that CD64 Mfs do not present the antigens, but provide trophic

factors to DCs to induce their maturation and enhance their role as antigen presenters. DP

DCs and CD11b SP DCs may be especially receptive to these trophic factors due to their

mixed origins. Macrophage-derived trophic signals have been shown to facilitate tissue

development (i.e. ductal branching, bone morphogenesis, and generation of adipose

tissue) and tumor invasion (i.e. angiogenesis, extracellular matrix manipulation) 326. GM-

CSF, G-CSF, M-CSF, Flt3L, TGFβ, and IL-1β are a few candidate factors that intestinal

CD64 Mfs could elicit to support development of DC subsets, enhance their antigen

presentation functions and promote production of specific cytokine milieus 327-329.

Alternatively, Wnt/β-catenin pathways have been shown to maintain intestinal

homeostasis, induce tolerogenic functions in dendritic cells and regulate inflammatory

responses to pathogens 330-332. Other extrinsic factors such as chemokines and

semaphorin-plexin systems influence DC viability, migration to lymph and induction of

antigen-specific T cell responses 329,333. Overall, trophic factors provided by macrophages

143

could augment the efficiency of the LP MNP system and thus, regulate immune

responses.

SFB colonization induces Th17 cell differentiation without altering the

representation of the other effector T cell lineages, but it increases the ratio of Th17 cells

to pTreg cells, since these two lineages are in a tightly regulated equilibrium with each

other. Our unpublished observations confirm that SFB do not induce pTreg

differentiation and that there is no difference in pTreg levels between WT or DC-

deficient mice before and after SFB colonization. We further confirm previous

observations that BATF3-/- mice show no defect in FoxP3+pTreg proportions and that

Flt3L-/- mice have reduced levels of FoxP3+pTregs. In contrast to Welty et al, we do not

notice significant impairment of FoxP3+pTregs in Langerin-DTA, BATF3-/- mice, which

suggests that pTreg differentiation can occur in the absence of all CD103+ LP DCs.

Because of the pTreg defect in Flt3L-/-, we propose that LP DCs act redundantly for

induction of pTregs. Since induction of colonic pTregs seems to resemble that of SI Th17

cells, it remains to be investigated whether individual or multiple colonic LP DCs are

required for commensal-mediated pTreg differentiation.

With respect to the non-SFB microbiota, we find that LP MNP subset-specific

deficiencies do not reduce further the few numbers of Th17 cells observed in mice

lacking SFB. This observation suggests that either LP MNPs act redundantly to induce

Th17 cell differentiation in response to luminal commensals or that other innate immune

cells are necessary altogether for this process. It will be interesting to determine whether

there is a specific luminal commensal species that induces the residual Th17 cell

differentiation or whether it is a joined effort of the microbial community. If specific

144

species are identified, it will be important to compare and contrast their mechanisms of

induction of Th17 cell differentiation to those of mucosa-attaching SFB. Genetic

manipulation of both the luminal species and SFB will prove useful for generating

targeted therapies for inflammatory bowel diseases.

In this thesis study, we have shown that among LP CD11c+MHCII+ subsets,

CD64 Mfs are necessary for induction of SFB-specific Th17 cell differentiation. Their

precise contribution to this immune response remains to be explored, but our data points

to collaboration with conventional LP DCs. Specifically, our data suggests that SFB

antigen acquisition, antigen processing, and induction of Th17 lineage-skewing cytokine

milieu are supported by LP Mfs, while SFB antigen presentation and activation of SFB-

specific CD4 T cells are facilitated by conventional LP DCs. In addition, in the absence

of LP DCs, CD64 Mfs can perform all of these functions. Our studies focus on de novo

induction of Th17 cell differentiation so we do not know which LP MNP subset plays a

role in maintenance of SFB-specific Th17 cell responses or in development of SFB-

specific memory Th17 cells, if these cells exist. Upon depletion of CD64 Mfs after SFB-

specific Th17 cell differentiation occurred, we found that Th17 cell levels were not

reduced showing that CD64 Mfs are not necessary for maintenance of Th17 cells. It is

possible that once SFB-specific Th17 cells are induced they can be maintained either by

any LP MNP subset or by autocrine/paracrine cytokine signaling. Of course, the latter

possibility excludes the need for constant SFB antigen presentation and suggests that

SFB-specific Th17 cells are stable and committed to their fate. In this regard, it will be

important to remove SFB after Th17 cell differentiation occurred and assess any changes

in Th17 cell levels over time. As mentioned earlier, adhesion to IECs is postulated to be

145

required for Th17 cell differentiation and it is a feature of both commensal SFB and

pathogenic bacteria and fungi such as Citrobacter rodentium and Candida albicans.

Because CD64 Mfs play an important role in SFB-specific Th17 cell differentiation, it is

worth exploring whether they are necessary for Th17 cell differentiation driven by all

epithelium-associated microbes.

Our work helps elucidate the mechanisms of induction of commensal-specific

mucosal T cell responses. Similarly to previous knowledge that colonic pTregs consist of

commensal-specific TCRs, we showed that small intestinal SFB-induced Th17 cells

consist mostly of SFB-specific TCRs. In addition, we propose that SFB-specific Th17

cells are primed and differentiate locally in the LP, which has also been suggested for

commensal-specific but not food antigen-specific pTregs. We then show that LP MNPs

present SFB antigens for induction of SFB-specific Th17 cell differentiation, and we find

that CD64 Mfs are especially important in this process. The precise roles that CD64 Mfs

play in SFB-specific Th17 cell induction have yet to be described. Furthermore, whether

CD64 Mfs play a role in commensal-specific pTregs has also not been explored.

Nevertheless, our work is the first to propose an MNP subset that is required for

induction of a commensal-specific mucosal T cell response. Looking forward, it is worth

exploring the interplay among different MNP subsets that leads to induction and

discrimination of commensal-specific from pathogen-specific T cell responses. It is

further important to understand whether differentiation of commensal-specific T cells can

be subverted to induce pathogenic mucosal T cell responses. Our knowledge of the

distinguishing features of commensal-specific versus pathogen-specific T cell response

will contribute to designing novel therapies that can differentiate between friend and foe.

146

REFERENCES

1 Mowat, A. M. & Agace, W. W. Regional specialization within the intestinal

immune system. Nature reviews. Immunology 14, 667-685, doi:10.1038/nri3738 (2014).

2 Rao, J. N. & Wang, J. Y. in Regulation of Gastrointestinal Mucosal Growth Integrated Systems Physiology: from Molecule to Function to Disease (2010).

3 Koscso, B., Gowda, K., Schell, T. D. & Bogunovic, M. Purification of dendritic cell and macrophage subsets from the normal mouse small intestine. Journal of immunological methods 421, 1-13, doi:10.1016/j.jim.2015.02.013 (2015).

4 Noah, T. K., Donahue, B. & Shroyer, N. F. Intestinal development and differentiation. Experimental cell research 317, 2702-2710, doi:10.1016/j.yexcr.2011.09.006 (2011).

5 Birchenough, G. M., Johansson, M. E., Gustafsson, J. K., Bergstrom, J. H. & Hansson, G. C. New developments in goblet cell mucus secretion and function. Mucosal immunology 8, 712-719, doi:10.1038/mi.2015.32 (2015).

6 Gribble, F. M. & Reimann, F. Enteroendocrine Cells: Chemosensors in the Intestinal Epithelium. Annual review of physiology 78, 277-299, doi:10.1146/annurev-physiol-021115-105439 (2016).

7 Gerbe, F. et al. Intestinal epithelial tuft cells initiate type 2 mucosal immunity to helminth parasites. Nature 529, 226-230, doi:10.1038/nature16527 (2016).

8 von Moltke, J., Ji, M., Liang, H. E. & Locksley, R. M. Tuft-cell-derived IL-25 regulates an intestinal ILC2-epithelial response circuit. Nature 529, 221-225, doi:10.1038/nature16161 (2016).

9 Gerbe, F., Legraverend, C. & Jay, P. The intestinal epithelium tuft cells: specification and function. Cellular and molecular life sciences : CMLS 69, 2907-2917, doi:10.1007/s00018-012-0984-7 (2012).

10 Barker, N. Adult intestinal stem cells: critical drivers of epithelial homeostasis and regeneration. Nature reviews. Molecular cell biology 15, 19-33, doi:10.1038/nrm3721 (2014).

11 Madara, J. L. Cup cells: structure and distribution of a unique class of epithelial cells in guinea pig, rabbit, and monkey small intestine. Gastroenterology 83, 981-994 (1982).

12 Clevers, H. The intestinal crypt, a prototype stem cell compartment. Cell 154, 274-284, doi:10.1016/j.cell.2013.07.004 (2013).

147

13 Vaishnava, S. et al. The antibacterial lectin RegIIIgamma promotes the spatial segregation of microbiota and host in the intestine. Science 334, 255-258, doi:10.1126/science.1209791 (2011).

14 Peterson, L. W. & Artis, D. Intestinal epithelial cells: regulators of barrier function and immune homeostasis. Nature reviews. Immunology 14, 141-153, doi:10.1038/nri3608 (2014).

15 Hooper, L. V., Littman, D. R. & Macpherson, A. J. Interactions between the microbiota and the immune system. Science 336, 1268-1273, doi:10.1126/science.1223490 (2012).

16 Knoop, K. A. & Newberry, R. D. Isolated Lymphoid Follicles are Dynamic Reservoirs for the Induction of Intestinal IgA. Frontiers in immunology 3, 84, doi:10.3389/fimmu.2012.00084 (2012).

17 Gabanyi, I. et al. Neuro-immune Interactions Drive Tissue Programming in Intestinal Macrophages. Cell 164, 378-391, doi:10.1016/j.cell.2015.12.023 (2016).

18 Muller, P. A. et al. Crosstalk between muscularis macrophages and enteric neurons regulates gastrointestinal motility. Cell 158, 300-313, doi:10.1016/j.cell.2014.04.050 (2014).

