Kenneth Martin Rosenberg Email: [email protected], [email protected] 660 West Redwood Street, Howard Hall Room 332D, Baltimore, MD, 21201

Kenneth Martin Rosenberg Email: Kenneth.Rosenberg@Som.Umaryland.Edu, Kenny.M.Rosenberg@Gmail.Com 660 West Redwood Street, Howard Hall Room 332D, Baltimore, MD, 21201

The impact of the non-immune chemiome on T cell activation

Item Type dissertation

Authors Rosenberg, Kenneth

Publication Date 2020

Abstract T cells are critical organizers of the immune response and rigid control over their activation is necessary for balancing host defense and immunopathology. It takes 3 signals provided by dendritic cells (DC) to fully activate a T cell response – T ce...

Keywords signaling; T cell; T-Lymphocytes--immunology

Download date 02/10/2021 13:41:58

Link to Item http://hdl.handle.net/10713/14477 Kenneth Martin Rosenberg Email: [email protected], [email protected] 660 West Redwood Street, Howard Hall Room 332D, Baltimore, MD, 21201

EDUCATION MD, University of Maryland, Baltimore, MD Expected May 2022 PhD, University of Maryland, Baltimore, MD December 2020 Graduate Program: Molecular Microbiology and Immunology (MMI) BS, University of Maryland, College Park, MD May 2013 Major: Bioengineering, cum laude University Honors Citation, Gemstone Citation

RESEARCH EXPERIENCE UMSOM Microbiology and Immunology Baltimore, MD July 2016-present PhD Candidate Principal Investigator: Dr. Nevil Singh Thesis: The impact of the non-immune chemiome on T cell activation Examined environmental stimuli from classically “non-immune” sources – growth factors, hormones, neurotransmitters, etc. – act to modulate T cell signaling pathways and the functional effects of activating encounters with dendritic cells. UMSOM Anatomy and Neurobiology Baltimore, MD May-August 2015 Rotating student Principal Investigator: Dr. Asaf Keller Studied the role of descending modulation pathways on affective pain transmission. Performed tract- tracing experiments using targeted injection of Cholera toxin subunit B into the lateral parabrachial nucleus and ventrolateral periaqueductal gray of anesthetized transgenic mice. FDA Center for Devices and Radiological Health Laurel, MD July 2013-June 2014 ORISE Research Fellow in the Laboratory of Cardiovascular and Interventional Therapeutics Principal Investigators: Dr. Bill Pritchard and Dr. John Karanian Studied the factors that contribute to endovascular implant failure. Analyzed CT angiograms using image-based geometric modeling techniques to characterize the in vivo loading environment in major blood vessels and cardiac structures. Assisting in imaging of animal subjects and other various surgical procedures. UMD Clark School of Engineering College Park, MD May 2012-May 2013 Orthopedic Mechanobiology Lab Faculty Mentor: Dr. Adam Hsieh, Department of Bioengineering Studied the role of the pericellular matrix in the mechanotransduction processes of human mesenchymal stem cells undergoing chondrogenesis. Analysis included real-time RT-PCR and immunofluorescence imaging techniques. Received an HHMI Undergraduate Fellowship to fund this research beginning spring 2013. Senior Capstone Project College Park, MD Fall 2012-Spring 2013 Physician Mentor: Dr. Ashutosh Sachdeva, University of Maryland Medical Center Faculty Mentor: Dr. Keith Herold, Department of Bioengineering Worked with four students to develop a sensor system to measure the reocclusion of silicone tracheal stents. The sensor was designed to detect the changing capacitance across the stent caused by the

buildup of mucus or granulation tissue. It then wirelessly transmits a digital signal to an Android smart phone where the user could be alerted to the measurement results. Gemstone-Team RODENT College Park, MD Fall 2010-Spring 2013 Faculty Mentor: Dr. Brian Bequette, Department of Animal and Avian Sciences Collaborated with eleven students and a faculty mentor to perform research studying the underlying metabolic dysregulation associated with obesity using a mouse model. Was responsible for care of the mice as well as several lab procedures including dissections and sample processing. Received research grants from HHMI and ACCIAC. NIH National Institute on Aging Baltimore, MD Summer 2011 Summer Intern in the MRI and Spectroscopy Section of the Laboratory of Clinical Investigation Principal Investigator: Dr. Richard Spencer Studied the efficacy of diagnosing cartilage degradation using multiple MRI parameter analysis.

TEACHING EXPERIENCE UMB, Dept. of Microbiology and Immunology Baltimore, MD Summer 2020 Lecturer – Overview of Immunology Summer Course Gave lectures regarding T cell antigen recognition and transplant immunology as part of a three- week virtual immunology course for summer internship students as well as incoming and current graduate students. University of Maryland SOM Baltimore, MD Fall 2018, Fall 2019 Small group leader – Host Defense and Infectious Diseases Led second year medical student discussion sections covering CAR-T cell therapy, virology basics, HPV vaccination, and Zika virus pathogenesis. Served as a leader for 4 sessions in 2018 and 3 in 2019. UMB Molecular Microbiology and Immunology Baltimore, MD Fall 2018, Fall 2019 Lecturer and discussion leader – Advances in Immunology Gave the Neuroimmunology overview lecture, including a history of the field and summary of recent literature highlighting the breadth of the field. Additionally, worked with a student in the course to select and present a paper discussion covering 2-3 recent journal articles. Was invited back to serve in the same role in fall of 2019. University of Maryland CMNS College Park, MD Fall 2011, Spring 2013 Undergraduate Teaching Assistant - Introduction to Genetics Fall 2011 - Led a class of 20 students in weekly discussion and held weekly office hours. Aided in the proctoring and grading of exams as well as graded other class assignments. Spring 2013 - Transitioned to a different role in which I managed the online resources available to the students and wrote weekly quizzes for students in all sections. Continued to maintain weekly office hours and aid in the proctoring and grading of exams.

UNIVERSITY SERVICE UMB Molecular Microbiology and Immunology Admissions Committee Fall 2019-Spring 2020 Student representative and interviewer University of Maryland MSTP Admissions and Advisory Committee Fall 2016-Spring 2018 Student representative and interviewer University of Maryland School of Medicine Admissions Fall 2015-Spring 2016

Student interviewer University of Maryland School of Medicine Fall 2014-Spring 2016 Class of 2018 Class Council Treasurer

PUBLICATIONS Rosenberg, KM, Singh, NJ. (2019) “Mouse T cells express a neurotransmitter-receptor signature that is quantitatively modulated in a subset- and activation-dependent manner.” Brain, Behavior, and Immunity. 80:275-285

Rosenberg, KM, Singh, NJ. VIP inhibits TCR-mediated ERK signaling to drive altered CD4 activation and differentiation. [Manuscript in prep]

Fioretti S, Matson CA, Rosenberg KM, Singh NJ. Dynamic downregulation of CD19 by B cells underlies the escape phenotype after anti-CD19 directed immunotherapy. [Manuscript in prep]

ABSTRACTS & PRESENTATIONS Oral Presentations Rosenberg, KM, Singh, NJ. “Neuronal regulation of immunity via T cell subset-specific neurotransmitter receptors.” Graduate Student Symposium, June 2019, UMB, Baltimore, MD.

Rosenberg, KM, Singh, NJ. “Differential expression of neurotransmitter receptors by T cell subsets modifies their antigen-specific activation.” Graduate Student Symposium, June 2019, UMB, Baltimore, MD.

Rosenberg, KM, Singh, NJ. “Subset-specific neurotransmitter receptor expression tunes T cell activation.” American Association of Immunologists: Immunology 2018. May 2018; Austin, TX.

Rosenberg, KM, Singh, NJ. “A potential codex for neuronal regulation of peripheral T cells.” Graduate Student Symposium, June 2019, UMB, Baltimore, MD.

Rosenberg, KM, Singh, NJ. “Neurotransmitters in the regulation of T cell function.” University of Maryland MSTP Summer Research Symposium. August 2016; Baltimore, MD.

Rosenberg, KM, Keller, A. “Descending modulation of affective pain pathways.” University of Maryland MSTP Summer Research Symposium. August 2015; Baltimore, MD.

Posters Rosenberg, KM, Singh, NJ. “Expression of T cell-subset-specific signatures of neurotransmitter receptors allows the nervous system to fine tune immune activation.” MD/PhD National Student Conference. July, 2019; Copper Mountain, CO.

Rosenberg, KM, Singh, NJ. “Subset-specific neurotransmitter receptor expression tunes T cell activation.” American Association of Immunologists: Immunology 2018. May 2018; Austin, TX.

Rosenberg, KM, Singh, NJ. “Families of neurotransmitter receptors are differentially regulated across T cell subsets.” American Association of Immunologists: Immunology 2017. May 2017; Washington, DC.

Rosenberg, KM, Gregg, K, Leonard, S, Ries, S, Karanian, JW, Pritchard, WF. “Geometry and Motion of the Aortic Arch Across Species: Implications for Stent-Graft Device Evaluation and Development.” FDA CDRH Summer Student Poster Symposium. August 2013; White Oak, MD.

Rosenberg, KM, Twomey, JD, Hsieh, AH. “Role of the pericellular matrix in the mechanotransduction pathways of hMSCs undergoing chondrogenesis.” HHMI Research Fellowship Program 14th Annual Research Symposium, University of Maryland. April 2013; College Park, MD.

GRANTS University of Maryland MSTP T32 trainee Fall 2015-Spring 2016 Maryland-HHMI Undergraduate Research Fellowship Spring 2013

HONORS & AWARDS AAI Immunology 2018 – Trainee Abstract Award May 2018 University of Maryland Banneker/Key Scholar Fall 2009-Spring 2013

Abstract

Title of Dissertation:

The impact of the non-immune chemiome on T cell activation

Kenneth M. Rosenberg, Doctor of Philosophy, 2020

Dissertation Directed by:

Nevil Singh, PhD, Assistant Professor Department of Microbiology and Immunology University of Maryland School of Medicine

T cells are critical organizers of the immune response and rigid control over their activation is necessary for balancing host defense and immunopathology. It takes 3 signals provided by dendritic cells (DC) to fully activate a T cell response – T cell receptor (TCR) engagement of antigen on MHC (Signal 1), co-stimulatory signals (Signal 2) and cytokines

(Signal 3). Yet, even before activation T cells are typically exposed to a universe of chemicals (a “chemiome”) including drugs, metabolites, hormones etc. which are not typically ascribed an immunological role. In this thesis, we hypothesized that members of this non-immune chemiome acting on T cells, prior to antigen encounter, flavor specific signaling pathways to differentially influence subsequent T cell activation and fate.

Unraveling these signals, which we termed “Signal 0”, could help us understand and manipulate tissue and time specific flavoring of immunity. In this thesis we first developed a pharmacological model for signal 0, by treating T cells with drugs that activate only subsets of the TCR-signaling network prior to full antigen exposure. We found that pharmacological pre-activation of the PKCƟ/ERK pathways modulates long time survival of T cells without changing proliferation or cytokine production. Next, we examined receptors for the non-immune chemiome that resting T cells express and identified

neurotransmitter receptors (NR) as a major family. All T cells expressed a core NR signature, but very few NR were also modulated in a T cell lineage-specific fashion. Of these, we focused on VPAC1, the receptor for vasoactive intestinal peptide (VIP). We found that VIP signaling attenuates ERK phosphorylation, but paradoxically drives increased differentiation towards IL-17 and IL-22 secretion. In addition ERK signaling induced by drugs (phorbol esters) versus the TCR followed differential kinetics and recruited non-overlapping negative feedback mechanisms, suggesting that even the same branch of TCR signaling is subject to different localization and temporal controls. Taken together, our data suggest that the branches of the TCR-signaling network integrate pre- existing signals (Signal 0) into the activation program of T cells, allowing localized cues, including neurotransmitter levels, to modify the long-term trajectory of the immune response.

The Impact of the Non-immune Chemiome on T cell Activation

by Kenneth M. Rosenberg

Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, Baltimore in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2020

Acknowledgements

To my mentor, Dr. Nevil Singh, thank you for so many things but especially your patience and support over the past four years. Thank you for making science fun, both inside and outside of the lab. There is not another person I would rather have learned how to do science with than you.

To Dr. Courtney Matson and (soon to be Dr.) Allison Gerber, thank you for being my teammates, those I could always lean on, though out this process. Simply put, I would not have made it through graduate school without you two.

To Dr. John Reiser and Gideon Wolf, my lab brothers, thank you for always being available to talk about science and about life. You guys made sure that coming to lab each was fun.

Thank you Zach Fasana, although you joined the lab late in my training, you were incredibly helpful in preparing this dissertation, providing excellent feedback.

Thank you to my thesis committee, Drs. Sergei Atamas, Martin Flajnik, Achsah Keegan, and John O’Shea, for your insightful critiques of both my project over the years and this dissertation as well as your enthusiastic support. I want to offer an extra thank you to Dr. Flajnik who regularly stopped by lab to check in and hand me a brand new paper I needed to see but also to talk about the latest sports news.

To Drs. Bret Hassel and Heather Ezelle, as well as the rest of the MMI community, thank you for your feedback on my project at department seminars, for keeping me on track though out the graduate program, and for just being great people to spend the last four years of my life with.

To the MSTP administration, Dr. Mary-Claire Roghmann, Dr. Achsah Keegan, and especially Jane Bacon, for helping me navigate between multiple schools and training programs as well as helping me to keep a clinical mindset during these years.

Lastly, to my family and friends, both old and new, thank you for being an unwavering support structure, through successes and setbacks.

Thank you all.

iii

Table of Contents

List of Tables ...... viii

List of Figures ...... ix

List of Abbreviations ...... xiv

Chapter 1: Introduction ...... 1 1.1. T cells are critical organizers of the adaptive immune response ...... 1 1.2. The road to the 3 signal model: the current paradigm for T cell activation and differentiation...... 7 1.2.1. The two-signal model ...... 7 1.2.2. A third signal emerges...... 13 1.2.3. The third signal is biochemically distinct from Signals 1 and 2 ...... 17 1.2.4. A more significant role for the TCR in differentiation and cell survival ...... 23 1.2.5. The components of TCR signaling ...... 24 1.2.6. Co-stimulatory receptor signaling ...... 30 1.3. Timing and order of signals matter...... 31 1.4. Other signals can affect TCR signaling ...... 33 1.5. The new concept: signal 0 ...... 35 1.6. Thesis hypothesis and Specific aims ...... 39

Chapter 2: Materials and methods ...... 40

Chapter 3: Pre-stimulating specific subsets of the TCR signaling pathway alters T cell activation ...... 63 3.1. Introduction ...... 63 3.1.1. Approach to studying testing the signal 0 hypothesis ...... 63 3.1.2. Chapter summary ...... 66 3.2. Results ...... 66 3.2.1. Transient exposure to PMA modulates subsequent T cell activation...... 66 3.2.2. PMA induces unique ERK phosphorylation kinetics...... 76 3.2.3. PMA and direct TCR-complex signaling recruit different kinds of negative feedback ...... 80 3.2.4. Mechanism for differential ERK activation is not clear ...... 89 3.2.5. PMA exposure may decrease T cell survival in vivo ...... 92

3.2.6. Ceramide (C6) does not induce ERK phosphorylation in our hands ...... 97 3.2.7. Early TCR signaling is modulated by prior antigen encounter ...... 102

3.2.8. Optimizing a comprehensive panel for further evaluation of signaling interactions ...... 111 3.3. Significant findings & Discussion ...... 115

Chapter 4: Receptors for the non-immune chemiome on T cells ...... 118 4.1. Introduction ...... 118 4.1.1. Defining a strategy for identifying relevant receptors for the Non-immune chemiome (rNIC) ...... 119 4.2. Results ...... 120 4.2.1. T cells express receptors for a broad array of environmental ligands...... 120 4.2.2. EGFR may not participate in signal 0 ...... 125 4.2.3. Insulin treatment does not modulate T cell proliferation ...... 129 4.3. Significant findings & Discussion ...... 133

Chapter 5: Neurotransmitter receptors constitute a major source of rNIC relevant to T cells ...... 135 5.1. Introduction ...... 135 5.1.1. Neuronal regulation of immunity ...... 135 5.1.2. NR classes and signaling modalities ...... 137 5.1.3. Proposed functional impacts of NR signaling on T cell biology ...... 142 5.1.4. Limitations of the prior literature and the necessity of its reexamination ...... 142 5.2. Results ...... 144 5.2.1. Peripheral T cells express a limited subset of Neurotransmitter receptors (NR) ...... 144 5.2.2. Validation of NR expression in FACS-sorted T cells...... 155 5.2.3. T cell expression of NR is dynamic with many showing rapid downregulation after activation...... 166 5.2.4. Tissue residency has minimal effect on the NR signature of activated T cells173 5.2.5. T cell NR expression patterns confirmed by RNA-Seq analysis ...... 177 5.2.6. Similar NR expression trends are found in human T cell populations...... 180 5.2.7. A working map of NR expression by T cells ...... 183 5.2.8. 2-adrenergic receptor signaling dampens T cell activation ...... 185 5.2.9. Transient agonism of other core signature NR do not inhibit T cell activation ...... 191 5.2.10. NR agonism does not induce ERK phosphorylation ...... 195 5.3 Significant findings & Discussion ...... 198