19 Akashi, K., Traver, D., Miyamoto, T. & Weissman, I. L. A clonogenic common myeloid progenitor that gives rise to all myeloid lineages. Nature 404, 193-197, doi:10.1038/35004599 (2000).

20 Auffray, C. et al. CX3CR1+ CD115+ CD135+ common macrophage/DC precursors and the role of CX3CR1 in their response to inflammation. The Journal of experimental medicine 206, 595-606, doi:10.1084/jem.20081385 (2009).

21 Fogg, D. K. et al. A clonogenic bone marrow progenitor specific for macrophages and dendritic cells. Science 311, 83-87, doi:10.1126/science.1117729 (2006).

22 Onai, N. et al. Identification of clonogenic common Flt3+M-CSFR+ plasmacytoid and conventional dendritic cell progenitors in mouse bone marrow. Nature immunology 8, 1207-1216, doi:10.1038/ni1518 (2007).

23 Naik, S. H. et al. Development of plasmacytoid and conventional dendritic cell subtypes from single precursor cells derived in vitro and in vivo. Nature immunology 8, 1217-1226, doi:10.1038/ni1522 (2007).

24 Liu, K. et al. In vivo analysis of dendritic cell development and homeostasis. Science 324, 392-397, doi:10.1126/science.1170540 (2009).

25 Satpathy, A. T. et al. Zbtb46 expression distinguishes classical dendritic cells and their committed progenitors from other immune lineages. The Journal of experimental medicine 209, 1135-1152, doi:10.1084/jem.20120030 (2012).

148

26 Serbina, N. V. & Pamer, E. G. Monocyte emigration from bone marrow during bacterial infection requires signals mediated by chemokine receptor CCR2. Nature immunology 7, 311-317, doi:10.1038/ni1309 (2006).

27 Tsou, C. L. et al. Critical roles for CCR2 and MCP-3 in monocyte mobilization from bone marrow and recruitment to inflammatory sites. The Journal of clinical investigation 117, 902-909, doi:10.1172/JCI29919 (2007).

28 Geissmann, F., Jung, S. & Littman, D. R. Blood monocytes consist of two principal subsets with distinct migratory properties. Immunity 19, 71-82 (2003).

29 Auffray, C. et al. Monitoring of blood vessels and tissues by a population of monocytes with patrolling behavior. Science 317, 666-670, doi:10.1126/science.1142883 (2007).

30 Tamoutounour, S. et al. CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. European journal of immunology 42, 3150-3166, doi:10.1002/eji.201242847 (2012).

31 Bain, C. C. et al. Resident and pro-inflammatory macrophages in the colon represent alternative context-dependent fates of the same Ly6Chi monocyte precursors. Mucosal immunology 6, 498-510, doi:10.1038/mi.2012.89 (2013).

32 Tamoutounour, S. et al. Origins and functional specialization of macrophages and of conventional and monocyte-derived dendritic cells in mouse skin. Immunity 39, 925-938, doi:10.1016/j.immuni.2013.10.004 (2013).

33 Ginhoux, F., Lim, S., Hoeffel, G., Low, D. & Huber, T. Origin and differentiation of microglia. Frontiers in cellular neuroscience 7, 45, doi:10.3389/fncel.2013.00045 (2013).

34 Perdiguero, E. G. & Geissmann, F. The development and maintenance of resident macrophages. Nature immunology 17, 2-8, doi:10.1038/ni.3341 (2016).

35 Gomez Perdiguero, E. et al. Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature 518, 547-551, doi:10.1038/nature13989 (2015).

36 Yona, S. et al. Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis. Immunity 38, 79-91, doi:10.1016/j.immuni.2012.12.001 (2013).

37 Hoeffel, G. & Ginhoux, F. Ontogeny of Tissue-Resident Macrophages. Frontiers in immunology 6, 486, doi:10.3389/fimmu.2015.00486 (2015).

38 Hoeffel, G. et al. C-Myb(+) erythro-myeloid progenitor-derived fetal monocytes give rise to adult tissue-resident macrophages. Immunity 42, 665-678, doi:10.1016/j.immuni.2015.03.011 (2015).

149

39 Molawi, K. et al. Progressive replacement of embryo-derived cardiac macrophages with age. The Journal of experimental medicine 211, 2151-2158, doi:10.1084/jem.20140639 (2014).

40 Hashimoto, D. et al. Tissue-resident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes. Immunity 38, 792-804, doi:10.1016/j.immuni.2013.04.004 (2013).

41 Bouwens, L., Baekeland, M., De Zanger, R. & Wisse, E. Quantitation, tissue distribution and proliferation kinetics of Kupffer cells in normal rat liver. Hepatology 6, 718-722 (1986).

42 Jenkins, S. J. et al. Local macrophage proliferation, rather than recruitment from the blood, is a signature of TH2 inflammation. Science 332, 1284-1288, doi:10.1126/science.1204351 (2011).

43 Bruttger, J. et al. Genetic Cell Ablation Reveals Clusters of Local Self-Renewing Microglia in the Mammalian Central Nervous System. Immunity 43, 92-106, doi:10.1016/j.immuni.2015.06.012 (2015).

44 Ginhoux, F. et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science 330, 841-845, doi:10.1126/science.1194637 (2010).

45 Kierdorf, K. et al. Microglia emerge from erythromyeloid precursors via Pu.1- and Irf8-dependent pathways. Nature neuroscience 16, 273-280, doi:10.1038/nn.3318 (2013).

46 Schulz, C. et al. A lineage of myeloid cells independent of Myb and hematopoietic stem cells. Science 336, 86-90, doi:10.1126/science.1219179 (2012).

47 Chorro, L. et al. Langerhans cell (LC) proliferation mediates neonatal development, homeostasis, and inflammation-associated expansion of the epidermal LC network. The Journal of experimental medicine 206, 3089-3100, doi:10.1084/jem.20091586 (2009).

48 Davies, L. C. et al. A quantifiable proliferative burst of tissue macrophages restores homeostatic macrophage populations after acute inflammation. European journal of immunology 41, 2155-2164, doi:10.1002/eji.201141817 (2011).

49 Sawyer, R. T., Strausbauch, P. H. & Volkman, A. Resident macrophage proliferation in mice depleted of blood monocytes by strontium-89. Laboratory investigation; a journal of technical methods and pathology 46, 165-170 (1982).

50 Yamada, M., Naito, M. & Takahashi, K. Kupffer cell proliferation and glucan-induced granuloma formation in mice depleted of blood monocytes by strontium-89. Journal of leukocyte biology 47, 195-205 (1990).

150

51 Bogunovic, M. et al. Origin of the lamina propria dendritic cell network. Immunity 31, 513-525, doi:10.1016/j.immuni.2009.08.010 (2009).

52 Varol, C. et al. Intestinal lamina propria dendritic cell subsets have different origin and functions. Immunity 31, 502-512, doi:10.1016/j.immuni.2009.06.025 (2009).

53 Merad, M., Sathe, P., Helft, J., Miller, J. & Mortha, A. The dendritic cell lineage: ontogeny and function of dendritic cells and their subsets in the steady state and the inflamed setting. Annual review of immunology 31, 563-604, doi:10.1146/annurev-immunol-020711-074950 (2013).

54 Shortman, K. & Liu, Y. J. Mouse and human dendritic cell subtypes. Nature reviews. Immunology 2, 151-161, doi:10.1038/nri746 (2002).

55 Satpathy, A. T., Wu, X., Albring, J. C. & Murphy, K. M. Re(de)fining the dendritic cell lineage. Nature immunology 13, 1145-1154, doi:10.1038/ni.2467 (2012).

56 Schlitzer, A. et al. IRF4 transcription factor-dependent CD11b+ dendritic cells in human and mouse control mucosal IL-17 cytokine responses. Immunity 38, 970-983, doi:10.1016/j.immuni.2013.04.011 (2013).

57 Schlitzer, A. et al. Identification of cDC1- and cDC2-committed DC progenitors reveals early lineage priming at the common DC progenitor stage in the bone marrow. Nature immunology 16, 718-728, doi:10.1038/ni.3200 (2015).

58 Watchmaker, P. B. et al. Comparative transcriptional and functional profiling defines conserved programs of intestinal DC differentiation in humans and mice. Nature immunology 15, 98-108, doi:10.1038/ni.2768 (2014).

59 Bain, C. C. & Mowat, A. M. CD200 receptor and macrophage function in the intestine. Immunobiology 217, 643-651, doi:10.1016/j.imbio.2011.11.004 (2012).

60 Rivollier, A., He, J., Kole, A., Valatas, V. & Kelsall, B. L. Inflammation switches the differentiation program of Ly6Chi monocytes from antiinflammatory macrophages to inflammatory dendritic cells in the colon. The Journal of experimental medicine 209, 139-155, doi:10.1084/jem.20101387 (2012).

61 Persson, E. K. et al. IRF4 transcription-factor-dependent CD103(+)CD11b(+) dendritic cells drive mucosal T helper 17 cell differentiation. Immunity 38, 958-969, doi:10.1016/j.immuni.2013.03.009 (2013).

62 Cerovic, V., Bain, C. C., Mowat, A. M. & Milling, S. W. Intestinal macrophages and dendritic cells: what's the difference? Trends in immunology 35, 270-277, doi:10.1016/j.it.2014.04.003 (2014).

63 Hume, D. A., Perry, V. H. & Gordon, S. The mononuclear phagocyte system of the mouse defined by immunohistochemical localisation of antigen F4/80: macrophages associated with epithelia. The Anatomical record 210, 503-512, doi:10.1002/ar.1092100311 (1984).

151

64 Denning, T. L. et al. Functional specializations of intestinal dendritic cell and macrophage subsets that control Th17 and regulatory T cell responses are dependent on the T cell/APC ratio, source of mouse strain, and regional localization. J Immunol 187, 733-747, doi:10.4049/jimmunol.1002701 (2011).