Chapter 6: VIP as a model NIC for subset specific modulation of T cell biology....201 6.1. Introduction ...... 201 6.1.1. VPAC1 or the VIP receptor, is known to have immunomodulatory roles...... 201 6.2. Results ...... 203 6.2.1. T cell activation is inhibited by transient VIP exposure ...... 203 6.2.2. VIP signaling inhibits phosphorylation of ERK in T cells ...... 214

6.2.3. VIP enhances TH22 and TH17 differentiation ...... 218

6.2.4. VPAC1 is most highly expressed in naïve CD4+ T cells ...... 224

6.2.5. Overexpression of VPAC1 alters T cell effector function and survival ...... 226

6.2.6. Overexpression of VPAC1 in bone marrow may promote peripheral Treg development...... 233 6.3. Significant findings & Discussion ...... 237

Chapter 7: Conclusions and future directions ...... 239 7.1. Pre-stimulating specific subsets of the TCR signaling pathway alters T cell activation ...... 239 7.1.1. Significant findings ...... 239 7.1.2. Future directions ...... 240 7.2. Receptors for the non-immune chemiome on T cells ...... 240 7.2.1. Significant findings ...... 240 7.2.2. Future directions ...... 240 7.3. Neurotransmitter receptors constitute a major source of rNIC relevant to T cells ...... 241 7.3.1. Significant findings ...... 241 7.3.2. Future directions ...... 241 7.4. VIP as a model NIC for subset specific modulation of T cell biology ...... 242 7.4.1. Significant findings ...... 242 7.4.2. Future directions ...... 242 7.5. Implications of the NIC as signal 0 ...... 243 7.5.1. Pre-conditioning of T cells by non-immune encounters likely contributes to cell-to-cell response heterogeneity...... 243 7.5.2 Our mechanistic studies offer a rational strategy to anticipate (and perhaps treat) variation in human immune responses ...... 244 7.5.3. The insights gained from this work can lead to new pharmacological approaches ...... 244

References ...... 246

vii

List of Tables

Table 2.1. Mouse model genetic designation and abbreviations...... 42

Table 2.2. Antibodies used for magnetic T cell selection...... 45

Table 2.3. Antibodies used for flow cytometric analysis...... 49

Table 2.4. Antibodies used for TCR-crosslinking in phospho-flow cytometry studies. ... 53

Table 2.5. Antibodies used for FACS-purification...... 53

Table 2.6. Primary antibodies used for western blot analysis...... 59

Table 2.7. Secondary antibodies used for western blot analysis...... 59

Table 4.1. T cells express receptors for NIC molecules...... 124

Table 5.1. Overview of neurotransmitters and their receptors...... 140

Table 5.2. NR expression by T cells using restrictive thresholding (100) of ImmGen data

...... 149

Table 5.3. NR expression by T cells using inclusive thresholding (40) of ImmGen data.

...... 150

Table 5.4. Statistical assessment of NR expression in T cells of ImmGen data...... 153

Table 5.5. Statistical comparison of NR expression between T cell subsets of ImmGen data...... 154

Table 5.6. Complete NanoString gene list and probe details...... 161

Table 5.7. Statistical analysis of NR expression dynamics within ImmGen data...... 170

viii

List of Figures

Figure 1.1. Overview of the physical and cellular barriers to infection...... 2

Figure 1.2. Overview of the receptor classes utilized by cells of the innate and adaptive immune systems...... 4

Figure 1.3. Time course of the immune response...... 5

Figure 1.4. Summary of CD4 and CD8 T cell function...... 6

Figure 1.5. Evolution of the 2-signal model...... 12

Figure 1.6. Summary of CD4 T cell differentiation...... 15

Figure 1.7. Common -chain cytokines control T cell proliferation, survival, and lineage commitment...... 16

Figure 1.8. The 3-signal model of T cell activation...... 17

Figure 1.9. Summary of JAK-STAT signaling by cytokine receptors...... 21

Figure 1.10. Schematic of TGFR signaling...... 22

Figure 1.11. Schematic of T cell receptor signaling pathways...... 28

Figure 1.12. Cooperation of several TF are required for the full activation of IL-2 transcription...... 29

Figure 1.13. Schematic of Signal 0 hypothesis...... 38

Figure 3.1. Transient PMA exposure enhances early antibody-mediated T cell activation.

...... 68

Figure 3.2. Transient PMA exposure enhances early antigen-mediated T cell activation.

...... 70

Figure 3.3. Effects of transient PMA exposure on early antigen-mediated T cell activation are replicated...... 71

Figure 3.4. Transient PMA exposure enhances T cell proliferation and IL-2 production. 73

Figure 3.5. Effects of transient PMA exposure on T cell proliferation are replicated...... 75

Figure 3.6. PMA activates ERK signaling in 5C.C7 cells...... 77

Figure 3.7. PMA-induced ERK phosphorylation follows distinct kinetics from TCR- mediated pERK signaling...... 79

Figure 3.8. PMA-induced ERK kinetics dominate TCR-mediated kinetics...... 81

Figure 3.9. Polyclonal T cells also show distinct kinetics between PMA-induced and

TCR-mediated pERK...... 83

Figure 3.10. CD3 is not downregulated by anti-CD3/CD4 stimulation...... 85

Figure 3.11. Transient PMA exposure augments subsequent TCR-mediated ERK phosphorylation kinetics...... 88

Figure 3.12. PMA- and TCR-induced ERK phosphorylation may show differential localization...... 91

Figure 3.13. Transient PMA exposure alters T cell survival in vivo...... 93

Figure 3.14. Transient PMA exposure does not alter in vivo T cell proliferation of differentiation...... 94

Figure 3.15. Lack of effect of PMA treatment of T cell proliferation and differentiation is repeated...... 96

Figure 3.16. C6 exposure does not alter CD68 expression in B6 T cells...... 98

Figure 3.17. C6 does not induce ERK phosphorylation...... 100

Figure 3.18. Acutely and chronically activated T cells show distinct TCR-mediated signaling patterns from naïve T cells...... 104

Figure 3.19. Distinct TCR-mediated signaling patterns of acutely and chronically activated T cells apparent in T cell replete hosts...... 108

Figure 3.20. Detailed TCR-mediated signaling kinetics can be observed using spectral flow cytometry...... 113

Figure 4.1. EGFR expression by T cells is not detected...... 126

Figure 4.2. EGF does not induce ERK phosphorylation in T cells...... 128

Figure 4.3. Insulin has no effect on 5C.C7 T cell proliferation...... 131

Figure 4.4. Insulin does not alter proliferation mediated by altered peptide ligands...... 132

Figure 5.1. Summary of NR signaling modalities...... 141

Figure 5.2. T cells express a limited diversity of neurotransmitter receptors in a subset- specific fashion...... 147

Figure 5.3. FACS gating strategy for isolating T cell subsets...... 157

Figure 5.4. Post-sort purities of isolated T cell populations...... 158

Figure 5.5. Quantitative comparison of NR expression by flow-sorted primary T cell subsets...... 159

Figure 5.6. NR expression differences between T cell subsets are verified by qPCR. ... 160

Figure 5.7. T cells NR at the protein level...... 165

Figure 5.8. Naïve T cells modulate NR expression following activation...... 169

Figure 5.9. Effector T cells show virtually identical NR profiles...... 172

Figure 5.10. T cell NR expression is largely intrinsic, rather than dominantly modulated by tissue residence...... 176

Figure 5.11. NR core signature is validated by RNA-Seq analysis...... 179

Figure 5.12. NR core signature is validated by RNA-Seq analysis...... 182

Figure 5.13. Maps of neurotransmitter receptor expression by T cells...... 184

Figure 5.14. NR agonism analysis gating strategy...... 186

Figure 5.15. 2 adrenergic receptor signaling inhibits early T cell activation...... 188

Figure 5.16. 2AR-mediated inhibition of T cell activation is likely mediated by cAMP.

...... 190

Figure 5.17. Histamine H2R signaling may limit T cell activation...... 192

Figure 5.18. Nicotinic acetylcholine receptor signaling does not alter T cell activation.194

Figure 5.19. ERK phosphorylation is not modified by NR agonism alone...... 196

Figure 6.1. VIP inhibits early T cell activation in a polyclonal population...... 204

Figure 6.2. VIP decreases T cell proliferation...... 206

Figure 6.3. VIP inhibits early antibody-mediated activation of 5C.C7 T cells...... 208

Figure 6.4. The effect of VIP on early antibody-mediated activation is replicated in

SMARTA T cells...... 209

Figure 6.5. VIP inhibits early peptide-mediated activation of 5C.C7 and SMARTA T cells...... 211

Figure 6.6. Proliferation of 5C.C7 T cells is unaffected by VIP...... 213

Figure 6.7. VIP inhibits ERK induction in T cells...... 215

Figure 6.8. Quantitation of TCR signaling following transient exposure to VIP...... 217

Figure 6.9. VIP enhances IL-22 production by T cells...... 219

Figure 6.10. VIP enhances IL-17 production by T cells...... 220

Figure 6.11. IFN production largely unaffected by VIP...... 221

Figure 6.12. IL-2 dampens IL-17 and IL-22 production...... 223

Figure 6.13. Vipr1 is downregulated after activation of naïve T cells...... 225

xii

Figure 6.14. Vipr1 overexpression plasmid diagram...... 227

Figure 6.15. Retroviral expression of Vipr1...... 228

Figure 6.16. Overexpression of Vipr1 increases IFN production by effector cells...... 230

Figure 6.17. Overexpression of Vipr1 may increase T cell survival...... 232

Figure 6.18. Vipr1 BM chimeras mice show increased peripheral CD4+CD25+ T cell frequency...... 235

xiii

List of Abbreviations

7AAD 7-aminoactinomycin D

ACh Acetylcholine

AChR Acetylcholine receptor

AP-1 Activator protein 1

APC Antigen presenting cell

AREG Amphiregulin

2AR 2-adrenergic receptor

BDNF Brain-derived neurotrophic factor

BM Bone marrow

BMM Bone marrow media

BMP Bone morphogenic protein cAMP cyclic adenosine monophosphate

CFSE Carboxyfluorescein succinimidyl ester

CGRP Calcitonin gene-related peptide

CTL Cytotoxic T lymphocyte

CTV Cell trace violet

DAG Diacylglycerol

DAMP Danger associated molecular patterns

DAPI 4′,6-diamidino-2-phenylindole

DC Dendritic cell

DNA Deoxyribonucleic acid

xiv

EGF Epidermal growth factor

EGFR Epidermal growth factor receptor

ELISA Enzyme-linked immunosorbent assay

Eph Ephrin receptor

EPO Erythropoietin

ERGIC endoplasmic reticulum-Golgi intermediate compartment

ERK Extracellular signal-regulated kinase

FACS Fluorescence assisted cell sorting

FcR Fc receptor

FCS Fetal calf serum

FGF Fibroblast growth factor

GABA -aminobutyric acid

Gads GRB2-related adapter downstream of Shc

GDP Guanosine diphosphate gMFI Geometric mean fluorescence intensity

GPCR G-protein-coupled receptor

GTP Guanosine triphosphate

HGF Hepatocyte growth factor

ICOS Inducible T cell costimulator

IFN Interferon

IFNγ Interferon γ

IGF Insulin-like growth factor

IKK Inhibitor of NFκB kinase

IL-1 Interleukin-1

IL-10 Interleukin-10

IL-12 Interleukin 12

IL-13 Interleukin-13

IL-15 Interlekin-15

IL-17 Interleukin-17

IL-2 Interleukin-2

IL-21 Interleukin-21

IL-22 Interleukin-22

IL-23 Interleukin-23

IL-3 Interleukin-3

IL-4 Interleukin-4

IL-5 Interleukin-5

IL-6 Interleukin-6

IL-7 Interleukin-7

ILC Innate lymphoid cell

INSR Insulin receptor i.p. Intraperitoneal

IP3 Inositol 1,4,5-triphosphate

IP3R Inositol 1,4,5-triphosphate receptor

ITAM Immunoreceptor tyrosine-based activation motif

ITK Interleukin-2-incudible T cell kinase

IκB Inhibitor of NFκB

xvi i.v. Intravenous

JAK Janus kinase

JNK c-Jun N-terminal kinase

KO Knock-out

LAG-3 Lymphocyte-activation gene 3

LAT Linker for activation

Lck Lymphocyte specific protein tyrosine kinase

LCMV Lymphocytic choriomeningitis virus

LN Lymph node

LPS Lipopolysaccharide

MAPK Mitogen activated protein kinase

MCC Moth cytochrome C

MFI Mean fluorescence intensity

MHC Major histocompatibility complex

MPEC Memory precursor effector cell

NFAT Nuclear factor of activated T cells

NFκB Nuclear factor-κB

NGF Nerve growth factor

NIC Non-immune chemiome

NK Natural killer

NMU Neuromedin U

NR Neurotransmitter receptor

NRP1 Neuropilin-1

xvii

NT neurotransmitter

NT-3 Neurotrophin-3 nRTK non-receptor tyrosine kinase

PAMP Pathogen associated molecular pattern

PBS Phosphate-buffered saline

PCC Pigeon cytochrome C pERK phosphorylated-ERK

PI3K Phosphoinositide 3-kinase

PIP2 Phosphatidylinositol 4,5-bisphosphate

PIP3 Phosphatidylinositol 3,4,5-trisphosphate

PKC Protein kinase C

PLCγ Phospholipase Cγ

PMA Phorbol 12-myristate 13-acetate pMHC peptide-MHC pQ-RV pQ2aB retrovirus pre-Tx Pre-treatment

PRR Pattern recognition receptors pSLP-76 phosphorylated-SLP-76 pV-RV Retrovirus encoding Vipr1 cloned into pQ2aB pZap70 phosphorylated-Zap70

RasGRP Ras guanyl nucleotide releasing protein

RNA Ribonucleic acid rNIC Receptor for a Non-immune chemiome molecule

xviii

RORt RAR-related orphan receptor 

RSTK Receptor serine/threonine kinase

RTK Receptor tyrosine kinase

SCF Stem cell factor

Sema3 Class 3 semaphorin

SH2 Src homology 2

SLEC Short-lived effector cell

SLO Secondary lymphoid organ

SLP-76 SH2 domain-containing leukocyte protein of 76 kDa

STAT Signal transducer and activator of transcription

TAAR Trace amine-associated receptor

T-bet T-box expressed in T cells

TCM T cell media

TCR T cell receptor

TCR-Tg T cell receptor transgenic

TF Transcription factor

TFH T follicular helper cell

TGF Transforming growth factor

TH T helper

TH1 T helper type 1

TH2 T helper type 2

TH17 T helper type 3

TIM-3 T cell immunoglobulin and mucin-domain containing protein 3

xix

TLR Toll-like receptor

TNF Tumor necrosis factor

Treg T regulatory cell

VEGF Vascular endothelial growth factor

VIP Vasoactive intestinal peptide

WT Wild-type

Zap70 Zeta chain associated protein kinase 70

Chapter 1: Introduction

1.1. T cells are critical organizers of the adaptive immune response

The immune system evolved to protect the body from a variety of threats – infectious organisms (including bacteria, viruses, parasites, and fungi), tumors as well as environmental toxins. To defend the body, this system, comprised of several specialized cell types and the molecules they release, must detect these threats, stop their replication and/or spread, minimize damage to additional host tissues, and ultimately eliminate or contain the threat. Due to the incredible diversity of potential threats to the integrity of the body, the cells and molecules of the immune system have developed a broad array of mechanisms and a complex division of labor to accomplish each of these tasks, as there is no “one-size-fits-all” strategy (Fig. 1.1).

Once a pathogen has breached one of the passive barriers that physically protect the body from invasion, including the skin, lungs, gut, and other mucosal surfaces (Fig. 1.1), the cells that it initially encounters (including macrophages, neutrophils, eosinophils, mast cells, etc.) are part of the innate immune system. Responding rapidly, within minutes to hours of invasion, these actors identify threatening, foreign organisms using genetically hard-coded (or “innate”) receptors. These pattern recognition receptors (PRR)1 bind conserved structures unique to pathogens (e.g. bacterial cell membranes, viral genomes, etc.), and thus “foreign” to the host, termed pathogen-associated molecular patterns

(PAMP). Additionally, PRR also recognize threats by binding molecules indicative of tissue injury, termed damage-associated molecular patterns (DAMP). Once an innate immune cell has been activated by binding PAMPs and/or DAMPs, it tries to eliminate the threat by a variety of mechanisms. These include engulfing and killing the pathogen within

the cell via a process called phagocytosis, killing the pathogen in the extracellular space by secreting antimicrobial and degradative proteins, as well as releasing chemokines which act as beacons to recruit additional immune cells to the site of infection/damage. The rapid, nonspecific response of the innate system acts as the first line of defense, attempting to stave off the spread of infection while the second line of defense develops: the adaptive immune response.