65 Bain, C. C. & Mowat, A. M. The monocyte-macrophage axis in the intestine. Cellular immunology 291, 41-48, doi:10.1016/j.cellimm.2014.03.012 (2014).

66 Scott, C. L. et al. CCR2(+)CD103(-) intestinal dendritic cells develop from DC-committed precursors and induce interleukin-17 production by T cells. Mucosal immunology 8, 327-339, doi:10.1038/mi.2014.70 (2015).

67 Grainger, J. R. et al. Inflammatory monocytes regulate pathologic responses to commensals during acute gastrointestinal infection. Nature medicine 19, 713-721, doi:10.1038/nm.3189 (2013).

68 Seo, S. U. et al. Intestinal macrophages arising from CCR2(+) monocytes control pathogen infection by activating innate lymphoid cells. Nature communications 6, 8010, doi:10.1038/ncomms9010 (2015).

69 Panea, C. et al. Intestinal Monocyte-Derived Macrophages Control Commensal-Specific Th17 Responses. Cell reports 12, 1314-1324, doi:10.1016/j.celrep.2015.07.040 (2015).

70 Koscso, B., Gowda, K. & Bogunovic, M. In vivo depletion and genetic targeting of mouse intestinal CX3CR1(+) mononuclear phagocytes. Journal of immunological methods 432, 13-23, doi:10.1016/j.jim.2015.12.009 (2016).

71 Bakri, Y. et al. Balance of MafB and PU.1 specifies alternative macrophage or dendritic cell fate. Blood 105, 2707-2716, doi:10.1182/blood-2004-04-1448 (2005).

72 Moriguchi, T. et al. MafB is essential for renal development and F4/80 expression in macrophages. Molecular and cellular biology 26, 5715-5727, doi:10.1128/MCB.00001-06 (2006).

73 Nagamura-Inoue, T., Tamura, T. & Ozato, K. Transcription factors that regulate growth and differentiation of myeloid cells. International reviews of immunology 20, 83-105 (2001).

74 Ginhoux, F. et al. The origin and development of nonlymphoid tissue CD103+ DCs. The Journal of experimental medicine 206, 3115-3130, doi:10.1084/jem.20091756 (2009).

75 Schreiber, H. A. et al. Intestinal monocytes and macrophages are required for T cell polarization in response to Citrobacter rodentium. The Journal of experimental medicine 210, 2025-2039, doi:10.1084/jem.20130903 (2013).

76 Sathaliyawala, T. et al. Mammalian target of rapamycin controls dendritic cell development downstream of Flt3 ligand signaling. Immunity 33, 597-606, doi:10.1016/j.immuni.2010.09.012 (2010).

152

77 Esterhazy, D. et al. Classical dendritic cells are required for dietary antigen-mediated induction of peripheral Treg cells and tolerance. Nature immunology 17, 545-555, doi:10.1038/ni.3408 (2016).

78 Cerovic, V. et al. Intestinal CD103(-) dendritic cells migrate in lymph and prime effector T cells. Mucosal immunology 6, 104-113, doi:10.1038/mi.2012.53 (2013).

79 Kusunoki, T. et al. TH2 dominance and defective development of a CD8+ dendritic cell subset in Id2-deficient mice. The Journal of allergy and clinical immunology 111, 136-142, doi:10.1067/mai.2003.29 (2003).

80 Hacker, C. et al. Transcriptional profiling identifies Id2 function in dendritic cell development. Nature immunology 4, 380-386, doi:10.1038/ni903 (2003).

81 Edelson, B. T. et al. Peripheral CD103+ dendritic cells form a unified subset developmentally related to CD8alpha+ conventional dendritic cells. The Journal of experimental medicine 207, 823-836, doi:10.1084/jem.20091627 (2010).

82 Kashiwada, M., Pham, N. L., Pewe, L. L., Harty, J. T. & Rothman, P. B. NFIL3/E4BP4 is a key transcription factor for CD8alpha(+) dendritic cell development. Blood 117, 6193-6197, doi:10.1182/blood-2010-07-295873 (2011).

83 Jackson, J. T. et al. Id2 expression delineates differential checkpoints in the genetic program of CD8alpha+ and CD103+ dendritic cell lineages. The EMBO journal 30, 2690-2704, doi:10.1038/emboj.2011.163 (2011).

84 Kobayashi, T. et al. NFIL3 is a regulator of IL-12 p40 in macrophages and mucosal immunity. J Immunol 186, 4649-4655, doi:10.4049/jimmunol.1003888 (2011).

85 Grajales-Reyes, G. E. et al. Batf3 maintains autoactivation of Irf8 for commitment of a CD8alpha(+) conventional DC clonogenic progenitor. Nature immunology 16, 708-717, doi:10.1038/ni.3197 (2015).

86 Lewis, K. L. et al. Notch2 receptor signaling controls functional differentiation of dendritic cells in the spleen and intestine. Immunity 35, 780-791, doi:10.1016/j.immuni.2011.08.013 (2011).

87 Satpathy, A. T. et al. Notch2-dependent classical dendritic cells orchestrate intestinal immunity to attaching-and-effacing bacterial pathogens. Nature immunology 14, 937-948, doi:10.1038/ni.2679 (2013).

88 Scott, C. L., Tfp, Z. M., Beckham, K. S., Douce, G. & Mowat, A. M. Signal regulatory protein alpha (SIRPalpha) regulates the homeostasis of CD103(+) CD11b(+) DCs in the intestinal lamina propria. European journal of immunology 44, 3658-3668, doi:10.1002/eji.201444859 (2014).

89 Jaensson, E. et al. Small intestinal CD103+ dendritic cells display unique functional properties that are conserved between mice and humans. The Journal of experimental medicine 205, 2139-2149, doi:10.1084/jem.20080414 (2008).

153

90 Leishman, A. J. et al. Precursors of functional MHC class I- or class II-restricted CD8alphaalpha(+) T cells are positively selected in the thymus by agonist self-peptides. Immunity 16, 355-364 (2002).

91 Staton, T. L. et al. CD8+ recent thymic emigrants home to and efficiently repopulate the small intestine epithelium. Nature immunology 7, 482-488, doi:10.1038/ni1319 (2006).

92 Cheroutre, H., Lambolez, F. & Mucida, D. The light and dark sides of intestinal intraepithelial lymphocytes. Nature reviews. Immunology 11, 445-456, doi:10.1038/nri3007 (2011).

93 Luda, K. M. et al. IRF8 Transcription-Factor-Dependent Classical Dendritic Cells Are Essential for Intestinal T Cell Homeostasis. Immunity 44, 860-874, doi:10.1016/j.immuni.2016.02.008 (2016).

94 Shale, M., Schiering, C. & Powrie, F. CD4(+) T-cell subsets in intestinal inflammation. Immunological reviews 252, 164-182, doi:10.1111/imr.12039 (2013).

95 Parrott, D. M. et al. Analysis of the effector functions of different populations of mucosal lymphocytes. Annals of the New York Academy of Sciences 409, 307-320 (1983).

96 Targan, S. R., Deem, R. L., Liu, M., Wang, S. & Nel, A. Definition of a lamina propria T cell responsive state. Enhanced cytokine responsiveness of T cells stimulated through the CD2 pathway. J Immunol 154, 664-675 (1995).

97 Wang, H. C., Zhou, Q., Dragoo, J. & Klein, J. R. Most murine CD8+ intestinal intraepithelial lymphocytes are partially but not fully activated T cells. J Immunol 169, 4717-4722 (2002).

98 Zhou, L., Chong, M. M. & Littman, D. R. Plasticity of CD4+ T cell lineage differentiation. Immunity 30, 646-655, doi:10.1016/j.immuni.2009.05.001 (2009).

99 Omenetti, S. & Pizarro, T. T. The Treg/Th17 Axis: A Dynamic Balance Regulated by the Gut Microbiome. Frontiers in immunology 6, 639, doi:10.3389/fimmu.2015.00639 (2015).

100 Gagliani, N. et al. Th17 cells transdifferentiate into regulatory T cells during resolution of inflammation. Nature 523, 221-225, doi:10.1038/nature14452 (2015).

101 Lee, Y. K. et al. Late developmental plasticity in the T helper 17 lineage. Immunity 30, 92-107, doi:10.1016/j.immuni.2008.11.005 (2009).

102 Lui, J. B., Devarajan, P., Teplicki, S. A. & Chen, Z. Cross-differentiation from the CD8 lineage to CD4 T cells in the gut-associated microenvironment with a nonessential role of microbiota. Cell reports 10, 574-585, doi:10.1016/j.celrep.2014.12.053 (2015).

154

103 Lui, J. B., McGinn, L. S. & Chen, Z. Gut microbiota amplifies host-intrinsic conversion from the CD8 T cell lineage to CD4 T cells for induction of mucosal immune tolerance. Gut microbes 7, 40-47, doi:10.1080/19490976.2015.1117737 (2016).

104 McDonald, K. G., McDonough, J. S. & Newberry, R. D. Adaptive immune responses are dispensable for isolated lymphoid follicle formation: antigen-naive, lymphotoxin-sufficient B lymphocytes drive the formation of mature isolated lymphoid follicles. J Immunol 174, 5720-5728 (2005).

105 Eberl, G. et al. An essential function for the nuclear receptor RORgamma(t) in the generation of fetal lymphoid tissue inducer cells. Nature immunology 5, 64-73, doi:10.1038/ni1022 (2004).

106 Jang, M. H. et al. CCR7 is critically important for migration of dendritic cells in intestinal lamina propria to mesenteric lymph nodes. J Immunol 176, 803-810 (2006).

107 Worbs, T. et al. Oral tolerance originates in the intestinal immune system and relies on antigen carriage by dendritic cells. The Journal of experimental medicine 203, 519-527, doi:10.1084/jem.20052016 (2006).

108 Spahn, T. W. et al. Induction of colitis in mice deficient of Peyer's patches and mesenteric lymph nodes is associated with increased disease severity and formation of colonic lymphoid patches. The American journal of pathology 161, 2273-2282, doi:10.1016/S0002-9440(10)64503-8 (2002).

109 Spahn, T. W. et al. Mesenteric lymph nodes are critical for the induction of high-dose oral tolerance in the absence of Peyer's patches. European journal of immunology 32, 1109-1113, doi:10.1002/1521-4141(200204)32:4&#60;1109::AID-IMMU1109&#62;3.0.CO;2-K (2002).

110 Berlin, C. et al. Alpha 4 beta 7 integrin mediates lymphocyte binding to the mucosal vascular addressin MAdCAM-1. Cell 74, 185-195 (1993).

111 Papadakis, K. A. et al. The role of thymus-expressed chemokine and its receptor CCR9 on lymphocytes in the regional specialization of the mucosal immune system. J Immunol 165, 5069-5076 (2000).

112 Kim, S. V. et al. GPR15-mediated homing controls immune homeostasis in the large intestine mucosa. Science 340, 1456-1459, doi:10.1126/science.1237013 (2013).

113 Welty, N. E. et al. Intestinal lamina propria dendritic cells maintain T cell homeostasis but do not affect commensalism. The Journal of experimental medicine 210, 2011-2024, doi:10.1084/jem.20130728 (2013).

114 Svensson, M. et al. Retinoic acid receptor signaling levels and antigen dose regulate gut homing receptor expression on CD8+ T cells. Mucosal immunology 1, 38-48, doi:10.1038/mi.2007.4 (2008).

155

115 Elgueta, R. et al. Imprinting of CCR9 on CD4 T cells requires IL-4 signaling on mesenteric lymph node dendritic cells. J Immunol 180, 6501-6507 (2008).

116 Yokota, A. et al. GM-CSF and IL-4 synergistically trigger dendritic cells to acquire retinoic acid-producing capacity. International immunology 21, 361-377, doi:10.1093/intimm/dxp003 (2009).

117 Wang, S. et al. MyD88-dependent TLR1/2 signals educate dendritic cells with gut-specific imprinting properties. J Immunol 187, 141-150, doi:10.4049/jimmunol.1003740 (2011).

118 Coombes, J. L. et al. A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. The Journal of experimental medicine 204, 1757-1764, doi:10.1084/jem.20070590 (2007).

119 Mucida, D. & Cheroutre, H. TGFbeta and retinoic acid intersect in immune-regulation. Cell adhesion & migration 1, 142-144 (2007).

120 Mucida, D. et al. Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 317, 256-260, doi:10.1126/science.1145697 (2007).

121 Sun, C. M. et al. Small intestine lamina propria dendritic cells promote de novo generation of Foxp3 T reg cells via retinoic acid. The Journal of experimental medicine 204, 1775-1785, doi:10.1084/jem.20070602 (2007).

122 Paidassi, H. et al. Preferential expression of integrin alphavbeta8 promotes generation of regulatory T cells by mouse CD103+ dendritic cells. Gastroenterology 141, 1813-1820, doi:10.1053/j.gastro.2011.06.076 (2011).

123 Worthington, J. J., Czajkowska, B. I., Melton, A. C. & Travis, M. A. Intestinal dendritic cells specialize to activate transforming growth factor-beta and induce Foxp3+ regulatory T cells via integrin alphavbeta8. Gastroenterology 141, 1802-1812, doi:10.1053/j.gastro.2011.06.057 (2011).

124 Molenaar, R. et al. Lymph node stromal cells support dendritic cell-induced gut-homing of T cells. J Immunol 183, 6395-6402, doi:10.4049/jimmunol.0900311 (2009).

125 Hammerschmidt, S. I. et al. Stromal mesenteric lymph node cells are essential for the generation of gut-homing T cells in vivo. The Journal of experimental medicine 205, 2483-2490, doi:10.1084/jem.20080039 (2008).

126 Lefrancois, L. & Puddington, L. Intestinal and pulmonary mucosal T cells: local heroes fight to maintain the status quo. Annual review of immunology 24, 681-704, doi:10.1146/annurev.immunol.24.021605.090650 (2006).

127 Fujihashi, K. et al. Peyer's patches are required for oral tolerance to proteins. Proceedings of the National Academy of Sciences of the United States of America 98, 3310-3315, doi:10.1073/pnas.061412598 (2001).

156

128 Tsuji, M. et al. Requirement for lymphoid tissue-inducer cells in isolated follicle formation and T cell-independent immunoglobulin A generation in the gut. Immunity 29, 261-271, doi:10.1016/j.immuni.2008.05.014 (2008).

129 Hamada, H. et al. Identification of multiple isolated lymphoid follicles on the antimesenteric wall of the mouse small intestine. J Immunol 168, 57-64 (2002).

130 Pabst, O. et al. Cryptopatches and isolated lymphoid follicles: dynamic lymphoid tissues dispensable for the generation of intraepithelial lymphocytes. European journal of immunology 35, 98-107, doi:10.1002/eji.200425432 (2005).

131 Pabst, O. et al. Adaptation of solitary intestinal lymphoid tissue in response to microbiota and chemokine receptor CCR7 signaling. J Immunol 177, 6824-6832 (2006).

132 Moser, M. & Murphy, K. M. Dendritic cell regulation of TH1-TH2 development. Nature immunology 1, 199-205, doi:10.1038/79734 (2000).

133 Murphy, K. M. et al. Signaling and transcription in T helper development. Annual review of immunology 18, 451-494, doi:10.1146/annurev.immunol.18.1.451 (2000).

134 Szabo, S. J. et al. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 100, 655-669 (2000).

135 Maynard, C. L. & Weaver, C. T. Intestinal effector T cells in health and disease. Immunity 31, 389-400, doi:10.1016/j.immuni.2009.08.012 (2009).

136 Cohen, S. B. et al. CXCR3-dependent CD4(+) T cells are required to activate inflammatory monocytes for defense against intestinal infection. PLoS pathogens 9, e1003706, doi:10.1371/journal.ppat.1003706 (2013).

137 Powrie, F. et al. Inhibition of Th1 responses prevents inflammatory bowel disease in scid mice reconstituted with CD45RBhi CD4+ T cells. Immunity 1, 553-562 (1994).

138 Christen, U. & von Herrath, M. G. Initiation of autoimmunity. Curr Opin Immunol 16, 759-767, doi:10.1016/j.coi.2004.09.002 (2004).

139 Becker, C. et al. Constitutive p40 promoter activation and IL-23 production in the terminal ileum mediated by dendritic cells. The Journal of clinical investigation 112, 693-706, doi:10.1172/JCI17464 (2003).

140 Oppmann, B. et al. Novel p19 protein engages IL-12p40 to form a cytokine, IL-23, with biological activities similar as well as distinct from IL-12. Immunity 13, 715-725 (2000).

141 Batten, M. et al. Interleukin 27 limits autoimmune encephalomyelitis by suppressing the development of interleukin 17-producing T cells. Nature immunology 7, 929-936, doi:10.1038/ni1375 (2006).

157

142 Troy, A. E. et al. IL-27 regulates homeostasis of the intestinal CD4+ effector T cell pool and limits intestinal inflammation in a murine model of colitis. J Immunol 183, 2037-2044, doi:10.4049/jimmunol.0802918 (2009).

143 Mazmanian, S. K., Liu, C. H., Tzianabos, A. O. & Kasper, D. L. An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system. Cell 122, 107-118, doi:10.1016/j.cell.2005.05.007 (2005).

144 Fujimoto, K. et al. A new subset of CD103+CD8alpha+ dendritic cells in the small intestine expresses TLR3, TLR7, and TLR9 and induces Th1 response and CTL activity. J Immunol 186, 6287-6295, doi:10.4049/jimmunol.1004036 (2011).

145 Muzaki, A. R. et al. Intestinal CD103(+)CD11b(-) dendritic cells restrain colitis via IFN-gamma-induced anti-inflammatory response in epithelial cells. Mucosal immunology 9, 336-351, doi:10.1038/mi.2015.64 (2016).

146 Hand, T. W. et al. Acute gastrointestinal infection induces long-lived microbiota-specific T cell responses. Science 337, 1553-1556, doi:10.1126/science.1220961 (2012).

147 Ivanov, II et al. Induction of intestinal Th17 cells by segmented filamentous bacteria. Cell 139, 485-498 (2009).

148 Zheng, W. & Flavell, R. A. The transcription factor GATA-3 is necessary and sufficient for Th2 cytokine gene expression in CD4 T cells. Cell 89, 587-596 (1997).

149 Ansel, K. M., Djuretic, I., Tanasa, B. & Rao, A. Regulation of Th2 differentiation and Il4 locus accessibility. Annual review of immunology 24, 607-656, doi:10.1146/annurev.immunol.23.021704.115821 (2006).

150 O'Garra, A. Cytokines induce the development of functionally heterogeneous T helper cell subsets. Immunity 8, 275-283 (1998).

151 Chu, D. K. et al. T helper cell IL-4 drives intestinal Th2 priming to oral peanut antigen, under the control of OX40L and independent of innate-like lymphocytes. Mucosal immunology 7, 1395-1404, doi:10.1038/mi.2014.29 (2014).

152 Kumamoto, Y. et al. CD301b(+) dermal dendritic cells drive T helper 2 cell-mediated immunity. Immunity 39, 733-743, doi:10.1016/j.immuni.2013.08.029 (2013).

153 Gao, Y. et al. Control of T helper 2 responses by transcription factor IRF4-dependent dendritic cells. Immunity 39, 722-732, doi:10.1016/j.immuni.2013.08.028 (2013).

154 Weaver, C. T. & Murphy, K. M. The central role of the Th17 lineage in regulating the inflammatory/autoimmune axis. Seminars in immunology 19, 351-352, doi:10.1016/j.smim.2008.01.001 (2007).

158

155 Stockinger, B. & Veldhoen, M. Differentiation and function of Th17 T cells. Curr Opin Immunol 19, 281-286, doi:10.1016/j.coi.2007.04.005 (2007).