A B

Figure 1.1. Overview of the physical and cellular barriers to infection. (A) There are many organs spread throughout the body that act as key sites for the development of the adaptive immune response. The thymus and bone marrow are known as primary lymphoid organs, as they are the sites of T cell and B cell development, respectively. The remaining structures, namely the spleen, lymph nodes, tonsils, Peyer’s patches, and the appendix are termed secondary lymphoid organs (SLO). After development, T and B cells circulate between these structures, moving through lymphatic vessels, and the blood stream, awaiting activation by encounter with their respective antigen. Reprinted with permission from: https://microdok.com/organs-of-the-immune-system-and-their-function/ (B) Several barriers act to stop infection by potential pathogens. The first includes physical and barriers that prevent entry into the body, sometimes aggregating and expelling pathogens in the case of mucus or presenting a hostile environment for pathogen survival in the case of stomach acid. Additionally, the skin and mucous membranes themselves, in which cells layers are sealed by tight junctions, serve as walls that provide among the greatest barriers to infection. Reprinted with permission from the Open University: https://www.open.edu/openlearn/ocw/mod/oucontent/view.php?id=28153§ion=4.1#back_longdesc_id m45097952660480

Unlike the innate system which recognizes broad patterns which are shared by multiple pathogens or danger, the cells of the adaptive immune system, T and B cells, utilize highly specific receptors that allow only a few cells to respond to a particular pathogen or malignancy that is currently threatening the host. In other words, while a PRR can identify a pathogen as a bacterium or virus, each T and B cell receptor recognizes individual structures from the specific strain and species of invading bacteria or virus (Fig.

1.2). Rather than being hard-coded in the DNA, T and B cell receptors (TCR and BCR) are generated randomly by combining DNA segments together, like piecing together LEGO blocks. The combination of gene segments as well as the addition of random nucleotides at sites of recombination yield a pool of cells, each expressing structurally unique antigen receptors, which can identify and respond to any pathogenic threat. Although this system generates an extensive diversity of T and B cells, only a small number are capable of responding to any given pathogen. As such, during an infection, these few, specific cells need time to multiply to sufficient numbers to mount a response. Thus, while the innate response is rapid but nonspecific, the adaptive response is highly specific but slow to develop (Fig. 1.3). Together, the two arms of the immune system work in concert to eliminate pathogenic threats to the body.

Figure 1.2. Overview of the receptor classes utilized by cells of the innate and adaptive immune systems. Cells of the immune system use a variety of receptors classes to recognize and eliminate pathogenic threats or malignancies. Cells of the innate system utilize PRR, including toll-like receptors (TLR), mannose receptors, etc., to recognize biochemical patterns specific to pathogens, triggering intracellular signaling pathways that activate and enhance the function of these cells. Fc-Receptors (FcR) can further activate innate cells by binding antibody opsonized pathogens, but they, along with mannose receptors, can also trigger engulfment and degradation of pathogens. Both innate and adaptive cells employ chemokine receptors that allow them to traffic to sites of infection or damage. Additionally, both cells express a variety of cytokine receptors that allow communication between the cells of the immune system and damaged tissue to coordinate and drive an effective response to the given threat. T cells, members of the adaptive immune system, use highly specialized T cell receptors in order to become activated by antigen presentation by APCs as well as to recognize and kill infected cells. Finally, B cells, the other adaptive immune cell type, express B cell receptors that similarly allow for activation, which are then truncated and released as antibodies that can circulate throughout the body and enact effector functions, including neutralization, complement- fixation, and antibody-mediated cell killing (or opsonization). Reprinted with permission from Bhat & Steinman, 20092, license number: 4893270006028.

Figure 1.3. Time course of the immune response. Both diagrams highlight the time course of the immune response to infection. (A) This panel emphasizes the cellular members of the innate and adaptive immune systems as well as indicates the time scale at which they act in defense. Further, it indicates that DCs act as a bridge between the two arms of the immune system, acting to activate the adaptive response. Reprinted with permission from: https://www.creative- diagnostics.com/innate-and-adaptive-immunity.htm.

A key intermediate between these two cellular phases is the dendritic cell (DC).

DCs are innate cells that typically migrate through the tissues in an immature state. When they encounter a pathogen, toxin, or danger (PAMP or DAMP), they are activated by PRR signaling. At the same time, their phagocytic abilities allow them to engulf and carry pieces of pathogen/threat-derived proteins to the nearest (draining) lymph node (Fig. 1.1A). The

DC, presenting antigens from the threats, waits there until the relatively rare T cells that have TCRs which are specific to the pathogen come along. The T cell then engages the peptide-MHC complex (pMHC) on this DC, which provides the signals necessary for the

T cell’s activation, proliferation, and differentiation. There are two classes of T cells, differing substantially in function. CD8 T cells (cytotoxic T cells or CTL) are concerned primarily with killing target cells and CD4 T cells (helper or TH) play multiple regulatory

roles during a response. These classes are identified by key cell surface expression of either the CD8 or CD4, respectively (Fig. 1.4). CD4 T cells, play a critical role in coordinating the overall immune response by secreting cytokines that enhance and direct the activity of all other immune cells. As such, CD4 T cells are also critical for dictating the type of response the immune system makes against an invading pathogen. This role is vitally important as the body must be able respond to the huge breadth of threats, from tiny viruses that replicate within cells to huge, multicellular, parasitic worms, each of which require different strategies for elimination. It is not completely understood how a CD4 T cell takes in the wealth of information it gets during a pathogenic invasion and “decides” how to respond.

Figure 1.4. Summary of CD4 and CD8 T cell function. Helper T cells (TH) are identified by the co-receptor CD4 that aids in binding of the TCR to MHC-II found on antigen presenting cells (APC). Once activated, TH secrete cytokines that help other cells of the immune system better fight this infection. The type of cytokines released specifically amplify the types of functions necessary for clearing the particular pathogenic threat. CD8 is the co-receptor that defines cytotoxic T lymphocytes (CTL), which are activated by and recognize antigen presented on MHC-I. This interaction allows CTL to identify cells of the body that are infected with the given pathogen, and directly eliminate these cells, preventing further spread of the infection. Reprinted with permission from: http://www.glycopedia.eu/e-chapters/Overview-of-Immune-Responses-A-Primer-72/Cellular-adaptive- immune-responses

The widely accepted paradigm of T cell activation and differentiation, known as the “3 signal model” (reviewed in detail below), focuses on the signaling cues provided from the DC while it presents the invading pathogen to the inactive T cell. Importantly, while T cells await activation signals from DC, they do not exist in isolation, but instead in a rich, dynamic environment of other immune and non-immune cells and signaling molecules. However, the 3-signal model does not provide a framework for understanding these environmental cues provided to the T cell before it interacts with the DC. This thesis aims to expand the 3-signal model by describing the implications of environmental signaling prior to T cell activation encompassed in an additional “Signal 0.”

1.2. The road to the 3 signal model: the current paradigm for T cell activation and differentiation

Activation and differentiation of T cells is currently understood in terms of the stepwise engagement of three sets of receptors defined within the three-signal model3,4.

This model emerged over decades of research, starting from the cloning of the TCR, understanding of its activation rules, elaboration of the two-signal model, and finally the appreciation of a role for the third signal 5,6. Below, we discuss the origins of these concepts and how it is relevant to the main topics considered in this thesis.

1.2.1. The two-signal model

Each T cell senses its antigenic target through a specific TCR, as we discussed above. The TCR engages peptides presented by MHC molecules. Once the TCR binds a cognate pMHC, a series of signaling pathways are engaged (downstream of the TCR- associated CD3 signaling complex)7. These signals, which indicate to the T cell that it has successfully found and engaged an appropriate pMHC, are collectively referred to as Signal

1. In many receptor-ligand interactions, the engagement of the receptor with the ligand is sufficient to trigger a functional response that is associated with the ligand. The TCR is unique in that even complete and strong signal transduction downstream of the TCR is insufficient for proper T cell activation. Experiments by Jenkins and Schwartz8 showed that by itself, this is inadequate to fully drive the activation of T cells, especially those in the naïve state. These experiments, showed that when antigen-presenting cells were chemically treated such that they were still able to present pMHC but could not upregulate other ligands for stimulating T cell surface molecules, the result was the inactivation of the

T cell rather than activation. Additionally, several signaling and metabolic changes are induced, leaving the cell in a state of “anergy” unable to respond to subsequent activating stimuli8-11. In time, this 2nd signal or co-stimulation has been defined as critical to drive full activation of the T cell12,13. Following the seminal Jenkins and Schwartz experiments, the precise molecular underpinnings of co-stimulation were resolved when the major co- stimulatory receptor, CD28, was cloned14. CD28, which binds B7 molecules CD80 and

CD86 on DCs, is the prototypic co-stimulatory receptor15, mostly owing to its essential role for activation of naïve T cells. Several co-stimulatory receptors have since been discovered, including ICOS, OX40, CD40L, 4-1BB, GITR, CD27, and CD30, that have varying influences on T cell proliferation, cytokine production, cytotoxic function, etc., and act in different contexts16; nevertheless, CD28 still functions as a unique player as it can help trigger naïve T cell activation.

The idea of multiple signals being essential for lymphocyte activation pre-dates 2- signal model for T cell activation12. The long historical record that has led to the modern

2-signal model (Fig. 1.5) originates from attempts to explain the fundamental question of

lymphocyte biology: what drives cell activation and response to an antigen versus tolerance and non-responsiveness, i.e. how does a lymphocyte “decide” what to respond to? An early model proposed by Lederberg postulated that newly developed, immature cells are poorly antigen reactive but maturation yields immune-competent cells17. Thus, this model only requires one antigenic signal to activate a lymphocyte as its responsiveness is dictated by developmental time. The observation that, while both “carrier” and “hapten” molecules could generate antibodies, haptens were only immunogenic when linked to carriers but carriers did not require haptens to elicit responses, necessitated a consideration of antigenic structure to explain responsiveness18-23. A second model, by Talmage and Pearlman, proposed that antigen alone yields low-level antibody production that fades to a tolerant state, while another protein that acts to aggregate the antigen in a non-specific way, such as complement, could provide a second, boosting signal that yields strong lymphocyte proliferation and antibody release24. Similarly, the observation that tolerized cells could be activated by linking the tolerizing antigen to an immunogenic antigen suggested that the binding of another protein, reactive to the second antigen, to the antigen complex provides a second, activating signal to the lymphocyte25,26. Bretscher and Cohn synthesized these observations by proposing that monomeric hapten-carrier molecules were insufficient for activation but that aggregation by carrier-specific antibodies triggered activation in hapten- specific lymphocytes, later revising this to dictate a signal provided directly by the carrier antibody13,27. This was the first model to require 2-signals, 1 antigenic signal from the hapten and a 2nd signal from the carrier antibody; however, it struggled to explain the origin of the 2nd signal in the activation of the carrier antibody, termed the “primer problem.” This model also illustrated that lymphocyte activation is likely a collaboration between cells as

the clonal selection theory28 eschewed the idea that any one lymphocyte would have receptors specific to multiple antigens. The notion of collaboration was significantly bolstered upon the discovery of the distinct B and T lymphocyte lineages and the observation that they act synergistically to in antibody production29-31. Further refinement of Bretscher and Cohn’s model to replace the carrier-specific antibody with the carrier- specific T cell essentially leaves us with the modern 2-signal model for B cell activation.

Importantly, however, the primer problem still remains as there was yet no understanding of what would provide the 2nd activation signal to T cells.

In parallel, work by Lafferty and Jones studying graft-versus-host disease models determined that the activation of lymphocytes was species specific, as xenogeneic cell transfer always yielded inferior responses to allogeneic32. On the basis of further studies using allografts, Lafferty proposed that, much like the Bretscher and Cohn model, rejection required recognition of antigen as well as signals to proliferate provided by the “stimulator” cells. It was further determined that “stimulator” cells are of hematopoietic origin, must be metabolically active, and are transferred to the host within the graft32,33. Of note, cytotoxic function of T cells was no longer dependent on the 2nd signal following initial activation33.

Summarizing these findings, Lafferty and Cunningham proposed a model in which

“stimulator” cells with rejection-mediating “responder” lymphocytes to provide both signal 1 in the form of antigen either expressed by these cells or merely bound to them as well as a second, boosting signal, activating responder cells in a way that removes the requirement for further secondary signal34. Upon discovery of MHC restriction and its role in alloreactivity by Zinkernagel and Doherty, it became clear that MHC molecules expressed by “stimulator” cells provided the molecular link between these cells and T cells

in providing antigenic signal 135. Returning to and incorporating the work of Jenkins and

Schwartz discussed above8 yields a 2-signal activation model of T cells summarized as follows: 1) APCs present peptide antigens bound to MHC molecules, providing the antigen-specific first activation signal to T cells; 2) upon engagement of antigen binding with T cells, APCs actively express a second, co-stimulatory molecule that when bound by

T cells provides a second, proliferative signal. However, although Jenkins and Schwartz showed that the absence of signal 2 provided by the APC yields unresponsiveness, this model did not allow for this situation to occur in living cells, and, thus, did not allow for tolerance.

The final piece of the puzzle was provided by Charles Janeway when he proposed that, instead of being induced by pMHC engagement with the T cell, co-stimulatory receptor expression by APCs is induced upon interaction with infectious organisms36.

Based on absolutely no data, Janeway predicted that APCs express receptors that are capable of identifying structural motifs specific to foreign organisms and that upon ligation of these receptors, the APC would then provide the key co-stimulatory second signal to T cells. This revolutionary idea elegantly explained a mechanism for self-non-self- discrimination and thus tolerance. Critically, the subsequent discovery of several classes of PRR provided immense support for Janeway’s model. However, Polly Matzinger’s

“danger hypothesis” instead proposed that APCs are induced to express co-stimulatory molecules in response to encounter with signs of tissue damage, such as intracellular molecules located extracellular indicating necrosis or other “inducible alarm signals” like

IFN release37. Her model allows for an understanding of commensal microorganisms that must express non-self motifs but do not elicit immune responses. A modern understanding

of these two theories suggests that integration of microbial structure and danger molecule- induced signaling drives APC activation and co-stimulation expression.

Figure 1.5. Evolution of the 2-signal model. Summary of the critical discoveries and the proposed mechanisms of lymphocyte activation. (A) The timeline covers a period beginning with the discovery that carriers and haptens are both capable of eliciting lymphocyte response but that haptens are dependent on the presence of carrier molecules to do so. It culminates following the discovery of two distinct lymphocyte classes and the proposal that they collaborate in their activation. (B) This timeline takes us through the definition of the modern 2-signal model and the discovery of the microbial pattern recognition receptor Janeway envisioned nearly a decade prior. Reprinted with permission from Baxter and Hodgkin, 200212, license number: 4893290685806.

1.2.2. A third signal emerges

The initial formulation of the 2-signal model dealt with one aspect of lymphocyte biology – i.e. the activation of T cells for the first time. It was soon evident that in addition to these ‘triggers’ there was a 2nd question to be resolved involving the acquisition and maintenance of effector functions or ‘classes’ of T cells and eventually contributing to immunological memory. The key aspects for this phase therefore involve T cell differentiation or the process by which a naïve T cell acquires effector functions.

When they leave the thymus, the mature T cell is in a naïve state which, as discussed above, is fundamentally inert. Upon activation, T cells release IL-2 and upregulate the third chain of the IL-2R, CD25, allowing for high affinity autocrine signaling and a period of robust proliferation, a process that has received intense examination for decades. During this time, T cells also initiate transcriptional programs that drive the effector functions they use to eliminate immunologic threats. As discussed in section 1.1, the immense breadth of potential pathogens one might encounter necessitates a similarly broad range of mechanisms to combat them. For example, in the case of viral infection, a critical clearance mechanism employed is the direct killing of infected cells by CD8 T cells. Thus, mature, functional CD8 T cells are elicited upon activation by upregulating the transcription factors

T-bet and Eomes which promote cytolytic programs, but also the production of IFN and

TNF which aid in eliminating compromised host cells. CD4 T cells can augment these processes by also producing IFN and TNF, and, as such, activated CD4 T cells will differentiate into TH1 cells which also express T-bet to drive cytokine production.

Alternatively, in the case of parasitic worm infection, CD4 T cells will become TH2 cells that express GATA-3 and produce IL-4, IL-5, and IL-13, which promote eosinophil and

mast cell degranulation, mucin secretion, etc. that act to degrade and expel the much larger pathogens. Thus, eliciting appropriate differentiation, particularly of CD4 T cells that can express a variety of effector programs to coordinate varying immune responses, is critical for clearing an infection. While the 2-signal model effectively describes how a pathogen- specific T cell “chooses” to initiate a response to the infection, it is unclear how these two signals alone can dictate the type of response necessary to combat the infection. T cells must receive some other input, a 3rd signal, which communicates the necessary differentiation programs to enact. Returning to Janeway’s PRR-mediated DC activation hypothesis provides insight on this process.

During an infection, contact with a microbial threat via PRR engagement will trigger DC activation. PRR signaling induces transcriptional changes, mediated predominantly by NFB and IRF3/738, that lead to the upregulation of MHC-II and co- stimulatory molecules, and migration to secondary lymphoid organs (SLO) to engage inactive T cells39. Further, mature DCs release cytokines, dictated by the type of PRR engaged, and thus the type of pathogen it bound. Thus, there is a unique cytokine profile secreted by DCs dependent on the class of microbial threat it encountered at the site of infection. During DC-T cell interactions in the SLO, it is these cytokines that “skew” the differentiation program of the activated T cells. Over decades of research, the specific cytokine milieu that elicits each type of CD4 differentiation program have been well characterized and are summarized in Figure 1.6. Critical skewing cytokines include IL-12 for TH1 cells, IL-4 for TH2 cells, IL-6 and TGF for TH17 cells, IL-6 and IL-21 for TFH

40-42 cells, and TGF yielding Treg .