156 Ishigame, H. et al. Differential roles of interleukin-17A and -17F in host defense against mucoepithelial bacterial infection and allergic responses. Immunity 30, 108-119, doi:10.1016/j.immuni.2008.11.009 (2009).

157 Furuzawa-Carballeda, J., Vargas-Rojas, M. I. & Cabral, A. R. Autoimmune inflammation from the Th17 perspective. Autoimmunity reviews 6, 169-175, doi:10.1016/j.autrev.2006.10.002 (2007).

158 Bettelli, E. et al. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441, 235-238, doi:10.1038/nature04753 (2006).

159 Mangan, P. R. et al. Transforming growth factor-beta induces development of the T(H)17 lineage. Nature 441, 231-234, doi:10.1038/nature04754 (2006).

160 Chung, Y. et al. Critical regulation of early Th17 cell differentiation by interleukin-1 signaling. Immunity 30, 576-587, doi:10.1016/j.immuni.2009.02.007 (2009).

161 Zhou, L. et al. IL-6 programs T(H)-17 cell differentiation by promoting sequential engagement of the IL-21 and IL-23 pathways. Nature immunology 8, 967-974, doi:10.1038/ni1488 (2007).

162 Ivanov, II et al. The orphan nuclear receptor RORgammat directs the differentiation program of proinflammatory IL-17+ T helper cells. Cell 126, 1121-1133, doi:10.1016/j.cell.2006.07.035 (2006).

163 Ciofani, M. et al. A validated regulatory network for Th17 cell specification. Cell 151, 289-303, doi:10.1016/j.cell.2012.09.016 (2012).

164 Brustle, A. et al. The development of inflammatory T(H)-17 cells requires interferon-regulatory factor 4. Nature immunology 8, 958-966, doi:10.1038/ni1500 (2007).

165 Mudter, J. et al. IRF4 regulates IL-17A promoter activity and controls RORgammat-dependent Th17 colitis in vivo. Inflammatory bowel diseases 17, 1343-1358, doi:10.1002/ibd.21476 (2011).

166 Schraml, B. U. et al. The AP-1 transcription factor Batf controls T(H)17 differentiation. Nature 460, 405-409, doi:10.1038/nature08114 (2009).

167 Bauquet, A. T. et al. The costimulatory molecule ICOS regulates the expression of c-Maf and IL-21 in the development of follicular T helper cells and TH-17 cells. Nature immunology 10, 167-175, doi:10.1038/ni.1690 (2009).

168 Quintana, F. J. et al. Control of T(reg) and T(H)17 cell differentiation by the aryl hydrocarbon receptor. Nature 453, 65-71, doi:10.1038/nature06880 (2008).

159

169 Veldhoen, M. et al. The aryl hydrocarbon receptor links TH17-cell-mediated autoimmunity to environmental toxins. Nature 453, 106-109, doi:10.1038/nature06881 (2008).

170 Yang, X. O. et al. T helper 17 lineage differentiation is programmed by orphan nuclear receptors ROR alpha and ROR gamma. Immunity 28, 29-39, doi:10.1016/j.immuni.2007.11.016 (2008).

171 Zheng, Y. et al. Interleukin-22, a T(H)17 cytokine, mediates IL-23-induced dermal inflammation and acanthosis. Nature 445, 648-651, doi:10.1038/nature05505 (2007).

172 Yosef, N. et al. Dynamic regulatory network controlling TH17 cell differentiation. Nature 496, 461-468, doi:10.1038/nature11981 (2013).

173 Acosta-Rodriguez, E. V. et al. Surface phenotype and antigenic specificity of human interleukin 17-producing T helper memory cells. Nature immunology 8, 639-646, doi:10.1038/ni1467 (2007).

174 Yamazaki, T. et al. CCR6 regulates the migration of inflammatory and regulatory T cells. J Immunol 181, 8391-8401 (2008).

175 Ivanov, II et al. Specific microbiota direct the differentiation of IL-17-producing T-helper cells in the mucosa of the small intestine. Cell host & microbe 4, 337-349, doi:10.1016/j.chom.2008.09.009 (2008).

176 Hu, W., Troutman, T. D., Edukulla, R. & Pasare, C. Priming microenvironments dictate cytokine requirements for T helper 17 cell lineage commitment. Immunity 35, 1010-1022, doi:10.1016/j.immuni.2011.10.013 (2011).

177 Shaw, M. H., Kamada, N., Kim, Y. G. & Nunez, G. Microbiota-induced IL-1beta, but not IL-6, is critical for the development of steady-state TH17 cells in the intestine. The Journal of experimental medicine 209, 251-258, doi:10.1084/jem.20111703 (2012).

178 Korn, T., Oukka, M., Kuchroo, V. & Bettelli, E. Th17 cells: effector T cells with inflammatory properties. Seminars in immunology 19, 362-371, doi:10.1016/j.smim.2007.10.007 (2007).

179 Fina, D. et al. Regulation of gut inflammation and th17 cell response by interleukin-21. Gastroenterology 134, 1038-1048, doi:10.1053/j.gastro.2008.01.041 (2008).

180 Stolfi, C. et al. Involvement of interleukin-21 in the regulation of colitis-associated colon cancer. The Journal of experimental medicine 208, 2279-2290, doi:10.1084/jem.20111106 (2011).

181 McKenzie, B. S., Kastelein, R. A. & Cua, D. J. Understanding the IL-23-IL-17 immune pathway. Trends in immunology 27, 17-23, doi:10.1016/j.it.2005.10.003 (2006).

160

182 McGeachy, M. J. & Cua, D. J. The link between IL-23 and Th17 cell-mediated immune pathologies. Seminars in immunology 19, 372-376, doi:10.1016/j.smim.2007.10.012 (2007).

183 Zaph, C. et al. Commensal-dependent expression of IL-25 regulates the IL-23-IL-17 axis in the intestine. The Journal of experimental medicine 205, 2191-2198, doi:10.1084/jem.20080720 (2008).

184 Duerr, R. H. et al. A genome-wide association study identifies IL23R as an inflammatory bowel disease gene. Science 314, 1461-1463, doi:10.1126/science.1135245 (2006).

185 Kullberg, M. C. et al. IL-23 plays a key role in Helicobacter hepaticus-induced T cell-dependent colitis. The Journal of experimental medicine 203, 2485-2494, doi:10.1084/jem.20061082 (2006).

186 Yen, D. et al. IL-23 is essential for T cell-mediated colitis and promotes inflammation via IL-17 and IL-6. The Journal of clinical investigation 116, 1310-1316, doi:10.1172/JCI21404 (2006).

187 Elson, C. O. et al. Monoclonal anti-interleukin 23 reverses active colitis in a T cell-mediated model in mice. Gastroenterology 132, 2359-2370, doi:10.1053/j.gastro.2007.03.104 (2007).

188 Wu, C. et al. Induction of pathogenic TH17 cells by inducible salt-sensing kinase SGK1. Nature 496, 513-517, doi:10.1038/nature11984 (2013).

189 Quatresooz, P. et al. Ustekinumab in psoriasis immunopathology with emphasis on the Th17-IL23 axis: a primer. Journal of biomedicine & biotechnology 2012, 147413, doi:10.1155/2012/147413 (2012).

190 Mease, P. J. et al. Secukinumab Inhibition of Interleukin-17A in Patients with Psoriatic Arthritis. The New England journal of medicine 373, 1329-1339, doi:10.1056/NEJMoa1412679 (2015).

191 Lubberts, E. The IL-23-IL-17 axis in inflammatory arthritis. Nature reviews. Rheumatology 11, 562, doi:10.1038/nrrheum.2015.128 (2015).

192 McGeachy, M. J. et al. TGF-beta and IL-6 drive the production of IL-17 and IL-10 by T cells and restrain T(H)-17 cell-mediated pathology. Nature immunology 8, 1390-1397, doi:10.1038/ni1539 (2007).

193 Lee, Y. et al. Induction and molecular signature of pathogenic TH17 cells. Nature immunology 13, 991-999, doi:10.1038/ni.2416 (2012).

194 Codarri, L. et al. RORgammat drives production of the cytokine GM-CSF in helper T cells, which is essential for the effector phase of autoimmune neuroinflammation. Nature immunology 12, 560-567, doi:10.1038/ni.2027 (2011).

161

195 El-Behi, M. et al. The encephalitogenicity of T(H)17 cells is dependent on IL-1- and IL-23-induced production of the cytokine GM-CSF. Nature immunology 12, 568-575, doi:10.1038/ni.2031 (2011).

196 Griseri, T., McKenzie, B. S., Schiering, C. & Powrie, F. Dysregulated hematopoietic stem and progenitor cell activity promotes interleukin-23-driven chronic intestinal inflammation. Immunity 37, 1116-1129, doi:10.1016/j.immuni.2012.08.025 (2012).

197 Coccia, M. et al. IL-1beta mediates chronic intestinal inflammation by promoting the accumulation of IL-17A secreting innate lymphoid cells and CD4(+) Th17 cells. The Journal of experimental medicine 209, 1595-1609, doi:10.1084/jem.20111453 (2012).

198 Gaboriau-Routhiau, V. et al. The key role of segmented filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity 31, 677-689, doi:10.1016/j.immuni.2009.08.020 (2009).

199 Atarashi, K. et al. ATP drives lamina propria T(H)17 cell differentiation. Nature 455, 808-812, doi:10.1038/nature07240 (2008).

200 Atarashi, K. et al. Th17 Cell Induction by Adhesion of Microbes to Intestinal Epithelial Cells. Cell 163, 367-380, doi:10.1016/j.cell.2015.08.058 (2015).

201 Hall, J. A. et al. Commensal DNA limits regulatory T cell conversion and is a natural adjuvant of intestinal immune responses. Immunity 29, 637-649, doi:10.1016/j.immuni.2008.08.009 (2008).

202 Goto, Y. et al. Segmented filamentous bacteria antigens presented by intestinal dendritic cells drive mucosal Th17 cell differentiation. Immunity 40, 594-607, doi:10.1016/j.immuni.2014.03.005 (2014).