Once the T cell has acquired effector functions, the related priority for the immune system is to sustain this response, both in the short term, long enough to clear the pathogen or malignancy, as well as in the long-term for the formation of memory. Many studies have investigated the role cytokines play in these functions, converging on signaling mediated by a family of cytokine receptors that share a “common -chain”, which includes the IL-2 receptor43,44. Members of this family, IL-7 receptor and IL-15 receptor, binding the cytokines for which they are named, have been well established as critical drivers of long- term survival and homeostatic proliferation of naïve and memory T cell, with naïve T cells especially dependent on IL-7 availability45,46 Additionally, during the contraction phase of the T cell response, these cytokines distinctly support the accumulation of short-lived effector cells (SLEC) and memory precursor effector cells (MPEC), with IL-15 favoring the former and IL-7 favoring the latter47.

Figure 1.6. Summary of CD4 T cell differentiation. Upon activation by cognate antigen presentation by DC, CD4 T cells differentiate and produce characteristic cytokine profiles. Which effector program the T cell adopts is driven by cytokine signaling it receives from the DC, driving the expression of transcription factors that enforce the effector function. Reprinted with permission from Nurieva & Chung, 201048, license number: 4893290829915.

Figure 1.7. Common -chain cytokines control T cell proliferation, survival, and lineage commitment. T cells require trophic factors at several phases of the immune response to proliferate and survive. The specific effects of these trophic factors are stage specific, with IL-2 most critical for driving clonal expansion after activation, IL-15 supporting effector T cells, and IL-7 predominantly responsible for maintaining naïve and memory T cells. Reprinted with permission from Schluns & Lefrancois, 200344, license number: 4893290916111.

Interestingly, cytokines have also been shown to effect T cells in ways that resemble input from signal 2. Inflammatory cytokines including IL-1 and IL-12 can promote proliferation of T cells, predominantly CD4 and CD8 T cells, respectively, replacing the role for adjuvant-mediated co-stimulation49. In CD8 T cells, IL-12 upregulates CD25, allowing for increased IL-2 dependent proliferation, even in the absence of co-stimulatory CD28 signaling50. Further, type I IFNs can act to support survival of CD8

T cells, cytotoxic function, and IFN release in a STAT4-dependent manner51. Taken together, these regulatory cytokines now form a third group of signals that completes the current definition of a three-signal model for T cell activation: 1) antigen recognition by the TCR, 2) co-stimulation, and 3) cytokine signaling, collaborate to provide the necessary stimulatory cues to generate full and effective T cell activation (Fig. 1.8).

Figure 1.8. The 3-signal model of T cell activation. The accepted model of T cell activation describes the provision of 3 signals from DCs that synergize to instruct T cells when and how to activate. 1) Cognate pMHC is presented by DC and engages the TCR providing the major activation signal. 2) Co-stimulatory molecules act to drive T cell survival and proliferation, preventing T cell non-responsiveness. 3) Cytokines provide differentiation cues, dictating which effector functions to employ while activated. Figure adapted from Murphy & Weaver, Janeway’s Immunobiology, 9th Ed.52

1.2.3. The third signal is biochemically distinct from Signals 1 and 2

A key aspect of signal 3 that is very relevant to this thesis is the intracellular nature of biochemical signals transduced by these cytokines. Signal 2 (at least that from CD28) is mostly reported to be a synergistic signaling pathway to the TCR. A more granular discussion of TCR signaling is found in section 1.2.5, but these are quite different from the signals used by cytokines. Most cytokine receptors utilize JAK-STAT receptor signaling pathways (Fig. 1.9A). Upon ligand binding, cytokine receptors dimerize, allowing the associated Janus kinase (JAK) – a non-receptor tyrosine kinase (nRTK) – molecules bound to each subunit to localize and undergo transphosphorylation of the opposing activation loop. This allows the JAKs to phosphorylate tyrosine residues on cytoplasmic tails of the receptors, allowing for binding of SH2 containing signal transducer and activator of 17

transcription (STAT) proteins. The STATs are then tyrosine-phosphorylated, releasing from the receptor tail, and forming homo- or heterodimers by binding the newly generated

SH2 binding domain on another pSTAT molecule. STAT dimers are then translocated to the nucleus where they act as transcription factors to drive diverse functions including proliferation and differentiation.

Among cytokine receptors, it is common for receptor subunits to be shared between receptor dimer pairs such that cytokines and the receptors they share are often thought of as families. As mentioned above, receptors for IL-2, -4, -7, -9, and -21 all share the common

-chain (CD132) are dubbed the “common -chain family” (Fig. 1.9B). As a consequence of sharing this receptor subunit and its associated JAK3, all of these receptors elicit STAT5 translocation, and, thus, similar functional consequences, e.g. T cell proliferation. The remaining subunits that are unique to the given receptor provide, then, varying concurrent

JAK and STAT combinations, yielding potentially cytokine specific programs. STAT6, elicited by IL-4 signaling, critically induces GATA-3 expression and TH2 differentiation

(Fig. 1.6). This combinatorial scheme allows for rich complexity of functional outputs from cytokine signaling while utilizing few unique gene sequences.

The IL-12 family of cytokines (Fig. 1.9C) similarly employs common receptor subunits and JAK-STAT signaling; however, it intriguingly also takes advantage of shared cytokine subunits53. The family name sake, IL-12, is a heterodimer of p40 and p35 proteins that bind and induce heterodimerization of IL-12R1 and IL-12R2, respectively. JAK2 and TYK2 activation yield homo-pSTAT4 dimers that are critical for inducing IFN production by TH1 cells, in collaboration with T-bet. The sister cytokine, IL-23, shares p40 with IL-12 but instead forms a heterodimer with p19. It then binds IL-12R1 with IL-23R,

which yields pSTAT3 and pSTAT4. Importantly, this blended pSTAT3/4 profile drives differentiation of TH17 cells. Thus, the swapping of one cytokine subunit for another dictates which effector functions are employed by the CD4 T cell and the type of immune response made to a given pathogen. This example effectively illustrates both the elegance and the power of a cytokine component to generate the 3-signal model.

These 2 families highlight a particularly salient point regarding cytokine signaling: shared signaling pathways beget overlapping changes in T cell behavior. As mentioned above, all of the common -chain cytokines trigger STAT5 translocation and promote T cell proliferation and survival. Importantly, cytokines from other receptor families including TSLP54, IL-3, and GM-CSF55-57 signal through STAT5 phosphorylation yielding similar promotion of T cell survival. Similarly, despite disparate receptor structures, IL-6,

IL-21, and IL-23 all signal via pSTAT3 and, perhaps unsurprisingly, all contribute to TH17 lineage commitment58.

While most cytokines trigger JAK-STAT signaling, TGF is an example of one that uses a unique, albeit similar, signaling pathway (Fig. 1.10)59. TGF receptors

(TGFR), heterodimers of TGFR1 and TGFR2, are serine/threonine kinase receptors, thus containing intrinsic kinase ability, unlike JAK coupled receptors. Upon cytokine- receptor ligation, two heterodimer pairs form a heterotetramer and allows the cytoplasmic kinase domain of TGF2 to phosphorylate TGF1, activating its own kinase function.

SMAD2 and SMAD3 can then be recruited to the receptor and phosphorylated by

TGFR1. pSMAD2/3 heterodimers form heterotrimeric complexes with SMAD4 allowing for nuclear translocation where SMADs, much like STATs, act as transcription factors.

Additionally, though not depicted, TGFR1 can also elicit activation of other signaling

pathways including Ras/MAPK and PI3K59,60. TGF has particularly pleiotropic effects on

T cell biology. It has been well established to act as 3rd signal to skew differentiating CD4

61-63 64 T cells away from TH1 and TH2 by the inhibition of T-bet and GATA-3 , respectively,

65,66 and instead promote Treg via induction of Foxp3 . TGF can also block the production of IL-2 and T cell proliferation67 by inhibiting TCR-proximal signaling nodes68,69, yielding blockade of activating signal 1. Intriguingly, a recent study found that tonic TGF signaling to naïve T cells from migratory DCs prior to antigen priming was necessary for tissue

70 resident memory T cell (TRM) formation . This point is incredibly powerful for two reasons: 1) at the point of antigenic activation, not all T cells have the same memory cell differentiation potential based on prior exposure to cytokines; and 2) the 3-signal model provides no framework for understanding these differences.

Figure 1.9. Summary of JAK-STAT signaling by cytokine receptors. (A) Many cytokines use a common mechanism for activating JAK-STAT signaling. Ligand-mediated receptor dimerization yields JAK transphosphorylation, followed by phosphorylation of the receptor tail and recruited STAT molecules. STATs then dimerize and translocate to the nucleus to act as transcription factors that mediate function. Reprinted with permission from Morris, et al. 201871, license number: 4893270820990. (B) The common -chain family of receptors share the -chain receptor subunit that couples to JAK3 and STAT5 signals. Thus the receptor family has overlapping functions based on this commonality, but the other recruited STATs dictated by the remaining receptor subunits provide unique flavoring to each cytokine response. Reprinted with permission from Rochman et al, 200943, license number: 4893291049860. (C) The IL-12 family of cytokines shows shared cytokine domains as well as receptor subunits, allowing for the swapping of cytokine domains to dictate CD4 differentiation. Reprinted with permission from Vignali & Kuchroo, 201253, license number: 4893291154641.

Figure 1.10. Schematic of TGFR signaling. Similar to JAK-STAT signaling, ligand binding triggers receptor dimerization and phosphorylation using intrinsic kinase function. SMADs are then recruited to the receptor tail, phosphorylated, and then translate to the nucleus to act as transcription factors. Reprinted with permission from: https://www.mycancergenome.org/content/pathways/TGF-beta-signaling/

1.2.4. A more significant role for the TCR in differentiation and cell survival

The initial formulation of a 3-signal model had a lasting impact on the field of T cell immunology in general. Over time, emphasis on understanding T cell differentiation and survival has largely focused on cytokines. Nevertheless, it was evident from even early studies that the nature of TCR signaling allows it to have a contribution in eliciting effector functions. Namely, the intensity of antigenic stimulation plays a critical role in driving T cell memory fate and CD4 differentiation.

Several studies have compared the magnitude and quality of T cell response elicited by varying intensities of TCR stimulations. While T cells have been shown to elicit effector functions, including IFN and cytotoxicity, when stimulated with low concentrations of antigenic peptide, proliferation is limited, requiring higher doses of antigen to elicit full activation and prolonged T cell responses72,73, possibly requiring increased duration of

TCR engagement with pMHC74. Similar findings were observed using low-affinity antigens, but these experiments also showed that, while low-affinity antigens promoted improved memory responses to high affinity antigens75. Collectively, these data suggest that, while weak TCR signaling can elicit function, increased signal strength is required for the robust proliferation needed to generate an effector response. Accordingly, it is currently believed that lower TCR signal strength results in the generation of memory, by default76,77.

Beyond the role for signal strength in CD8 driving effector and memory responses,

TCR signaling intensity has been shown to regulate CD4 differentiation. In vitro studies have shown that TH1 and TH2 differentiation and cytokine production exist on a spectrum, with high doses of antigen, which elicit increased ERK phosphorylation, yielding the former helper subset while low antigen doses promote the latter78,79. A similar paradigm

has been observed between TH17 differentiation and Treg induction, with lower antigen

80,81 doses or lower affinity antigens yielding increased frequencies of Foxp3+ Treg .

1.2.5. The components of TCR signaling

The classical biochemical network downstream of a TCR has been well characterized and reviewed7,82-88 (Fig. 1.11). After briefly discussing the initiation of TCR signaling and its early propagation events, this section will focus on the distinct, downstream signaling modules that are elicited and the contribution of each to T cell function.

The complete TCR complex itself is comprised of 8 protein subunits: the TCR  and  chains, which are randomly generated during thymic development by V(D)J recombination, that underlie antigen specific binding of the pMHC complex; and 6 intracellular CD3 molecules (1, 1, 2, 2 chains) that coordinate the initiation of signal transduction upon TCR ligation. It is yet unclear exactly how TCR ligation is propagated to intracellular signaling molecules; however, when the TCR binds its cognate pMHC, Lck, a Src-family kinase that associates with the cytoplasmic tail of the co-receptor molecule

(CD4 or CD8), is activated and phosphorylates tyrosine residues within ITAM domains of the CD3 cytoplasmic tails, most critically CD3, which contain three ITAMs each compared to one among other CD3 isoforms. Phosphorylated ITAMs allow for the recruitment of Zap70 via binding SH2 domains. Lck phosphorylates docked Zap70, activating its kinase activity, and triggering a cascade of signaling events. Lck and Zap70 yield the phosphorylation of Itk and the assembly of the scaffolding complex comprised of pLAT:Gads:pSLP-76. The LAT complex is a critical docking site for phosphorylation of

PLC by pItk, which elicits several branching signaling modalities. LAT also coordinates

the initiation of signaling through SOS1 and Vav1, triggering MAPK cascades that integrate with PLC-mediated signaling (discussed further below). Thus, signaling proximal to the TCR is accumulated until it is sufficient to establish the activated LAT complex. From there disparate signaling modalities are triggered that spread to trigger the necessary cellular changes to support T cell activation and differentiation. We next consider these signaling modules individually.

A common framework for understanding distal TCR signaling surrounds considering the resulting changes in transcription they elicit. More specifically, because the production of IL-2 by newly stimulated T cells is a definitive marker of activation that requires the integration of several transcription factors TF (TF) (Fig. 1.12), minimally

NFAT, NFB, and AP-1, we can consider those that drive each of these 3 TF as independent signaling modules. As indicated above, activation of PLC as is critical initiating signaling node for all 3 modules, with pPLC cleaving membrane phospholipid

IP2 into DAG and IP3. While DAG remains localized to the plasma membrane, IP3 diffuses to the ER membrane where it engages its receptor, IP3R, releasing ER Ca2+ stores into the cytosol. This Ca2+ flux is positively reinforced by activating calcium release activated channel (CRAC) opening and additional cytosolic Ca2+ influx from the extracellular space.

Calcineurin, a Ca2+ activated protein phosphatase, then dephosphorylates cytosolic p-

NFAT, allowing the TF to translocate into the nucleus. In addition to contributing to IL-2 transcription, NFAT has been shown to be critical for generating CD40L-dependent help to B cells89. Additionally, NFAT complexes with the CD4 lineage defining TF (T-bet,

GATA3, RORt, Foxp3, etc.) and relative expression of the NFAT isoforms NFAT1

(NFATc2), NFAT2 (NFATc1), and NFAT4 (NFATc3) have been shown to differentially

promote TF activity90. Most notably, a predominant hypothesis for the molecular mechanism of anergy induction is that NFAT is induced in the absence of AP-1, promoting a unique transcriptome that enforces T cell nonresponsiveness91.

2+ In parallel to IP3-mediated Ca signaling, DAG recruits and activates PKC that phosphorylates CARMA1 to initiate a scaffolding complex that results in activation of IKK via phosphorylation. IKK, in turn, phosphorylates IB triggering its proteasomal degradation and allowing NFB nuclear translocation. NFB TFs are ubiquitous across mammalian cells with diverse functional roles in each cell type dependent on the individual

NFB subunits expressed; however, in T cells a major role of NFB is to promote cell survival through the expression of Bcl-xL92 among other mechanisms93,94.

DAG also recruits RasGRP to the membrane where it activates Ras, inducing the

ERK MAPK signaling cascade. ERK signaling is reinforced by activation of SOS1 by Grb2 at the LAT complex, similarly inducing Ras/Raf signal transduction. Other MAPK p38 and

JNK are induced by via VAV1 activation of Rho GTPases Rac and Cdc42. All 3 MAPK pathways yield phosphorylation of AP-1 TF, allowing for nuclear translocation and transcriptional regulation. Although AP-1 TF encompass a variety of subunit isoforms95,96, in T cells, heterodimeric AP-1 is typically formed by Fos and Jun. Despite convergence at the TF level, each MAPK pathway has separable impacts on T cell behavior, with ERK2 and JNK1/2 contributing to survival and proliferation97, while the extent of ERK signaling

79,98-100 specifically tilts the balance of TH1 vs TH2 differentiation . Further, ERK-induced

AP-1 activity has been shown to be critical for CD69 expression101,102, which acts to regulate T cell egress from SLO and tissues103-105. Additionally, as mentioned above, AP-

1 is critical for balancing NFAT activity to prevent anergy91.

Thus signal 1, recognition by the TCR of antigen presented on MHC, initiates a complex orchestra of signaling cascades, eliciting the activity of several TF that collaborate to dictate several aspects of T cell biology including proliferation, survival, migration, and differentiation. Yet, this complexity provides diverse opportunities for signals 2 and 3 to synergize with or regulate T cell behavior.

Figure 1.11. Schematic of T cell receptor signaling pathways. TCR engagement of cognate pMHC triggers phosphorylation by Lck of ITAMs on the associated intracellular CD3 molecules. Zap70 is recruited to pCD3 where it is activated by Lck, initiating a cascade of kinase activity and the generation of the LAT complex. This is a critical signaling node, as it serves as the branching site for several signaling modules, each culminating in activation of TF that collaborate to activate the T cell. Reprinted with permission from Gaud et al. 2018106, license number: 4893291279547.

Figure 1.12. Cooperation of several TF are required for the full activation of IL-2 transcription. The TCR signaling modules illustrated in Figure 1.11 each drive activation of individual TF that have diverse roles in mediating T cell activation. However, all of these TF must act cooperatively at the IL-2 promoter to elicit transcription and effective T cell activation, as incomplete subsets of these TF yield hypo- or non-responsive T cells. Reprinted with permission from Rothenberg & Ward, 1996107, Copyright 1996 National Academy of Sciences.

1.2.6. Co-stimulatory receptor signaling

This section will briefly discuss the signaling modalities costimulatory receptors, providing signal 2 of T cell activation, employ to mediate their effects. Owing to its obligatory role in naïve T cell activation, CD28 is by far the best characterized co- stimulatory molecule. The cytoplasmic tail of CD28 contains several tyrosine containing motifs that, upon ligation with B7, are phosphorylated by protein tyrosine kinases including Lck108-110. This provides the necessary docking site for the regulatory p85 domain of PI3K via SH2 interactions, allowing for the activation of PI3K111, which is thought to provide the major signaling effect of co-stimulation. PI3K catalyzes the phosphorylation of PIP2 to PIP3 which allows for the recruitment and activation via PH domains of PDK1. There are two critical targets of PDK1 that appear to underpin the costimulatory effects of CD28: PKC and Akt. Activating PKC provides direct input into the TCR pathway described above, reinforcing NFB and AP-1 pathways. In this manner, CD28 signaling can bolster AP-1 induction and provide balance to NFAT induced by Ca2+ signaling and preventing anergy91. Further, Akt signaling is complex and contributes in many diverse ways to T cell activation and differentiation, but, generally,

Akt promotes cell survival and proliferation and regulates T cell metabolism via mTOR112,113.

Beyond the activation of PI3K, CD28 acts as a dock for many other signaling proteins via phosphorylation of its serine and threonine containing motifs as well as proline- and lysine-rich domains108. Recruiting scaffolding molecules like Grb2 and Gads allows for signaling cascades resulting in JNK/AP-1 and NFB activation; however the

relative contributions of these pathways as stimulated by CD28 are poorly understood108,114.

1.3. Timing and order of signals matter.

A central tenet of the 2-signal and 3-signal models for T cell activation is the ordering of these events. In the classic 2-signal model, TCR signaling has to precede or be concomitant with costimulatory signals. Studies using “superagonist” CD28 engagement provided some evidence that CD28 could act to stimulate T cells without the presence of ongoing TCR signaling115, potentially challenging the 2-signal model. However, subsequent studies found that CD28 superagonism was dependent on low-level, tonic TCR signaling as the elimination of TCR-proximal SLP-76:Vav1116 or Zap70117 negated the superagonistic effect. These observations reinforce the model that TCR-mediated signal 1 is absolutely necessary for activation and that signal 2 alone is insufficient, thus requiring that both occur concurrently. Interestingly, signal 2 may not be required at the exact same time as signal 1, as a delay in CD28-ligation of up to 2 hours shows minimal diminishment of T cell proliferation, with increased dose or duration of TCR stimulation showing even less sensitivity to delayed CD28 signaling118. Although there is some flexibility in the precise timing requirements of CD28 signaling relative to that of TCR, it is clear that TCR engagement must occur with or slightly before co-stimulation in order to effectively activate a T cell.

Beyond the temporal requirements of signals 1 and 2, emerging evidence emphasizes that the timing of receiving cytokine-mediated signal 3 is also critical for successful activation and differentiation. It is generally thought that naïve T cells are refractory to cytokine signaling as many cytokine receptors, including IL-2R, IL-12R,

IL-4R, IL-23R, etc., are minimally or not expressed but show strong upregulation upon activation. However, observations from studies in which naïve T cells were incubated with

IL-2 prior to receiving antigen and co-stimulatory stimuli show that these T cells exhibit decreased proliferation upon activation in vitro and in vivo119. Pre-treatment imposes altered cytokine signaling as STAT5 phosphorylation is diminished upon activation among

T cells that had previously received IL-2 stimulation, suggesting the induction of negative feedback119. Interestingly, these effects are transient as IL-2-treated cells recover comparable proliferative capacity within 8 and 20 days, depending on experimental conditions119. Additional studies illustrate a similar effect with respect to T cell differentiation, as naïve T cells merely exposed to IL-4 during a TH2 response to parasitic worm infection become unable to mount an effective TH1 response to subsequent bacterial

120 infection, showing significant reduction in IFN production in favor of IL-4 . The TH2 cytokine milieu, elicited widespread STAT6 phosphorylation in lymph nodes, suggesting that naïve cells are in fact responsive to cytokine signals while in an inactive state120. These data show that cytokine signaling as part of signal 3 shows dramatic, temporal-specific effects on T cell activation and differentiation.

At the core of the 3-signal model is a framework that enforces tolerance: ensuring that the DC that encountered the pathogenic threat via pattern recognition directly engages antigen specific T cells and delivers the requisite co-stimulation locally to prevent widespread activation of potentially self-reactive T cells. For this model to ensure tolerance, naïve T cells must show a similar baseline propensity for activation and differentiation. A T cell that had previously received a cytokine signal, for example, that lowered the threshold for activation, poising it to respond to antigenic stimulation, may not

require co-stimulatory signaling and could activate on the basis of TCR-ligation alone.

Such a scenario would allow an autoreactive T cell, one that would normally be unable to activate as it would not encounter its antigenic peptide in the context of an activated, co- stimulatory receptor presenting DC, to be aberrantly receptive to activation. The current 3- signal model is unable to account for differences in naïve T cell baseline threshold for activation that can be modulated by prior signaling, forming the basis for several questions this dissertation seeks to answer.

1.4. Other signals can affect TCR signaling

While the 3-signal model has helped evolve a big-picture understanding of how different kinds of signaling entities, are sequentially required to complete T cell activation and differentiation, it certainly does not capture all the possible players in these processes either. As currently framed, only T cell-APC interactions yielding antigenic and co- stimulatory signals or paracrine acting cytokines released within the SLO can be understood to contribute to T cell activation and differentiation. It is patently clear that the

3-signal model does not account for all potential inputs T cells encounter.

One class of physiological signaling molecules are perfectly suited to challenge the

3-signal model: hormones. Most hormones are released into circulation, making them rather long-acting in both time and space. As such, they have the potential to broadly alter the baseline potential of inactive T cells, as discussed in the previous section. Fibroblast growth factor (FGF), for example is a systemically circulating growth factor that has been shown to enhance T cell activation by inducing PLC signaling121-124. As such, FGF binding to T cells would be predicted to lower the threshold for activation, potentially eliminating the requirement for co-stimulation to activate. In fact, it was observed that in a

model of myocardial injury, FGF-2 yielded increased T cell infiltration into the heart tissue, including FGF-R expressing cells, as well as increased immunopathology125. In addition to modulating the activation of T cells, hormones may alter signal 3, biasing differentiation.

It was recently shown that IGF signaling through IGF-1R promotes TH17 effector differentiation by driving enhancing Akt and STAT3 pathways126.

Similarly, how the T cell interacts with metabolites in its environment can have a significant impact on activation and differentiation. As the proliferative burst that comes with activation is an energy intense process, differential access to energy sources such as glucose would likely yield distinct proliferative capacities. A recent study examined T cell activation under hypoxic conditions, observing Glut1, a glucose transporter, upregulation by these cells resulting in increased activation127. Similarly, when cells were sorted on the basis of Glut1 expression, Glut1-Hi cells showed greater proliferation and IFN production than the Glut-Lo counterparts127. It may seem obvious that limited access to energy would yield hypo-responsive cells; however, another characteristic of the cellular milieu, salinity, has been shown to be impactful to T cell differentiation as a high salt environment enhances

128 129 acquisition of pathogenic TH17 effector functions at the expense of Treg induction .

These examples illustrate that the 3-signal model is insufficient to explain the serious impact that the context under which APC-T cell interactions occur can have on T cell activation.

The 3-signal model fails to account for the sheer complexity of the environment a

T cell exists in at all times. Although naïve T cells are often thought of as being functionally inert, they are constantly bathed in a diverse milieu of cytokines, metabolites, hormones, cell-cell interactions, etc., of which the naïve T cell will be responsive to at least a subset.

Importantly, the signaling modules induced by many of these stimuli are well known: FGF acts broadly to elicit ERK, Ca2+, NFB, and Akt signaling130; IGF-1 is more limited, triggering JNK and Akt131; the erythropoietin receptor (EPO-R) is coupled to JAK2 and induces STAT5 nuclear translocation132. A T cell that is exposed to these factors during antigen engagement would necessarily show distinct signaling and functional consequences compared to a naïve T cell within a different cellular environment. Further, because we have an understanding of how the individual signaling modules contribute and translate to T cell function, the augmentation or dampening of these individual pathways by combining with environmental signals should be predictable by a model that can incorporate the T cell environment into an activation scheme.

1.5. The new concept: signal 0

Based on these considerations, we propose an update to the 3-signal model: the inclusion of a “signal 0.” Simply, signal 0 encompasses all of the signaling events triggered by encounter with environmental stimuli prior to TCR signaling. Collectively, signal 0 defines that baseline state at which the T cell encounters signals 1-3 (Fig. 1.13). As discussed above, the timing at which cytokine and other environmental cues are received by T cells has a critical role in biasing their subsequent responses. Further, the open-source nature of TCR, co-stimulatory, and cytokine receptor-induced signaling modules, as these are widely shared by most receptor classes, provides an opportunity for signal 0 events to integrate with and bias these pathways when triggered at the time of activation. We believe this reframing provides an opportunity to better understand and predict functional outcomes of T cell responses.

1.5.1. The nature of signal 0

By definition, for a signaling event to contribute to signal 0, it must (a) occur before the initiation of antigenic peptide recognition by the TCR, and (b) trigger a change in signaling that is maintained through the time of antigen encounter. In addition to the cytokine examples provided above, precedence exists in the literature considering prior signaling through the TCR. Prior stimulation with an altered-peptide ligand yields decreased T cell proliferation and expression of activation markers upon subsequent activation with the parent peptide133,134. Further, naïve CD4 T cells receive tonic stimulation in vivo through interactions with MHC-II triggering maintenance of phosphorylated CD3 (p21) and improved proliferative capacity135. Intriguingly, this effect requires constant exposure to MHC-II as the effect is lost among cells in blood circulation135. Additionally, T cells that are repeated or serially triggered through the TCR show an accumulation of phosphorylated ERK136. These data clearly show that signaling differences elicited by prior TCR signals provide a variable baseline for T cell activation.

1.5.2. The source of ligands for signal 0: The non-immune chemiome.

In order to properly conceive signal 0, we must consider the breadth of potential signal 0 events. It must be first noted that the concept of signal 0 does not discount signaling events through TCR, co-stimulatory, or cytokine receptors that occurs prior to encounter with agonistic antigen; however, these signals have been widely studied. Opportunity for particularly novel insights come from broadening our thinking to consider all of the environmental stimuli T cells are exposed to. We use the term “non-immune chemiome”

(NIC) to refer to the totality of these environmental ligands, including hormones, metabolites, neurotransmitters, exogenous drugs, etc. Chapter 4 includes a more detailed discussion of the NIC.

1.5.3. Questions arising a. Does previous tickling of the TCR or the signaling modules it activates affect the subsequent fate of the T cell? b. What are the sources of the NIC in vivo, and which are T cells responsive to? c. Can we develop a strategy to study the NIC, both individual constituents and globally?

Figure 1.13. Schematic of Signal 0 hypothesis. T cells (A) are activated by signals downstream of the TCR, which are coupled to biochemical cascades (X,Y,Z etc.). In the absence of antigen (B), they can also be sensitive to other ligands in our chemiome if they express receptors (Rc) for them. Rc can share some downstream pathways (e.g. X) with the TCR. We predict that these “preconditioning” signals affect eventual TCR activation.

1.6. Thesis hypothesis and Specific aims The overarching hypothesis of this thesis is that non-immunological molecules in the environment, continuously alter basal signaling networks in T cells to bias subsequent T cell activation in a microenvironment-dependent fashion and affect long term T cell fate. We sought to test this overarching hypothesis through the following specific hypotheses and experimental aims: Specific Aim 1. Evaluate the hypothesis that the previous activation of a subset of cellular signaling pathways that are part of the TCR signaling network will alter TCR-driven activation and differentiation. We explore this hypothesis using in vitro and in vivo model systems in Chapter 3. Specific Aim 2. Evaluate the hypothesis that expression of distinct receptors for members of the non-immune chemiome confers differential susceptibility to T cells to modulation by the steady state environmental cues. We examined this by determining the breadth of NIC molecules that may act on T cells by identifying all rNIC expressed by T cells via literature review in Chapter 4. Specific Aim 3. Evaluate the sub-hypothesis derived from #2 that discrete neurotransmitter receptor expression profiles by T cell subsets allows neuronal impulses to differentially impact each T cell subset. In order to evaluate this, we examined neurotransmitters as a model for understanding how the NIC can modulate resting T cells by providing signal 0 using transcriptomic analyses and in vitro models in Chapters 5. Specific Aim 4. Based on studies above, we further explored the hypothesis that modulation of specific TCR signaling modules allows VIP to alter T cell activation and differentiation. We described the specific effects of one NR, vasoactive intestinal peptide (VIP), using in vitro culture and in vivo receptor overexpression systems to model T cell activation and differentiation in Chapters 6.

Chapter 2: Materials and methods

Cell culture

Primary murine thymocytes and T cells were maintained in RPMI-1640 (Gibco) supplemented with 10% FCS (Gemini), 1% glutamine (Gibco), 1% antibiotic-antimycotic

(Gibco), and 0.000014% β-mercaptoethanol referred to as T cell media (TCM). Primary murine bone marrow was cultured in RPMI 1640 (Gibco) supplemented with 10% FCS

(Gemini), 2% glutamine (Gibco), 1% sodium pyruvate (Gibco), 1% antibiotic/antimycotic

(Gibco), 20 ng/mL IL-3 (R&D 403-ML), 50 ng/mL SCF (R&D 455-MC), and 50 ng/mL

IL-6 referred to as bone marrow media (BMM). Phoenix-GP cells were obtained from

ATCC and maintained in DMEM (Gibco) supplemented with 10% FCS (Gemini), 2% glutamine (Gibco), 1% sodium pyruvate (Gibco), and 1% antibiotic/antimycotic (Gibco) referred to as complete DMEM (cDMEM). NIH 3T3 cells were obtained from ATCC. Cells were maintained in RPMI 1640 (Gibco) supplemented with 10% FCS (Gemini), 2% glutamine (Gibco), 1% sodium pyruvate (Gibco), and 1% antibiotic/antimycotic (Gibco) referred to as completed RPMI (cRPMI).

Mice

Wild-type mice were obtained from both Jackson Laboratories (C57BL/6J and

B6.SJL-PtprcaPepcb/BoyJ) and Taconic Biosciences (B6NTac and B6.SJL-

Ptprca/BoyAiTac). B6 mice were bred for dual congenic expression using a dam or sire from opposing place of purchase to eliminate any strain drift between the two wild-type strains. Wild-type B10.Q/Ai mice were originally obtained from Ron Schwartz, Institute of Allergy and Infection Diseases, NIH, Bethesda, MD. Wild-type B10.A-H2a H2-

T18a/SgSnJ mice were obtained from Jackson laboratories. B10.A mice were bred for

dual-congenic expression using a dam or sire from opposing genotypes. 5C.C7 TCR transgenic mice were obtained from Taconic Biosciences (B10.A-Rag2tm1FwaH2-

T18aTg (Tcra5CC7,Tcrb5CC7)lwep). The A1M TCR transgenic on a CBA/Ca,RAG1-/- background were a kind gift of Dr. Steven P. Cobbold, University of Oxford, UK137. They were then bred 9 times to B10.A,RAG2-/- and maintained as a homozygous A1M TCR+ line. SMARTA TCR transgenic mice were obtained from Jackson Laboratories (B6.Cg-

Ptprca Pepcb Tg(TcrLCMV)1Aox/PpmJ). TCRαβ knockout mice were obtained from

Jackson Laboratories (B6.129S2-Tcratm1Mom/J). CD3ε knockout mice were obtained from NIAID/Taconic mouse contract (B10.A-Cd45a(Ly5a)/NAi N5). PCC-CD3ε knockout mice were generated as described 138-140 and obtained from Ron Schwartz,

Institute of Allergy and Infection Diseases, NIH, Bethesda, MD. Briefly, these mice have expression of a membrane targeted form of pigeon cytochrome C (PCC) under the control of the MHC class I promoter and an Ig enhancer and carry a deletion of CD3ε gene such that they do not generate a functional T cell repertoire. PCC+ mice were generated by crossing PCC-CD3ε knock out mice with B10.A wild-type mouse to generate heterozygotes that carry the PCC transgene and generate a full compartment of T cells.