203 Hepworth, M. R. et al. Innate lymphoid cells regulate CD4+ T-cell responses to intestinal commensal bacteria. Nature 498, 113-117, doi:10.1038/nature12240 (2013).

204 Hepworth, M. R. et al. Immune tolerance. Group 3 innate lymphoid cells mediate intestinal selection of commensal bacteria-specific CD4(+) T cells. Science 348, 1031-1035, doi:10.1126/science.aaa4812 (2015).

205 Thornton, A. M. et al. Expression of Helios, an Ikaros transcription factor family member, differentiates thymic-derived from peripherally induced Foxp3+ T regulatory cells. J Immunol 184, 3433-3441, doi:10.4049/jimmunol.0904028 (2010).

206 Weiss, J. M. et al. Neuropilin 1 is expressed on thymus-derived natural regulatory T cells, but not mucosa-generated induced Foxp3+ T reg cells. The Journal of experimental medicine 209, 1723-1742, S1721, doi:10.1084/jem.20120914 (2012).

162

207 Fontenot, J. D., Gavin, M. A. & Rudensky, A. Y. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nature immunology 4, 330-336, doi:10.1038/ni904 (2003).

208 Zheng, S. G., Wang, J., Wang, P., Gray, J. D. & Horwitz, D. A. IL-2 is essential for TGF-beta to convert naive CD4+CD25- cells to CD25+Foxp3+ regulatory T cells and for expansion of these cells. J Immunol 178, 2018-2027 (2007).

209 Barnes, M. J. & Powrie, F. Regulatory T cells reinforce intestinal homeostasis. Immunity 31, 401-411, doi:10.1016/j.immuni.2009.08.011 (2009).

210 Vieira, P. L. et al. IL-10-secreting regulatory T cells do not express Foxp3 but have comparable regulatory function to naturally occurring CD4+CD25+ regulatory T cells. J Immunol 172, 5986-5993 (2004).

211 Maynard, C. L. et al. Regulatory T cells expressing interleukin 10 develop from Foxp3+ and Foxp3- precursor cells in the absence of interleukin 10. Nature immunology 8, 931-941, doi:10.1038/ni1504 (2007).

212 Rubtsov, Y. P. et al. Regulatory T cell-derived interleukin-10 limits inflammation at environmental interfaces. Immunity 28, 546-558, doi:10.1016/j.immuni.2008.02.017 (2008).

213 Chaudhry, A. et al. Interleukin-10 signaling in regulatory T cells is required for suppression of Th17 cell-mediated inflammation. Immunity 34, 566-578, doi:10.1016/j.immuni.2011.03.018 (2011).

214 Atarashi, K. et al. Induction of colonic regulatory T cells by indigenous Clostridium species. Science 331, 337-341, doi:10.1126/science.1198469 (2011).

215 Arpaia, N. et al. Metabolites produced by commensal bacteria promote peripheral regulatory T-cell generation. Nature 504, 451-455, doi:10.1038/nature12726 (2013).

216 Furusawa, Y. et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells. Nature 504, 446-450, doi:10.1038/nature12721 (2013).

217 Smith, P. M. et al. The microbial metabolites, short-chain fatty acids, regulate colonic Treg cell homeostasis. Science 341, 569-573, doi:10.1126/science.1241165 (2013).

218 Round, J. L. et al. The Toll-like receptor 2 pathway establishes colonization by a commensal of the human microbiota. Science 332, 974-977, doi:10.1126/science.1206095 (2011).

219 Neutra, M. R., Frey, A. & Kraehenbuhl, J. P. Epithelial M cells: gateways for mucosal infection and immunization. Cell 86, 345-348 (1996).

163

220 Mabbott, N. A., Donaldson, D. S., Ohno, H., Williams, I. R. & Mahajan, A. Microfold (M) cells: important immunosurveillance posts in the intestinal epithelium. Mucosal immunology 6, 666-677, doi:10.1038/mi.2013.30 (2013).

221 Kanaya, T. et al. The Ets transcription factor Spi-B is essential for the differentiation of intestinal microfold cells. Nature immunology 13, 729-736, doi:10.1038/ni.2352 (2012).

222 Hase, K. et al. Uptake through glycoprotein 2 of FimH(+) bacteria by M cells initiates mucosal immune response. Nature 462, 226-230, doi:10.1038/nature08529 (2009).

223 Yoshida, M. et al. Human neonatal Fc receptor mediates transport of IgG into luminal secretions for delivery of antigens to mucosal dendritic cells. Immunity 20, 769-783, doi:10.1016/j.immuni.2004.05.007 (2004).

224 Rescigno, M. et al. Dendritic cells express tight junction proteins and penetrate gut epithelial monolayers to sample bacteria. Nature immunology 2, 361-367, doi:10.1038/86373 (2001).

225 Vallon-Eberhard, A., Landsman, L., Yogev, N., Verrier, B. & Jung, S. Transepithelial pathogen uptake into the small intestinal lamina propria. J Immunol 176, 2465-2469 (2006).

226 Chieppa, M., Rescigno, M., Huang, A. Y. & Germain, R. N. Dynamic imaging of dendritic cell extension into the small bowel lumen in response to epithelial cell TLR engagement. The Journal of experimental medicine 203, 2841-2852, doi:10.1084/jem.20061884 (2006).

227 McDole, J. R. et al. Goblet cells deliver luminal antigen to CD103+ dendritic cells in the small intestine. Nature 483, 345-349, doi:10.1038/nature10863 (2012).

228 Niess, J. H. et al. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance. Science 307, 254-258, doi:10.1126/science.1102901 (2005).

229 Schulz, O. et al. Intestinal CD103+, but not CX3CR1+, antigen sampling cells migrate in lymph and serve classical dendritic cell functions. The Journal of experimental medicine 206, 3101-3114, doi:10.1084/jem.20091925 (2009).

230 Mazzini, E., Massimiliano, L., Penna, G. & Rescigno, M. Oral tolerance can be established via gap junction transfer of fed antigens from CX3CR1(+) macrophages to CD103(+) dendritic cells. Immunity 40, 248-261, doi:10.1016/j.immuni.2013.12.012 (2014).

231 Karlsson, M. et al. "Tolerosomes" are produced by intestinal epithelial cells. European journal of immunology 31, 2892-2900, doi:10.1002/1521-4141(2001010)31:10&#60;2892::AID-IMMU2892&#62;3.0.CO;2-I (2001).

164

232 Hershberg, R. M. & Mayer, L. F. Antigen processing and presentation by intestinal epithelial cells - polarity and complexity. Immunology today 21, 123-128 (2000).

233 Dudziak, D. et al. Differential antigen processing by dendritic cell subsets in vivo. Science 315, 107-111, doi:10.1126/science.1136080 (2007).

234 Uematsu, S. et al. Regulation of humoral and cellular gut immunity by lamina propria dendritic cells expressing Toll-like receptor 5. Nature immunology 9, 769-776, doi:10.1038/ni.1622 (2008).

235 Kinnebrew, M. A. et al. Interleukin 23 production by intestinal CD103(+)CD11b(+) dendritic cells in response to bacterial flagellin enhances mucosal innate immune defense. Immunity 36, 276-287, doi:10.1016/j.immuni.2011.12.011 (2012).

236 Diehl, G. E. et al. Microbiota restricts trafficking of bacteria to mesenteric lymph nodes by CX(3)CR1(hi) cells. Nature 494, 116-120, doi:10.1038/nature11809 (2013).

237 Liu, H. et al. TLR5 mediates CD172alpha(+) intestinal lamina propria dendritic cell induction of Th17 cells. Scientific reports 6, 22040, doi:10.1038/srep22040 (2016).

238 Laffont, S., Siddiqui, K. R. & Powrie, F. Intestinal inflammation abrogates the tolerogenic properties of MLN CD103+ dendritic cells. European journal of immunology 40, 1877-1883, doi:10.1002/eji.200939957 (2010).

239 Loschko, J. et al. Absence of MHC class II on cDCs results in microbial-dependent intestinal inflammation. The Journal of experimental medicine 213, 517-534, doi:10.1084/jem.20160062 (2016).

240 Samstein, M. et al. Essential yet limited role for CCR2(+) inflammatory monocytes during Mycobacterium tuberculosis-specific T cell priming. eLife 2, e01086, doi:10.7554/eLife.01086 (2013).

241 Geem, D. et al. Specific microbiota-induced intestinal Th17 differentiation requires MHC class II but not GALT and mesenteric lymph nodes. J Immunol 193, 431-438, doi:10.4049/jimmunol.1303167 (2014).

242 Geem, D. et al. Contribution of Mesenteric Lymph Nodes and GALT to the Intestinal Foxp3+ Regulatory T-Cell Compartment. Cellular and Molecular Gastroenterology and Hepatology 2, 274-280.e273, doi:10.1016/j.jcmgh.2015.12.009 (2016).

243 Pabst, O. & Mowat, A. M. Oral tolerance to food protein. Mucosal immunology 5, 232-239, doi:10.1038/mi.2012.4 (2012).

244 Hadis, U. et al. Intestinal tolerance requires gut homing and expansion of FoxP3+ regulatory T cells in the lamina propria. Immunity 34, 237-246, doi:10.1016/j.immuni.2011.01.016 (2011).

165

245 Mellor, A. L. & Munn, D. H. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nature reviews. Immunology 4, 762-774, doi:10.1038/nri1457 (2004).

246 Cong, Y., Wang, L., Konrad, A., Schoeb, T. & Elson, C. O. Curcumin induces the tolerogenic dendritic cell that promotes differentiation of intestine-protective regulatory T cells. European journal of immunology 39, 3134-3146, doi:10.1002/eji.200939052 (2009).

247 Baba, N., Samson, S., Bourdet-Sicard, R., Rubio, M. & Sarfati, M. Commensal bacteria trigger a full dendritic cell maturation program that promotes the expansion of non-Tr1 suppressor T cells. Journal of leukocyte biology 84, 468-476, doi:10.1189/jlb.0108017 (2008).