All animals were bred and maintained in modified specific-pathogen free (SPF) facilities at the University of Maryland Baltimore. Experiments were performed with animals at least six weeks of age and approved by University of Maryland Baltimore

Institutional Animal Care and Use Committee. Table 2.1 contains the genetic designation, description of phenotype, and the abbreviation used within this thesis.

Table 2.1. Mouse model genetic designation and abbreviations.

Mouse Genetic Strain Description Abbreviation Used C57BL/6J Wild-type B6 animal B6 B6.SJL-PtprcaPepcb/BoyJ Wild-type B6 animal B6 B6NTac Wild-type B6 animal B6 B6.SJL-Ptprca/BoyAiTac Wild-type B6 animal B6 B10.Q/Ai Wild-type B10.A animal B10.A B10.A-H2a H2-T18a/SgSnJ Wild-type B10.A animal B10.A B10.A-Rag2tm1Fwa H2-T18aTg CD4+ TCR-Tg specific for pigeon 5C.C7 (Tcra5CC7,Tcrb5CC7)lwep or moth cytochrome C peptide B10.A-Rag2tm1Fwa CD4+ TCR-Tg specific for male A1M Tg(TcraA1,TcrbA1)1Alm antigen DbY peptide B6.Cg-Ptprca Pepcb CD4+ TCR-Tg specific for LCMV SMARTA Tg(TcrLCMV)1Aox/PpmJ peptide B6.129S2-Tcratm1Mom/J Animals lack expression of TCR TCRαβ-KO alpha and beta chain which prevents T cell development (B10.A-Cd45a(Ly5a)/NAi N5 Animals lack expression of CD3ε CD3ε-KO which prevents T cell development B10.A, mPCC (Tg), CD3e-/- Animals lack expression of CD3ε PCC-CD3ε-KO which prevents T cell development. Animals also have transgenic expression of membrane localized pigeon cytochrome C under the MHC class I promoter and Ig enhancer. B10.A, mPCC (Tg) Animals have single allele of PCC+ CD3ε and transgenic expression of membrane localized pigeon cytochrome C under the MHC class I promoter and Ig enhancer

Primary lymphocyte isolation

Mice were euthanized by cervical dislocation and spleens and/or lymph nodes

(inguinal, axillary, brachial, superficial cervical, lumbar, sacral, mesenteric, and pancreatic) – see figure legends for whether splenocytes were included in the experiment

– were collected into sterile buffer (1xPBS supplemented with 5% FBS and 1x antimycotic/antibiotic, “crushing buffer”). A single cell suspension was prepare by mashing tissues through sterilized 100 M nylon mesh.

Magnetic T cell selection (MCS)

To ensure sufficient cell numbers following enrichment, most experiments pool single cell suspensions from 2-4 littermates, as indicated in individual figure legends. Cell suspensions were first Cells were applied to Ficoll-paque Plus (GE Healthcare) and the mononuclear cell layer was collected, diluted with fresh buffer and pelleted by centrifugation. To enrich for T cells using a negative selection strategy, cells were incubated on ice for 15 min with unconjugated or biotinylated primary antibodies

(depending on reagent availability, see Table 2.2), washed with buffer, and incubated with a mixture of anti-mouse and anti-rat coated Dynabeads (Invitrogen) or MyOne Streptavidin

T1 DynaBeads (Invitrogen) for 20 min at 4C. This suspension was placed against a separation magnet (Dynal) and the magnetic-particle-free supernatant containing the T cells pipetted out.

For some experiments (as indicated in figure legends), cells underwent a second magnetic negative selection using MACS Anti-Biotin MicroBeads protocol (Miltenyi

Biotec). Briefly, cells were incubated on ice for 10 min with the biotinylated antibody panel in Table 2.2. Primary antibodies were washed away before incubation with anti-

biotin MicroBeads (Miltenyi Biotec; 20 µL/107 cells) for 15 min at 4°C. After an additional wash with buffer, cells were separated using the autoMACS Pro (Miltenyi

Biotec) “Depletes” protocol.

Table 2.2. Antibodies used for magnetic T cell selection.

Concentration Antibody Conjugate Clone (μg/mL) Source Cat. # NK1.1 - PK136 6.25 purified in-house CD11b - M1/70 10.0 BD Pharmingen 553308 MHC-II (I-A/I-E) - M5/114 5.0 eBioscience 14-5321-85 B220 - RA3-6B2 5.0 eBioscience 14-0452-86 NK1.1 biotin PK136 5.0 BioLegend 108704 CD11b biotin M1/70 10.0 BioLegend 101204 MHC-II (I-A/I-E) biotin M5/114 5.0 BioLegend 107604 B220 biotin RA3-6B2 5.0 BioLegend 103204 CD8 biotin 53-6.7 5.0 BioLegend 100704

T cell drug pre-treatment scheme

Either MCS-enriched polyclonal T cells or TCR-Tg lymph node single cell suspensions (as indicated in figure legends) in TCM were plated on 96-well plates and treated with the indicated signaling inducer or receptor agonist for 15 or 30 min at 37C.

After centrifugation (1500 rpm x 5 min), media was replaced with fresh warmed TCM.

Cells were then supplemented with APCs and activating peptide or transferred to an anti-

CD3 coated 96-well plate and supplemented with anti-CD28 (see next method). Cells were maintained at 37C until prepared for the given assay, typically 24 or 72 hr later.

Ex vivo T cell activation and culture

Primary T cells were activated and cultured ex vivo in many experimental contexts; however, in all cases, either an antibody-mediated or APC/antigen activation strategy was used with a narrow range of technical variation. For antibody-mediated activation, plate- bound anti-CD3 (clone: 145-2C11, BD Pharmingen #553058) was prepared by adding

100 L of 1 or 3 g/mL antibody solution to a flat-bottom, 96-well plate. After 2 hr of incubation at 37C, plates were removed and washed twice with PBS. T cells in TCM (50-

100x103 cells/well) containing 2 g/mL anti-CD28 (clone: 37.51, purified in-house) were transferred to the coated plate. In these experiments, no stimulation controls were always included by transferring an identical number of cells in normal TCM to uncoated wells on the plate (i.e. culture without anti-CD3 or anti-CD28).

For APC/antigen activation, 20-100x103 TCR-Tg T cells were cultured in round- bottom 96-well plates in TCM with splenocytes from a compatible T cell depleted mouse line (CD3-KO or TCR-KO) at a 1:5 ratio. Antigenic peptide was added at 10 nM to 3

M (per experiment). In these experiments, no stimulation controls were prepared by not transferring antigenic peptide to control wells (typically labeled “0 nM” or “0 M”).

Ex vivo T cell differentiation

T cells were stimulated using the antibody-mediated activation scheme in the presence of a TH17 skewing cocktail (10 ng/mL IL-1, 30 ng/mL IL-6, 20 ng/mL IL-23,

10 g/mL anti-IFN, 10 g/mL anti-IL-4), a TH22 skewing cocktail (10 ng/mL IL-1, 30 ng/mL IL-6, 20 ng/mL IL-23, 400 nM FICZ, 10M galunisertib, 10 g/mL anti-IFN, 10

g/mL anti-IL-4) or no skewing control. In one experiment, galunisertib was left out of the

TH22 skewing cocktail (indicated in figure legend), but the remaining components were the same. Depending on the experiment, cultures were then supplemented with 100 nM

VIP (Tocris), 10 U/mL IL-2 (Peprotech), or both. After either 3 or 4 days in culture, cells received fresh media and were reactivated using 50 ng/mL PMA and 1 g/mL ionomycin for 5 hours. After 1 hour, brefeldin A was added to culture. Cells were then prepared for flow cytometry. (A) Gating strategy for identifying CD4+ T cells. (B) IL-22 production by

CD4+ T cells following treatment with or without VIP and the indicated skewing condition.

Proliferation dye labeling

T cells were labeled with 2μM carboxyfluorescein succinmidyl ester (CFSE,

Molecular Probes) or Cell Trace Violet (CTV, Thermo) in PBS supplemented with 0.5%

FCS in 37°C water bath for 12 minutes. The cells were washed with neat FCS to quench labeling and then resuspended in crushing buffer or TCM dependent upon assay.

Flow cytometry staining

Fc-receptor blocking was performed for 10-15 minutes at 4°C in PBS supplemented with 2% FCS, 0.01% azide, 1% mouse serum, 1% hamster serum, and 1% rat serum

(FcBlock). Surface staining was performed for 15 minutes at 4°C in PBS supplemented with 2% FCS, 0.01% azide, and 0.02% EDTA (FACS buffer) using the antibodies provided in Table 2.3. Cells were washed once with FACS buffer. Cells were then analyzed on one of the following flow cytometers: LSR II (BD), Canto II (BC), LSRFortessa (BD), or

Aurora (Cytek). All flow cytometry data analysis was performed using FlowJo v10.4.

Table 2.3. Antibodies used for flow cytometric analysis.

Antibody Fluorochrome Manufacturer Catalog # Clone 7AAD BioLegend 420403 B220 eFlour450 eBioscience 48-0452-82 RA3.6B2 B220 FITC Invitrogen 11-0452-82 RA3-6B2 c-kit PE-Cy7 BD Pharmingen 561681 2B8 c-kit PE BioLegend 105807 2B8 CD11b eFluor450 eBioscience 48-0112-82 M1/70 CD11c eFluor450 eBioscience 48-0114-82 N418 CD25 eFluor450 eBioscience 48-0251-82 PC61 CD25 APC BioLegend 102012 PC61 CD25 PE eBioscience 12-0251-82 PC61 CD3 APC-Cy7 BD Pharmingen 557596 145-2C11 CD3 APC BioLegend 100235 17A2 CD4 BV605 BioLegend 100451 GK1.5 CD4 APC-Cy7 BD Pharmingen 552051 GK1.5 CD4 eFluor450 eBioscience 48-0041-82 GK1.5 CD4 PE-Cy7 eBioscience 25-0041-82 GK1.5 CD4 BV750 BD Pharmingen 747102 GK1.5 CD4 PE eBioscience 12-0042-83 RM4-5 CD4 APC BioLegend 100515 RM4-5 CD44 APC eBioscience 17-0441-83 IM7 CD44 PE-Cy7 BioLegend 103030 IM7 CD44 PE eBioscience 12-0441-83 IM7 CD44 PerCPCy5.5 BioLegend 103032 IM7 CD44 BV421 BioLegend 103039 IM7 CD45.1 APC-Cy7 BioLegend 110716 A20 CD45.1 PE-Cy7 eBioscience 25-0453-82 A20 CD62L AF700 BioLegend 104426 MEL-14 CD69 PE Invitrogen 12-0691-82 H1.2F3 CD8 FITC eBioscience 11-0083-85 eBioH35-17.2 CD8 PE-Cy7 eBioscience 48-0081-82 eBioH35-17.2 EGFR AF700 CST 42675 D38B1 ERK AF647 CST 5376 137F5 I-Ab AF647 BioLegend 115310 KH74 IFN FITC BioLegend 505806 XMG1.2 IL-17A eFluor450 eBioscience 48-7177-82 eBio17B7 IL-22 APC Invitrogen 17-7222-82 IL22JOP LiveDead Far red Invitrogen L34973 LiveDead Near IR Invitrogen L34975

LiveDead Violet Invitrogen L34964 MHC II eFluor450 eBioscience 48-5321-82 M5/114.15.2 Jackson Mouse IgG ImmunoResearch 115-005-044 NK1.1 eFluor450 eBioscience 48-5941-82 PK136 pAkt PE-Vio770 Miltenyi Biotec 130-105-343 REA359 pCREB PE BD Pharmingen 558436 J151-21 pERK AF488 CST D13.14.4E pERK AF647 CST D13.14.4E pERK AF488 CST 4780 197G2 pERK PerCP-Cy5.5 BioLegend 369511 6B8B69 pp38 PE-CF594 BD Pharmingen 563569 36/p38 pS6 APC Invitrogen 12-9007-42 cupk43k pSLP-76 AF647 BD Biosciences 558438 J141.668.36.58 pSLP-76 PE BD Biosciences 558437 J141.668.36.58 pZap70 (Y292) PE BD Biosciences 558510 J34-602 pZap70 (Y319) PE-Cy7 BD Biosciences 561458 17A/P-ZAP70 Jackson Rabbit IgG FITC ImmunoResearch 111-095-003 RORt APC eBioscience 17-6988-80 AFJKS-9 Sca-1 APC-Cy7 BD Pharmingen 560654 D7 T-bet PE-Cy7 eBioscience 25-5825-82 eBio4B10 T-bet PerCP-Cy5.5 eBioscience 45-5825-82 eBio4B10 TCRβ PE-Cy7 BioLegend 109222 H57-597 TCRβ FITC BD Pharmingen 553171 H57-597 Vβ3 FITC BD Pharmingen 553208 KJ25 Vβ3 PE BD Pharmingen 553209 KJ25 Zap70 FITC eBioscience 11-6695-82 1E7.2

Intracellular cytokine staining

After either 3 or 4 days in culture, ex vivo differentiated T cells received fresh media and were reactivated using 50 ng/mL PMA and 1 g/mL ionomycin for 5 hours. After 1 hour, brefeldin A was added to culture. Surface staining was performed as described above.

Intracellular staining was performed with eBioscience Fixation/Permeabilization Kit. Cells were resuspended in 1X eBioscience Fixation/Permeabilization buffer and incubated overnight at 4°C. Intracellular staining was performed overnight at 4°C in 1X eBioscience

Permeabilization Buffer using the antibodies listed in Table 2.3. Stained cells were washed twice with FACS buffer and then analyzed using one of the above listed flow cytometers.

Transcription factor staining

Surface staining was performed as described above. Transcription factor staining was performed using the True-Nuclear Transcription Factor Buffer kit (BioLegend). Cells were resuspended in True-Nuclear 1X Fix Concentrate and incubated at room temperature for 1 hr. Cells were washed twice with True-Nuclear 1X Perm Buffer before staining with the antibodies listed in Table 2.3 in True-Nuclear 1X Perm Buffer for 45 min. Cells were washed once with FACS buffer and analyzed using one of the above listed flow cytometers

Cytokine capture assay

LN and splenocyte single cell suspension recovered from mice with in vivo activated T cells (see below) were cultured TCM with 10 M PCC protein overnight at

37C to reactivate 5C.C7 T cells. IFN capture assay was performed using mouse IFN secretion assay-detection kit (Miltenyi Biotec). Briefly, cells were transferred to 50 mL conical tubes and resuspended in 90 L cold capture buffer (PBS supplemented with 0.5%

BSA and 2 mM EDTA). 10 L of IFN capture reagent was added and incubated on ice

for 5 min. 50 mL of warm capture media (RPMI-1640 (Gibco) supplemented with 5% FCS

(Gemini)) was added and cells were incubated at 37C for 45 min, with rotation every 5 min. Cells were washed with 1 mL capture buffer and prepared for flow cytometry. Surface staining was performed as above, except anti-IFN-PE reagent was added 1:10 during the antibody incubation period.

Phospho-flow cytometry

T cell stimulation for phospho-flow cytometry included TCR-crosslinking using biotinylated antibodies and streptavidin, drug treatment, or both. Cell samples to be activated by TCR-crosslinking were stained in PBS with antibodies listed in Table 2.4 (as indicated by individual figure legends), washed, and resuspended in PBS. Cells were added to FACS tubes containing 5 g/mL streptavidin in PBS, pre-warmed to 37C in a water bath. Cell samples to be treated with drugs were resuspended in PBS and added to pre- warmed FACS tubes containing the intended drug in PBS. Cell samples to be drug treated and antibody-activated were stained with biotinylated antibodies as above and then added to pre-warmed FACS tubes containing streptavidin and the intended drug. Once cells are added to the tube, they were vortexed briefly and returned to the water bath. After the intended stimulation time, cells were fixed with 1.5% PFA. Cell membranes were permeabilized by incubation in 2 mL of ice cold 100% MeOH on ice for 30 min. After centrifugation (2000 rpm x 10 min), cells were rehydrated by incubation in 2 mL FACS buffer for 30 min on ice. Cells were then stained overnight at 4C with antibodies listed in

Table 2.3. Cells were washed twice with FACS buffer and analyzed using one of the above listed flow cytometers.

Table 2.4. Antibodies used for TCR-crosslinking in phospho-flow cytometry studies.

Concentration Antibody Conjugate Clone (µg/mL) Source Cat. Number CD3 biotin 17A2 10.0 BioLegend 100244 CD4 biotin GK1.5 10.0 BioLegend 100404 CD8 biotin 53-6.7 10.0 BD Pharmingen 553029 CD28 biotin 37.51 10.0 BD Pharmingen 553296

FACS-purification T cells enriched using by one round of MCS were stained using the antibodies in

Table 2.5 and 7AAD (BioLegend). Sorting was performed using the FACS Aria II (BD).

The gating strategy used is described in Chapter 5.

Table 2.5. Antibodies used for FACS-purification.