248 Delgado, M., Gonzalez-Rey, E. & Ganea, D. The neuropeptide vasoactive intestinal peptide generates tolerogenic dendritic cells. J Immunol 175, 7311-7324 (2005).

249 Fernandez-Martin, A., Gonzalez-Rey, E., Chorny, A., Ganea, D. & Delgado, M. Vasoactive intestinal peptide induces regulatory T cells during experimental autoimmune encephalomyelitis. European journal of immunology 36, 318-326, doi:10.1002/eji.200535430 (2006).

250 Gonzalez-Rey, E., Fernandez-Martin, A., Chorny, A. & Delgado, M. Vasoactive intestinal peptide induces CD4+,CD25+ T regulatory cells with therapeutic effect in collagen-induced arthritis. Arthritis and rheumatism 54, 864-876, doi:10.1002/art.21652 (2006).

251 Wenzel, U. A., Jonstrand, C., Hansson, G. C. & Wick, M. J. CD103+ CD11b+ Dendritic Cells Induce Th17 T Cells in Muc2-Deficient Mice with Extensively Spread Colitis. PloS one 10, e0130750, doi:10.1371/journal.pone.0130750 (2015).

252 Ishifune, C. et al. Differentiation of CD11c+ CX3CR1+ cells in the small intestine requires Notch signaling. Proceedings of the National Academy of Sciences of the United States of America 111, 5986-5991, doi:10.1073/pnas.1401671111 (2014).

253 Sano, T. et al. An IL-23R/IL-22 Circuit Regulates Epithelial Serum Amyloid A to Promote Local Effector Th17 Responses. Cell 163, 381-393, doi:10.1016/j.cell.2015.08.061 (2015).

254 Harrison, O. J. et al. Epithelial-derived IL-18 regulates Th17 cell differentiation and Foxp3(+) Treg cell function in the intestine. Mucosal immunology 8, 1226-1236, doi:10.1038/mi.2015.13 (2015).

255 Reading, N. C. & Kasper, D. L. The starting lineup: key microbial players in intestinal immunity and homeostasis. Frontiers in microbiology 2, 148, doi:10.3389/fmicb.2011.00148 (2011).

166

256 Schnupf, P. et al. Growth and host interaction of mouse segmented filamentous bacteria in vitro. Nature 520, 99-103, doi:10.1038/nature14027 (2015).

257 Klaasen, H. L. et al. Intestinal, segmented, filamentous bacteria in a wide range of vertebrate species. Laboratory animals 27, 141-150 (1993).

258 Ericsson, A. C., Hagan, C. E., Davis, D. J. & Franklin, C. L. Segmented filamentous bacteria: commensal microbes with potential effects on research. Comparative medicine 64, 90-98 (2014).

259 Tannock, G. W., Miller, J. R. & Savage, D. C. Host specificity of filamentous, segmented microorganisms adherent to the small bowel epithelium in mice and rats. Applied and environmental microbiology 47, 441-442 (1984).

260 Chung, H. et al. Gut immune maturation depends on colonization with a host-specific microbiota. Cell 149, 1578-1593, doi:10.1016/j.cell.2012.04.037 (2012).

261 Prakash, T. et al. Complete genome sequences of rat and mouse segmented filamentous bacteria, a potent inducer of th17 cell differentiation. Cell host & microbe 10, 273-284, doi:10.1016/j.chom.2011.08.007 (2011).

262 Pamp, S. J., Harrington, E. D., Quake, S. R., Relman, D. A. & Blainey, P. C. Single-cell sequencing provides clues about the host interactions of segmented filamentous bacteria (SFB). Genome research 22, 1107-1119, doi:10.1101/gr.131482.111 (2012).

263 Sczesnak, A. et al. The genome of th17 cell-inducing segmented filamentous bacteria reveals extensive auxotrophy and adaptations to the intestinal environment. Cell host & microbe 10, 260-272, doi:10.1016/j.chom.2011.08.005 (2011).

264 Snel, J. et al. Interactions between gut-associated lymphoid tissue and colonization levels of indigenous, segmented, filamentous bacteria in the small intestine of mice. Canadian journal of microbiology 44, 1177-1182 (1998).

265 Jiang, H. Q., Bos, N. A. & Cebra, J. J. Timing, localization, and persistence of colonization by segmented filamentous bacteria in the neonatal mouse gut depend on immune status of mothers and pups. Infection and immunity 69, 3611-3617, doi:10.1128/IAI.69.6.3611-3617.2001 (2001).

266 Ohashi, Y. et al. Colonization of segmented filamentous bacteria and its interaction with the luminal IgA level in conventional mice. Anaerobe 16, 543-546, doi:10.1016/j.anaerobe.2010.07.006 (2010).

267 Jonsson, H. Segmented filamentous bacteria in human ileostomy samples after high-fiber intake. FEMS microbiology letters 342, 24-29, doi:10.1111/1574-6968.12103 (2013).

268 Yin, Y. et al. Comparative analysis of the distribution of segmented filamentous bacteria in humans, mice and chickens. The ISME journal 7, 615-621, doi:10.1038/ismej.2012.128 (2013).

167

269 Okada, Y. et al. Effects of fecal microorganisms and their chloroform-resistant variants derived from mice, rats, and humans on immunological and physiological characteristics of the intestines of ex-germfree mice. Infection and immunity 62, 5442-5446 (1994).

270 Umesaki, Y., Okada, Y., Matsumoto, S., Imaoka, A. & Setoyama, H. Segmented filamentous bacteria are indigenous intestinal bacteria that activate intraepithelial lymphocytes and induce MHC class II molecules and fucosyl asialo GM1 glycolipids on the small intestinal epithelial cells in the ex-germ-free mouse. Microbiology and immunology 39, 555-562 (1995).

271 Goto, Y. et al. Innate lymphoid cells regulate intestinal epithelial cell glycosylation. Science 345, 1254009, doi:10.1126/science.1254009 (2014).

272 Lecuyer, E. et al. Segmented filamentous bacterium uses secondary and tertiary lymphoid tissues to induce gut IgA and specific T helper 17 cell responses. Immunity 40, 608-620, doi:10.1016/j.immuni.2014.03.009 (2014).

273 Suzuki, K. et al. Aberrant expansion of segmented filamentous bacteria in IgA-deficient gut. Proceedings of the National Academy of Sciences of the United States of America 101, 1981-1986, doi:10.1073/pnas.0307317101 (2004).

274 Wu, H. J. et al. Gut-residing segmented filamentous bacteria drive autoimmune arthritis via T helper 17 cells. Immunity 32, 815-827, doi:10.1016/j.immuni.2010.06.001 (2010).

275 Lee, Y. K., Menezes, J. S., Umesaki, Y. & Mazmanian, S. K. Proinflammatory T-cell responses to gut microbiota promote experimental autoimmune encephalomyelitis. Proceedings of the National Academy of Sciences of the United States of America 108 Suppl 1, 4615-4622, doi:10.1073/pnas.1000082107 (2011).

276 Palm, N. W. et al. Immunoglobulin A coating identifies colitogenic bacteria in inflammatory bowel disease. Cell 158, 1000-1010, doi:10.1016/j.cell.2014.08.006 (2014).

277 Mortha, A. et al. Microbiota-dependent crosstalk between macrophages and ILC3 promotes intestinal homeostasis. Science 343, 1249288, doi:10.1126/science.1249288 (2014).

278 Davis, C. P. & Savage, D. C. Habitat, succession, attachment, and morphology of segmented, filamentous microbes indigenous to the murine gastrointestinal tract. Infection and immunity 10, 948-956 (1974).

279 Jepson, M. A., Clark, M. A., Simmons, N. L. & Hirst, B. H. Actin accumulation at sites of attachment of indigenous apathogenic segmented filamentous bacteria to mouse ileal epithelial cells. Infection and immunity 61, 4001-4004 (1993).

280 Geuking, M. B. et al. Intestinal bacterial colonization induces mutualistic regulatory T cell responses. Immunity 34, 794-806, doi:10.1016/j.immuni.2011.03.021 (2011).

168

281 Yang, Y. et al. Focused specificity of intestinal TH17 cells towards commensal bacterial antigens. Nature 510, 152-156, doi:10.1038/nature13279 (2014).

282 Lochner, M. et al. Restricted microbiota and absence of cognate TCR antigen leads to an unbalanced generation of Th17 cells. J Immunol 186, 1531-1537, doi:10.4049/jimmunol.1001723 (2011).

283 Round, J. L. & Mazmanian, S. K. Inducible Foxp3+ regulatory T-cell development by a commensal bacterium of the intestinal microbiota. Proceedings of the National Academy of Sciences of the United States of America 107, 12204-12209, doi:10.1073/pnas.0909122107 (2010).

284 Lathrop, S. K. et al. Peripheral education of the immune system by colonic commensal microbiota. Nature 478, 250-254, doi:10.1038/nature10434 (2011).

285 Cebula, A. et al. Thymus-derived regulatory T cells contribute to tolerance to commensal microbiota. Nature 497, 258-262, doi:10.1038/nature12079 (2013).

286 Heimesaat, M. M. et al. Gram-negative bacteria aggravate murine small intestinal Th1-type immunopathology following oral infection with Toxoplasma gondii. J Immunol 177, 8785-8795 (2006).

287 Cosgrove, D. et al. Mice lacking MHC class II molecules. Cell 66, 1051-1066 (1991).

288 Grusby, M. J., Johnson, R. S., Papaioannou, V. E. & Glimcher, L. H. Depletion of CD4+ T cells in major histocompatibility complex class II-deficient mice. Science 253, 1417-1420 (1991).

289 Cardell, S. et al. CD1-restricted CD4+ T cells in major histocompatibility complex class II-deficient mice. The Journal of experimental medicine 182, 993-1004 (1995).

290 Muranski, P. et al. Tumor-specific Th17-polarized cells eradicate large established melanoma. Blood 112, 362-373, doi:10.1182/blood-2007-11-120998 (2008).

291 Kitamura, D., Roes, J., Kuhn, R. & Rajewsky, K. A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin mu chain gene. Nature 350, 423-426, doi:10.1038/350423a0 (1991).

292 Hashimoto, K., Joshi, S. K. & Koni, P. A. A conditional null allele of the major histocompatibility IA-beta chain gene. Genesis 32, 152-153 (2002).

293 Caton, M. L., Smith-Raska, M. R. & Reizis, B. Notch-RBP-J signaling controls the homeostasis of CD8- dendritic cells in the spleen. The Journal of experimental medicine 204, 1653-1664, doi:10.1084/jem.20062648 (2007).

294 Darrasse-Jeze, G. et al. Feedback control of regulatory T cell homeostasis by dendritic cells in vivo. The Journal of experimental medicine 206, 1853-1862, doi:10.1084/jem.20090746 (2009).

169

295 Lemos, M. P., Fan, L., Lo, D. & Laufer, T. M. CD8alpha+ and CD11b+ dendritic cell-restricted MHC class II controls Th1 CD4+ T cell immunity. Journal of immunology 171, 5077-5084 (2003).

296 Hershberg, R. M. & Mayer, L. F. Antigen processing and presentation by intestinal epithelial cells - polarity and complexity. Immunol Today 21, 123-128 (2000).

297 Madison, B. B. et al. Cis elements of the villin gene control expression in restricted domains of the vertical (crypt) and horizontal (duodenum, cecum) axes of the intestine. The Journal of biological chemistry 277, 33275-33283, doi:10.1074/jbc.M204935200 (2002).

298 Eberl, G. & Littman, D. R. Thymic origin of intestinal alphabeta T cells revealed by fate mapping of RORgammat+ cells. Science 305, 248-251 (2004).

299 Newberry, R. D., McDonough, J. S., McDonald, K. G. & Lorenz, R. G. Postgestational lymphotoxin/lymphotoxin beta receptor interactions are essential for the presence of intestinal B lymphocytes. Journal of immunology 168, 4988-4997 (2002).

300 Barnden, M. J., Allison, J., Heath, W. R. & Carbone, F. R. Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements. Immunology and cell biology 76, 34-40, doi:10.1046/j.1440-1711.1998.00709.x (1998).

301 White, J. et al. Two better cell lines for making hybridomas expressing specific T cell receptors. Journal of immunology 143, 1822-1825 (1989).

302 Pacholczyk, R. et al. Nonself-antigens are the cognate specificities of Foxp3+ regulatory T cells. Immunity 27, 493-504, doi:10.1016/j.immuni.2007.07.019 (2007).

303 Edelson, B. T. et al. Peripheral CD103+ dendritic cells form a unified subset developmentally related to CD8alpha+ conventional dendritic cells. The Journal of experimental medicine 207, 823-836, doi:10.1084/jem.20091627 (2010).

304 Tamoutounour, S. et al. CD64 distinguishes macrophages from dendritic cells in the gut and reveals the Th1-inducing role of mesenteric lymph node macrophages during colitis. European journal of immunology 42, 3150-3166, doi:10.1002/eji.201242847 (2012).

305 Cerovic, V. et al. Intestinal CD103(-) dendritic cells migrate in lymph and prime effector T cells. Mucosal immunology 6, 104-113, doi:10.1038/mi.2012.53 (2013).

306 Scott, C. L. et al. CCR2CD103 intestinal dendritic cells develop from DC-committed precursors and induce interleukin-17 production by T cells. Mucosal immunology, doi:10.1038/mi.2014.70 (2014).

307 Denning, T. L. et al. Functional specializations of intestinal dendritic cell and macrophage subsets that control Th17 and regulatory T cell responses are

170

dependent on the T cell/APC ratio, source of mouse strain, and regional localization. Journal of immunology 187, 733-747, doi:10.4049/jimmunol.1002701 (2011).

308 Welty, N. E. et al. Intestinal lamina propria dendritic cells maintain T cell homeostasis but do not affect commensalism. The Journal of experimental medicine 210, 2011-2024, doi:10.1084/jem.20130728 (2013).

309 Kaplan, D. H., Jenison, M. C., Saeland, S., Shlomchik, W. D. & Shlomchik, M. J. Epidermal langerhans cell-deficient mice develop enhanced contact hypersensitivity. Immunity 23, 611-620, doi:10.1016/j.immuni.2005.10.008 (2005).

310 Cerovic, V. et al. Lymph-borne CD8alpha+ dendritic cells are uniquely able to cross-prime CD8+ T cells with antigen acquired from intestinal epithelial cells. Mucosal immunology 8, 38-48, doi:10.1038/mi.2014.40 (2015).

311 Ginhoux, F. et al. The origin and development of nonlymphoid tissue CD103+ DCs. The Journal of experimental medicine 206, 3115-3130, doi:10.1084/jem.20091756 (2009).

312 Schulz, O. et al. Intestinal CD103+, but not CX3CR1+, antigen sampling cells migrate in lymph and serve classical dendritic cell functions. The Journal of experimental medicine 206, 3101-3114, doi:10.1084/jem.20091925 (2009).

313 Koscso, B., Gowda, K., Schell, T. D. & Bogunovic, M. Purification of dendritic cell and macrophage subsets from the normal mouse small intestine. J Immunol Methods, doi:10.1016/j.jim.2015.02.013 (2015).

314 Bain, C. C. et al. Resident and pro-inflammatory macrophages in the colon represent alternative context-dependent fates of the same Ly6Chi monocyte precursors. Mucosal immunology 6, 498-510, doi:10.1038/mi.2012.89 (2013).

315 Hohl, T. M. et al. Inflammatory monocytes facilitate adaptive CD4 T cell responses during respiratory fungal infection. Cell host & microbe 6, 470-481, doi:10.1016/j.chom.2009.10.007 (2009).

316 Farkas, A. M. et al. Induction of Th17 cells by segmented filamentous bacteria in the murine intestine. J Immunol Methods, doi:10.1016/j.jim.2015.03.020 (2015).

317 Sudo, T. et al. Functional hierarchy of c-kit and c-fms in intramarrow production of CFU-M. Oncogene 11, 2469-2476 (1995).

318 Hashimoto, D. et al. Pretransplant CSF-1 therapy expands recipient macrophages and ameliorates GVHD after allogeneic hematopoietic cell transplantation. The Journal of experimental medicine 208, 1069-1082, doi:10.1084/jem.20101709 (2011).

319 Uematsu, S. & Akira, S. Immune responses of TLR5(+) lamina propria dendritic cells in enterobacterial infection. Journal of gastroenterology 44, 803-811, doi:10.1007/s00535-009-0094-y (2009).

171

320 Kinnebrew, M. A. et al. Bacterial flagellin stimulates Toll-like receptor 5-dependent defense against vancomycin-resistant Enterococcus infection. The Journal of infectious diseases 201, 534-543, doi:10.1086/650203 (2010).

321 Zigmond, E. et al. Ly6C hi monocytes in the inflamed colon give rise to proinflammatory effector cells and migratory antigen-presenting cells. Immunity 37, 1076-1090, doi:10.1016/j.immuni.2012.08.026 (2012).

322 Tacke, F. et al. Immature monocytes acquire antigens from other cells in the bone marrow and present them to T cells after maturing in the periphery. The Journal of experimental medicine 203, 583-597, doi:10.1084/jem.20052119 (2006).

323 Farkas, A. M. et al. Induction of Th17 cells by segmented filamentous bacteria in the murine intestine. Journal of immunological methods 421, 104-111, doi:10.1016/j.jim.2015.03.020 (2015).

324 Qu, C., Brinck-Jensen, N. S., Zang, M. & Chen, K. Monocyte-derived dendritic cells: targets as potent antigen-presenting cells for the design of vaccines against infectious diseases. International journal of infectious diseases : IJID : official publication of the International Society for Infectious Diseases 19, 1-5, doi:10.1016/j.ijid.2013.09.023 (2014).

325 Kambayashi, T. & Laufer, T. M. Atypical MHC class II-expressing antigen-presenting cells: can anything replace a dendritic cell? Nature reviews. Immunology 14, 719-730, doi:10.1038/nri3754 (2014).

326 Pollard, J. W. Trophic macrophages in development and disease. Nature reviews. Immunology 9, 259-270, doi:10.1038/nri2528 (2009).

327 Shi, Y. et al. Granulocyte-macrophage colony-stimulating factor (GM-CSF) and T-cell responses: what we do and don't know. Cell research 16, 126-133, doi:10.1038/sj.cr.7310017 (2006).

328 Martins, A., Han, J. & Kim, S. O. The multifaceted effects of granulocyte colony-stimulating factor in immunomodulation and potential roles in intestinal immune homeostasis. IUBMB life 62, 611-617, doi:10.1002/iub.361 (2010).

329 Arango Duque, G. & Descoteaux, A. Macrophage cytokines: involvement in immunity and infectious diseases. Frontiers in immunology 5, 491, doi:10.3389/fimmu.2014.00491 (2014).

330 Hoffman, J., Kuhnert, F., Davis, C. R. & Kuo, C. J. Wnts as essential growth factors for the adult small intestine and colon. Cell Cycle 3, 554-557 (2004).

331 Swafford, D. & Manicassamy, S. Wnt signaling in dendritic cells: its role in regulation of immunity and tolerance. Discovery medicine 19, 303-310 (2015).

332 Suryawanshi, A. et al. Canonical wnt signaling in dendritic cells regulates Th1/Th17 responses and suppresses autoimmune neuroinflammation. J Immunol 194, 3295-3304, doi:10.4049/jimmunol.1402691 (2015).

172

333 Worzfeld, T. & Offermanns, S. Semaphorins and plexins as therapeutic targets. Nature reviews. Drug discovery 13, 603-621, doi:10.1038/nrd4337 (2014).