Concentration Antibody Fluorophore Clone (µg/mL) Source Cat. Number CD4 APC-Cy7 GK1.5 2.0 BD Pharmingen 552051 CD8b FITC eBioH35-17.2 2.0 eBioscience 11-0083-85 TCR β Chain PE-Cy7 H57-597 2.0 BD Pharmingen 560729 CD25 PE PC61 2.0 eBioscience 12-0251-82 CD44 APC IM7 2.0 eBioscience 17-0441-82 NK1.1 eFlour 450 PK136 1.0 eBioscience 48-5941-82 CD11b eFlour 450 M1/70 0.6 eBioscience 48-0112-82 CD11c eFlour 450 N418 0.6 eBioscience 48-0114-82 B220 eFlour 450 RA3-6B2 0.6 eBioscience 48-0452-82

Thymidine incorporation assay

“Pre-activated” 5C.C7 cells were generated using ex vivo culture with APCs from

CD3-KO mice and 1 M MCC for 3 d. Cells were then rested in the absence of antigenic stimuli but IL-2 supplementation for 8 d. Pre-activated and freshly isolated naïve 5C.C7 were activated using anti-CD3/28 or with APC (CD3-KO) and 1 M MCC peptide in the presence of indicated dose of insulin (combination of bovine and human in a 1:1 mix).

After 48 hr in culture, 1 Ci of 3H-thymidine was added to each well. After an additional

24 hr, cells were harvested onto a nylon wool filter using a Tomtec cell harvester.

Scintillation count was then determined using a Wallac Betalux betacounter.

RNA extraction and RT-PCR

To isolate RNA, T cells (either FACS sorted or MCS purified, as required) were spun down, the pellets loosened and resuspended in TRIZOL (Invitrogen). These cell lysates were further homogenized by applying to QIAshredder columns (Qiagen). The filtrate from the column was combined with 100 µL of chloroform and incubated at room temperature for 2 min. After centrifugation (17000g for 15 min), the aqueous layer was collected and combined with an equal volume of 70% ethanol. The RNA mixture was then purified using the Qiagen RNeasy Mini kit. Briefly, the RNA mixture was applied to a

Qiagen RNeasy, washed once each with 700 µL of RPE buffer and 500 µL 80% ethanol, and eluted with 50 µL PCR-grade water with 20 U of SUPERase RNase inhibitor added

(Invitrogen). RNA yield and quality were determined using a NanoVue Plus (GE). For reverse transcription, ~250 ng of RNA per sample was converted to cDNA using the

Qiagen RT2 First Strand kit, including the genomic DNA elimination step. For NR expression validation (Chapter 5), cDNA was applied to the Qiagen RT2 Profiler PCR

Array Mouse Neurotransmitter Receptors kit, performed per manufacturer’s instructions and normalized to house-keeping genes (Actb, B2m, Gapdh, Gusb, Hsp90ab1; geometric mean Ct for all 5 genes). Vipr1 expression (Chapter 7) was measured using the following primers: Vipr1 5’-ATGGATGAGCAGCAACAGACT-3’ and 5’-

GGCCATGACGCAATACTGG-3’; Actb 5’- GGAGCACCCTGTGCTGCTCACCGAGG

3’ and 5’- ATCTACGAGGGCTATGCTCTCCC -3’. The expression of Vipr1 was normalized to Actb. All qPCR was performed using ABI 7900HT.

NanoString RNA count analysis

To take an unbiased approach to assess NR expression, we generated a comprehensive list of NR genes, adapting from Iwama and Gojobori 2002141, and including additional genes identified through further literature review. Working with representatives from NanoString, this gene list was developed into a set of 179 NR targeting probes and 6 probes targeting housekeeping genes (see Table 5.6 for gene list and probe details). Our probeset (available for purchase under the label CodeSet NR-Mm-20113) arrived as part of the NanoString nCounter Gene Expression Assay kit, which includes all necessary reagents. To prepare a NanoString reaction, briefly, the provided reporter probes were diluted in hybridization buffer and added to the provided 12 well plate. ~100 ng of isolated

RNA per sample was applied to the probes and heated to 65°C. The capture probes were then added immediately before beginning hybridization for 12 hours. Resulting expression data were analyzed using nSolver 4.0 software (NanoString). Background thresholding was applied using the mean of the negative control wells.

ImmGen microarray analysis

Microarray data sets corresponding to mature, peripheral αβ T cell populations collected by the Immunological Genome Project (www.immgen.org) were downloaded from the NCBI Gene Expression Omnibus (GEO) (accession number GSE15907).

Standardized cell isolation protocols can be found on the ImmGen website. Briefly, male,

5-week-old C57BL/6J mice were euthanized and the indicated tissues from 3 or more mice were pooled to prepare a single cell suspension. Cells underwent two rounds of selection, either 2 rounds of FACS purification or one round of magnetic pre-purification using Dynal beads followed by FACS purification, to achieve >99% purity of the indicated T cell population. Cells were directly sorted into Trizol before RNA isolation. Microarray analysis was performed using Affymetrix Mouse Gene1.0ST array.

Microarray datasets were analyzed in Microsoft Excel using the BRB-ArrayTools v4.5 package. Individual microarray data files were imported and collated using the

“Affymetrix Gene ST Array Importer” function, applying quantile normalization and removing spot and gene filters. Biological replicates were averaged and thresholding was applied to the NR genes in the above gene list, modified from the ImmGen quality control analysis (Bezman et al. 2012, Supplementary Note 2). Briefly, after quantile normalization was applied to the >650 arrays collected, expression values >120 were determined to correspond with a ≥95% probability of true expression while values <47 have a ≥95% probability of being silent. In our hands, quantile normalization using the subset of 78 microarrays for peripheral ab T cells yielded mean expression values approximately 20% lower than that of the ImmGen normalization protocol. As such, we used respective threshold values of 100 and 40 for our analysis. Parallel analyses using both threshold values were performed and presented as indicated.

GEO microarray analysis

Microarray data sets were downloaded from GEO and analyzed in Microsoft Excel as above. The GEO accession numbers and citations for the datasets analyzed are as follows: GSE89037142, GSE41870143,144, GSE39152145, and GSE47045146. Details regarding experimental methodology for all studies are included in the Chapter 5.2. Data files were imported and collated as ImmGen datasets (above). A gene subset defined by the above NR gene list was then analyzed using either the “Class comparison between groups of arrays” function, setting a significance threshold of 0.01, or “Time course analysis,” setting a false discovery rate of 0.05. Genes identified by these analyses were then further tested for significantly different expression between groups using a Student’s

T-test.

RNA-Seq analysis

Collated FPKM or TPM text files were downloaded from GEO and analyzed using

Microsoft Excel. For the minority of data series that were available as individual files, the web-based Galaxy Project (www.galaxyproject.org) tool was used to collate the data set.

The above NR gene set was used to filter the data and thresholding was applied to eliminate background expression. The GEO accession numbers and citations for the datasets analyzed are as follows: GSE111143147, GSE107281148, GSE78247149, GSE85294150,

GSE97863151, and GSE112706 (unpublished).

Primary T cell, cell line, and tissue lysate preparation

Primary T cell samples, either purified or following drug treatment, were flash frozen using liquid N2 and thawed before the addition of 2x cell lysis buffer (2% NP-40,

20 mM Tris-HCl, 300 mM NaCl, 4 mM EDTA, 100 mM β-glycerophosphate, 4 mM

sodium orthovanidate, 20 mM sodium fluoride, 2 mM dithiotheitol, 2 mM PMSF, 1x cOmplete protease inhibitor cocktail (Sigma), 1:1000 phosphatase inhibitor cocktail set IV

(Calbiochem)). Confluent P6 Neuro-2a and P38 SH-SY5Y cells were trypsinized, counted, flash frozen, and thawed before the addition of cell lysis buffer (as above).

Brain, lung, and small intestine were collected from unperfused C57BL/6 mice euthanized by cervical dislocation. Small intestine was cut to 5 cm length and flushed with normal saline. Tissues were then weighed and incubated with tissue lysis buffer (50 mM

Tris-HcL, 150 mM NaCl, 0.25% SDS, 0.25% sodium deoxylcholate, 1 mM EDTA, 1x cOmplete protease inhibitor cocktail (Sigma)) for 10 min before bead homogenization using a Roche MagNA Lyser at speed 6000 for 60s. Homogenate underwent centrifugation

(10,000g x 10 min) and the supernatant was collected.

Western blot analysis

Cell and tissue lysates were combined with 2x Laemmli sample buffer (Bio-Rad), boiled for 10 min, and rested on ice for 3 min before loading onto NuPage 4-12% Bis-Tris

1.5 mm protein gel (Invitrogen) or hand-poured 10% polyacrylamide gels using the TGX

FastCast Acrylamide kit (Bio-Rad). Additionally, SeeBlue Plus2 Prestained Standard

(Invitrogen) was loaded as a size marker. Gels were transferred to 0.45 µm pore nitrocellulose membrane (Novex) using a TransBlot Turbo (Bio-Rad, 25 V for 20 min).

Blots were blocked using 5% Blotto (SantaCruz) for 1 hr at room temperature. Blots were then incubated with primary antibody (Table 2.6) for either 1 hr at room temperature or overnight at 4°C. Following washing, blots were incubated with secondary antibody (Table

2.7) for 1 hr at room temperature. Blots were then incubated with SuperSignal West Dura extended duration substrate or SuperSignal West Femto maximum sensitivity substrate

(Thermo Scientific) and visualized with ChemiDoc MP Imaging System (Bio-Rad). When a blot was stripped for re-blotting, it was incubated in stripping buffer (62.5 mM Tris-HCl,

2% SDS, 100 mM -mercaptoethanol in H2O) at 50C with gentle rotation for 30 min. It was then washed 6 times before restarting protocol at the blocking step. Band densitometry was measured using ImageJ software.

Table 2.6. Primary antibodies used for western blot analysis. When in italics, the gene name for the target protein is listed.

Concentration (μg/mL or Antibody target Host Conjugate dilution) Source Cat. Number Adora2a mouse - 1.0 SantaCruz sc-32261 Adrb2 rabbit - 0.38 Proteintech 13096-1-AP Gabrr2 rabbit - 4.0 Alomone AGA-007 Gria3 rabbit biotin 3.3 Bioss bs-1799R-Biotin Lpar6 rabbit - 4.1 Alomone ALR-036 Sigmar1 mouse - 1.0 SantaCruz sc-137075 Vipr1 goat - 2.0 Abcam ab123517 -actin mouse - 1.0 Abcam ab8226 pERK rabbit - 1:1000 CST 9101S ERK2 rabbit - 1:2000 SantaCruz sc-153

Table 2.7. Secondary antibodies used for western blot analysis.

Target Host Conjugate Dilution Source Cat. Number Mouse IgG (H+L) Goat HRP 1:5000 Bio-Rad 170-6516 Rabbit IgG (H+L) Goat HRP 1:5000 Bio-Rad 170-6515 Goat IgG (H+L) Rabbit HRP 1:5000 Bio-Rad 170-6520 Streptavidin - HRP 1:1000 BD Pharmingen 554066

Immunofluorescence imaging

Cells were stimulated and prepared for intracellular antibody staining as described in the phospho-flow cytometry section. Cells were stained overnight at 4C with primary antibodies: AF488-anti-pERK (CST #13214) and mouse-anti-GM130 (BD #610822). The cells were washed the next day with PBS and stained with Cy3-donkey-anti-mouse IgG secondary antibodies (Jackson ImmunoResearch #715-166-020). After additional washing, cells were adhered to silanated glass slides using cytocentrifugation (900 g x 3 min). A drop of ProLong Gold Antifade Mountant with DAPI (Invitrogen #P36935) was added and overlaid with a class coverslip. All images were taken using the 100x objective.

Adoptive transfer and in vivo T cell activation

TCR-Tg T cells were adoptively transferred in several experimental contexts in order to study T cell activation in vivo. Individual figure legends will specify the TCR-Tg mouse used, the number of T cells transferred, prior activation or treatment status, the mouse strain transferred to, the dose of LPS used to activate cells, and the time delay between transfer and LPS/peptide injection. This section will describe the common protocol elements.

T cells were resuspended in PBS at a concentration of the transferred cell number

(105 to 106) per 100 L. Cells were transferred to host mice via retro-orbital, intravenous

(i.v.) injection. Transferred T cells were activated in vivo by intraperitoneal (i.p.) injection of LPS (0.5-2.5 g) and agonistic peptide (30 g) in 100 L PBS immediately, 6 hr, or 24 hr after T cell transfer.

ELISA

Supernatants were collected form ex vivo cultures and cytokine concentrations were measured using commercially available ELISA kits: IFN (eBioscience mouse IFN

ELISA Ready-Set-Go! Kit), IL-17A (eBioscience mouse IL-17A ELISA Ready-Set-Go!

Kit), IL-4 (R&D Mouse IL-4 Quantikine ELISA kit).

Vipr1 retroviral overexpression

Vipr1 cDNA was amplified in two, overlapping segments from B6 FACS-purified

CD4+CD44lo T cell cDNA. Primers used: Vipr1 (5’ segment): 5’-

AGGAATTGAGCGGCCGCCACCATGCGCCCTCCGAG-3’ and 5’-

GGCCATGACGCAATACTGG-3’; Vipr1 (3’ segment): 5’-

ATGGATGAGCAGCAACAGACT-3’ and 5’-

CGCTAGCGAGGCCTCCTAGGACCAGGGAGACCTCCGCTTGGAAG-3’. The Vipr1 gene segments were cloned into the pQ2AB retroviral vector (Addgene, #124887) using an In-Fusion reaction (Takara). Vipr1 and vector control retroviral plasmids were used to transfect Phoenix-GP cells with resulting retroviral supernatants collected and concentrated. Retroviruses were used to transduce TCR-Tg T cells by spinoculation.

Briefly, T cells were activated ex vivo using the APC/antigen activation scheme described above. After 30 hours of activation, cells were incubated with retroviral expression vectors expressing Vipr1 (pV-RV, 1:10 final dilution) or empty vector (pQ-RV, 1:100 final dilution) and centrifuged (2000 rpm x 2 hr, 32C). A second round of transduction was repeated 10 hours later. 24 hr later, 1 g/mL puromycin was added to culture to select for transduced cells.

Bone marrow chimeric mouse preparation

Bone marrow (BM) chimeras using pV-RV and pQ-RV transduced BM were established as follows. B6 BM was collected by flushing the femur and tibia with crushing buffer. A single cell suspension was generated by passing cells through 100 M nylon mesh. BM was cultured in 6-well plates in 3 mL BMM for 24 hr. pQ-RV (1:200 final dilution) and pV-RV (1:20 final dilution) were transduced using spinoculation as above. A second round of transduction was performed 10 hours later followed by puromycin selection as above, 24 hr later. After an additional 24 hr, 100,000 BM cells were each transferred to irradiated (600 rad) TCR-KO mice by retro-orbital, i.v. injection. Mice were maintained on an antibiotic regimen of sulfamethoxazole/trimethoprim for 2 weeks following transfer. Cells were collected from mice ~10 weeks later for analysis.

Statistical analysis

Statistical analysis of ImmGen data was not possible as only mean signal intensity data for 2-4 samples is available, without SD or SEM data. For remaining analyses, the two-tailed unpaired Student’s T-test was applied using GraphPad Prism software.

Chapter 3: Pre-stimulating specific subsets of the TCR signaling pathway alters T

cell activation

3.1. Introduction

3.1.1. Approach to studying testing the signal 0 hypothesis

A central tenet of the signal 0 hypothesis that we investigated in this thesis is that triggering of individual modules of the TCR signaling pathway (see section 1.2.5, Fig. 1.1) should differentially bias subsequent ‘full’ T cell activation due to all signaling components being efficiently engaged (Fig. 1.13). In the context of an in vivo model, this is quite challenging to investigate. As discussed in the introduction, signals from cytokines, metabolites, and multiple environmental cues are available to T cells all the time in different tissue sites. So, although the goal is to understand how these very cues may impact on overall T cell function, examining them using purely in vivo models would be challenging. On the other hand, using in vitro experiments offers a minimal or reductionistic system that would allow a clearer dissection of how prior activation of individual TCR signaling modules bias subsequent activation.

The goal for this chapter was then to select reagents that would specifically activate individual pathways of TCR signaling such that we could target each one separately with drug treatment before activating the T cell with antibody or antigenic peptide and examine changes in markers of activation, proliferation, and cytokine production. Ideally, the terminal kinases of each pathway, calcineurin, IKK, ERK, JNK, and p38, could be targeted specifically such that the effects of each module could be cleanly isolated.

A problem with this approach, is that while there are several specific inhibitors for these and other enzymes within these modules, virtually no specific activators have been

identified. Interestingly, this is a challenge for the study of most biological systems beyond

T cells as enzymes appear to be uniquely difficult to isolate in this way and new methods for identifying specific inducers are being developed152. Instead of an activator based approach to isolate signaling modalities, we considered a reverse strategy in which combinations of specific pathway inhibitors could be paired with TCR ligation, allowing only a selected pathway to signal – i.e. combining MAPK and NFB inhibition with TCR stimulation to allow for isolated Ca2+ signaling. We opted against this approach two reasons: 1) while small molecule inhibitors are more specific than activators, they still have off-target effects that would confound investigation, and 2) although we could potentially block the TCR-singling modalities that lead to changes in gene transcription, completely eliminating all of the pathways evoked by TCR stimulation that may also confound results is challenging. Thus we chose to utilize the few small molecules that have been used in varied contexts to stimulate intracellular signaling cascades. Those relevant to the work in this thesis are discussed below.

Ceramide

Ceramide is a sphingolipid, generated by metabolism of either sphingosine-1- phosphate or sphingomyelin, which is found in cell membranes and is thought, predominantly, to provide structural support to lipid bilayers; however, it was discover that ceramide also acts as a signaling intermediate as well153,154. Because the endogenous form of ceramide is typically membrane embedded and difficult to probe, truncated, cell- permeable analogs including C2 and C6 have been used to examine its signaling properties.

Through mechanisms that are unclear, ceramide has been reported to activate ERK155,156 as well as protein phosphatase 2A (PP2A)157, but has been reported to modulate Akt and

NFB signaling in certain cell types153. Little to no investigation of ceramide signaling in

T cells has been pursued to date, but incubation of T cells with ceramide during activation decreases proliferation158.

PMA and ionomycin

While PMA is most commonly used in combination with ionomycin to activate T cells in vitro, their independent signaling roles have been studied. PMA is a membrane permeable analog of DAG and thus induces the activation of PKC in T cells, eliciting

NFB159 and ERK activation159,160. PMA has also been shown to activate guanylate cyclase via a Ca2+-dependent process161. Ionomycin is a Ca2+ ionophore acting to directly allow

Ca2+ flux into the cytosol, inducing calmodulin-dependent signaling pathways; however, treatment of T cells with ionomycin also triggers PKC activation162. Thus, in combination,

PMA and ionomycin act synergistically to activate the major modules of TCR signaling.

In the absence highly specific signaling activators or even physiological receptor ligands that elicit single pathway signaling, we elected to model signal 0 using PMA pre- treatment, as it has been well studied. Thus, we can simulate a receptor-ligand interaction that selectively yields NFB and AP-1 activation prior to full engagement of the TCR signaling pathways.

The next set of questions relates to choosing the methods and readouts. We decided to measure events close to the point of T cell activation initially, to get a better window into the immediate signaling and proximal functional consequences of signal 0 events.

Future experiments to measure cytokine production would be necessary to fully explore how signal 3 is impacted. Importantly, cell culture assays utilized highly enriched populations T cell, using a pan-T cell enrichment strategy, in order to ensure we are

capturing a direct effect on T cell signaling rather than bystander effects from other cells in culture. We also relied on phospho-flow cytometric approaches to study intracellular signaling as it allows us to consider single cell effects and heterogeneity within the population.

3.1.2. Chapter summary

In this chapter, we investigated for the first time whether activation of a subset of

TCR signaling pathways can alter the signaling events triggered by antigen recognition and the impact this has on the T cell activation process. We modeled a signal 0 event using brief exposure to PMA prior to activation, finding that early activation markers including

CD69 and CD25 as well as T cell proliferation were enhanced in this context. We also revealed unique ERK phosphorylation kinetics following pre-treatment, suggesting that distinct pathway regulation is induced. Further, although we find that PMA pre-treatment

(pre-Tx) does not alter in vivo proliferation or differentiation, cell recovery was decreased suggesting changes to T cell survival or migration. Lastly, we compare signal 0 effects between naïve and activated T cells showing that unique signal integration occurs depending on the prior activation state. In addition to these findings, this work allowed for the optimization of many assays used throughout the remaining chapters in which physiological signal 0 stimuli are considered.

3.2. Results

3.2.1. Transient exposure to PMA modulates subsequent T cell activation

To model the effects of selective TCR-pathway engagement on subsequent activation, we focused on PMA-induced signaling. Although PMA and ionomycin are typically used in combination, as discussed above, we chose to use PMA alone as a model

for partial stimulation of TCR signaling pathways. To assess the functional impact of brief exposure to PMA prior to full, TCR-mediated activation, we first considered the expression of the activation markers CD69 and CD44 in this paradigm. We used the 5C.C7 mouse model, in which TCR-Tg CD4+ T cells specific to pigeon (PCC) and moth cytochrome C

(MCC) peptides, in order to examine a uniform population of T cells under multiple activation conditions. 5C.C7 T cells were treated with PMA for 15 min before washing the cells with fresh media and activated with anti-CD3ε and anti-CD28 antibodies. After 24 h in culture, both CD69 and CD44 showed increased expression, both by percent of cells positive for the markers and at an individual cell level measured by geometric mean fluorescence intensity (gMFI), following PMA pre-Tx across all doses of anti-CD3ε used

(Fig. 3.1). Notably, CD69 and CD44 were induced by brief PMA exposure even without the addition of stimulating antibodies (Fig. 3.1 B-D).

C D

Figure 3.1. Transient PMA exposure enhances early antibody-mediated T cell activation. Lymph nodes were isolated from 5C.C7 mice, treated with 50 ng/mL PMA or PBS for 15 min, and then stimulated using the indicated dose of anti-CD3ε and 2 g/mL anti-CD28. 24 hours later, cells were analyzed by flow cytometry. (A) Gating strategy for analyzing live 5C.C7 T cells. (B) Representative histograms for CD69 (left) and CD44 (right) expression among 5C.C7 T cells following the respective pre-stimulation treatment. (C-D) Quantification of expression in respective tables (far right). Summary quantification of technical replicates for CD69 (C) and CD44 (D) expression by percentage of cells positive (upper) and gMFI (lower).

We next sought to examine the effects of PMA pre-Tx in the context of antigen- mediated activation of T cells. Accordingly, T cells were treated with PMA, washed with fresh media, and then added to splenocytes from PCC-CD3-KO mice (Fig. 3.2) or CD3-

KO mice (Fig. 3.3). Both sources of APCs were supplemented with 1 M MCC or no MCC as control. Consistent with effects of PMA on antibody-mediated T cell activation (Fig.

3.1), CD69 and CD44 showed increased expression by both %CD69+ and CD69+ gMFI under both antigen-mediated stimulation conditions (Fig. 3.2-3.3). While both APC sources showed similar CD69 and CD44 expression with the addition of MCC, T cells activated with PCC-expressing APCs showed strongly increased expression of CD69 and

CD44 compared to non-PCC-expressing APCs when PMA was added to culture. However, even when the PCC antigen was present in the PCC-CD3e-/- mice, CD69 was not expressed without the addition of PMA or MCC. This suggests that PCC expression in the

PCC-CD3- APCs only provided a minor, subthreshold level of stimulation that could be augmented by brief PMA pre-treatment of the T cells. In previous studies with these mice, it has been shown that the level of PCC expressed is relatively low for in vitro stimulation, but sufficient to activate naïve T cells in vivo140. Additionally, after 24 hr of stimulation,

CD4 was downregulated in a dose-responsive manner by PMA pre-Tx (Fig. 3.2B,E;

3.3B,E), consistent with prior literature163.

C D

Figure 3.2. Transient PMA exposure enhances early antigen-mediated T cell activation. Lymph nodes were isolated from 5C.C7 mice, treated with indicated dose of PMA or PBS for 15 min, and then stimulated using the indicated dose of MCC combined with PCC-CD3-εKO splenocytes. 24 hours later, cells were analyzed by flow cytometry. (A) Gating strategy for analyzing live 5C.C7 T cells. (B) Representative histograms for CD69 (left), CD44 (middle), and CD4 (right) expression among 5C.C7 T cells following the indicated pre-stimulation treatment and activation condition. (C-E) Summary quantification of technical replicates for CD69 (C), CD44 (D), and CD4 (E) expression by percentage of cells positive (upper) and gMFI (lower).

B C

Figure 3.3. Effects of transient PMA exposure on early antigen-mediated T cell activation are replicated. Lymph nodes were isolated from 5C.C7 mice, treated with indicated dose of PMA or PBS for 15 min, and then stimulated using the indicated dose of MCC combined with CD3-εKO splenocytes. 24 hours later, cells were analyzed by flow cytometry. Cells gated as in Figure 3.2A. (A) Representative histograms for CD69 (left), CD44 (middle), and CD4 (right) expression among 5C.C7 T cells following the indicated pre- stimulation treatment and activation condition. (B-D) Summary quantification of technical replicates for CD69 (B), CD44 (C), and CD4 (D) expression by percentage of cells positive (upper) and gMFI (lower).

We next used this same antigen-mediated T cell activation scheme to determine the effects of PMA pre-Tx on other activation metrics, including proliferation, surface marker expression, and IL-2 production. 5C.C7 T cells were loaded with either CFSE (Fig. 3.4) or

CTV (Fig. 3.5), treated with PMA, and activated with PCC-CD3-KO (Fig. 3.4) or CD3-

KO APCs (Fig. 3.5) as before. After 72 hr of culture, expression of CD44 and CD25 were increased among T cells that were pre-treated with PMA compared to PBS, under both stimulation conditions (Fig. 3.4B-5; 3.5A-B). As in the earlier time point, PCC-expressing

APC induced stronger expression of these markers than non-PCC-expressing APC in the presence of PMA pre-Tx. Consistent with activation marker expression, PMA pre-Tx showed a dose-dependent increase in T cell proliferation as measured by CFSE or CTV dilution (Fig. 3.4D; 3.5C). Further, in the presence of PMA, T cells could be induced to proliferate when activated by PCC-expressing APC even in the absence of MCC supplementation (Fig. 3.4D). Consistent with the increased activation, T cells also showed a dose-dependent increase in the amount of IL-2 production they secreted, with PMA pre-

Tx, as measured by ELISA of recovered supernatant after 72 hr of culture (Fig. 3.4E).

Collectively, these experiments reveal that even brief induction of PKC by PMA prior to full, TCR-mediated stimulation, enhances T cell activation, as measured by several metrics.

B C

Figure 3.4. Transient PMA exposure enhances T cell proliferation and IL-2 production. Lymph nodes were isolated from 5C.C7 mice, loaded with CFSE, treated with indicated dose of PMA or PBS for 15 min, and then stimulated using 0 or 1 M MCC combined with PCC-CD3ε-KO splenocytes. 72 hours later, culture supernatants were collected and cells were analyzed by flow cytometry. (A) Gating strategy for analyzing live 5C.C7 T cells. (B) Representative histograms for CD44 (left) and CD25 (right) expression among 5C.C7 T cells following the indicated pre-stimulation treatment and activation condition. (C) Summary quantification of technical replicates for CD25 (upper) and CD44 (lower) expression by percentage of cells positive.

D E

Figure 3.4. Transient PMA exposure enhances T cell proliferation and IL-2 production (cont.). Lymph nodes were isolated from 5C.C7 mice, loaded with CFSE, treated with indicated dose of PMA or PBS for 15 min, and then stimulated using 0 or 1 M MCC combined with PCC-CD3ε-KO splenocytes. 72 hours later, culture supernatants were collected and cells were analyzed by flow cytometry. (D) Histograms of CFSE dilution among 5C.C7 cells with indicated pre-stimulation treatment and activation conditions. Gate indicates the undivided population. Adjacent histograms of the same color constitute technical replicates. (E) IL-2 concentration in culture supernatant as measured by ELISA. Concentration represented by measured optical density.

A B

Figure 3.5. Effects of transient PMA exposure on T cell proliferation are replicated. Lymph nodes were isolated from 5C.C7 mice, loaded with CFSE, treated with indicated dose of PMA or PBS for 15 min, and then stimulated using 0 or 1 M MCC combined with CD3ε-KO splenocytes. 72 hours later, cells were analyzed by flow cytometry. Cells gated as in Figure 3.4A. (A) Representative histograms for CD44 (left) and CD25 (right) expression among 5C.C7 T cells following the indicated pre-stimulation treatment and activation condition. (B) Summary quantification of technical replicates for CD25 (upper) and CD44 (lower) expression by percentage of cells positive. (C) Histograms of CFSE dilution among 5C.C7 cells with indicated pre-stimulation treatment and activation conditions. Gate indicates the undivided population. Adjacent histograms of the same color constitute technical replicates.

3.2.2. PMA induces unique ERK phosphorylation kinetics

Having established that a brief pre-exposure to PMA pre-Tx functionally increases

T cell activation, we next sought to understand the changes in signaling that underlie these effects. The assumption underlying these experiments (from the signal 0 framework) was that ‘tickling’ of the small arm of the TCR signaling network selectively augments that pathway to make the T cell more sensitive to subsequent stimulation. Noting that PMA treatment enhances CD69 expression (Fig. 3.1-3.3), we focused on the activity of ERK as it is both known to be induced by PKC signaling in T cells160,164,165 and that CD69 expression is dependent on ERK-mediated activation of AP-1101,102. As such, we used phospho-flow cytometry to examine the kinetics of ERK phosphorylation within PMA- treated 5C.C7 cells (Fig. 3.6-3.7). Within 5 min of PMA exposure, T cells showed strong phosphorylation of ERK (p44/42 MAPK T202/Y204) which began to diminish at around

90 min of treatment (Fig. 3.6B). Additionally, total ERK remains constant through 2 hr of

PMA treatment (Fig. 3.6C), indicating that the eventual decrease in measured pERK at >90 min. This is likely due to diminishing signaling induction instead of protein degradation, though experiments with proteasome inhibition would be necessary to definitively rule this explanation out.

B C

Figure 3.6. PMA activates ERK signaling in 5C.C7 cells. Lymph nodes were isolated from 5C.C7 mice and cells were treated with 50 ng/mL PMA at 37C for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy used for analyzing cells. (B-C) Representative histograms of pERK (T202/Y204) (B) and total ERK (C) expression for up to 120 min of PMA treatment.

We next sought to compare PMA-induced ERK phosphorylation to that induced by the TCR-associated signaling complex using an antibody targeting the CD3 protein.

5C.C7 T cells were stained with biotinylated-anti-CD3 as well as anti-CD4 and then treated with streptavidin to induce crosslinking of both CD3 and CD4 at time 0 of the analysis.

Stimulated cells were fixed to stop the reaction at various times and the levels of phospho-

ERK measured as discussed in the methods. Interestingly, we found that PMA treatment typically results in maximal ERK phosphorylation with a slower kinetics than anti-CD3 stimulation. In the case of PMA stimulated cells, ERK phosphorylation was first apparent after 5 min of treatment, and it peaked quickly to remain sustained for >60 min (Fig.

3.7B,D). In contrast, antibody-induced ERK phosphorylation can be seen within 1 min of stimulation but diminishes by 15 min of stimulation (Fig. 3.7C-D). Additionally, the fraction of 5C.C7 cells which phosphorylated ERK after stimulation with CD3/CD4- crosslinking was consistently lower than those stimulated with PMA. The amount of pERK, as measured by the peak gMFI of pERK in the antibody-stimulated cells was also

~5 fold lower than that of PMA-induced pERK. Taken together, these data suggest that chemical activation of TCR signaling via PMA follows a distinct kinetics that those induced via the TCR, even though the former was eventually stronger in activating this particular MAPK pathway in a greater proportion of T cells to a greater extent.

B C D

Figure 3.7. PMA-induced ERK phosphorylation follows distinct kinetics from TCR-mediated pERK signaling. Lymph nodes were isolated from 5C.C7 mice and cells were stained with 10 g/mL biotinylated-anti-CD3 and anti-CD4 or left unstained. Unstained cells were treated with 50 ng/mL PMA while stained cells were treated with 5 g/mL streptavidin. All treatments occurred at 37C for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy used for analyzing cells. (B-C) Representative histograms of pERK (T202/Y204) expression by PMA-treated (B) and antibody-crosslinked (C) cells for up to 60 min of treatment. (D) The gMFI of pERK among pERK+ cells plotted as the percentage of the maximum gMFI reached by each treatment group.

3.2.3. PMA and direct TCR-complex signaling recruit different kinds of negative feedback

The activity of ERK in T cells is tightly regulated by many processes. Several molecules acting to dampen ERK signaling are known, including the phosphatases SHP-

1166,167 and DUSP6168-170 as well as the E3-ubiquitin ligase cbl-b171-173. Each of these are involved in turning off ERK activation using different mechanisms. Given the rapid loss or phospho-ERK in the antibody-stimulated T cells (Fig. 3.7C) vs the PMA-stimulated ones (Fig. 3.7B), we examined if these negative regulators are involved differentially between these two triggers. We hypothesized that TCR-mediated ERK activation is necessary for the induction of these negative regulation mechanisms, such that PMA treatment concurrent with TCR-mediated activation would exhibit rapidly attenuated ERK signal. Surprisingly, when both stimuli are concurrently used to activate T cells, ERK phosphorylation occurs rapidly within 1 min of stimulation and more strongly than antibody-mediated induction alone (Fig. 3.8). Further, ERK remains uniformly highly phosphorylated for at least 60 min (Fig. 3.8), indicating that PMA-induced ERK signaling is not attenuated by TCR-induced negative regulation.

pERK+

pERK

Figure 3.8. PMA-induced ERK kinetics dominate TCR-mediated kinetics. Lymph nodes were isolated from 5C.C7 mice and cells were stained with 10 g/mL biotinylated-anti-CD3 and anti-CD4. Cells were treated with 5 g/mL streptavidin and 50 ng/mL PMA for the indicated times at 37C before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. Cells gated as in Figure 3.7A. Representative histograms of pERK (T202/Y204) expression by cells for up to 60 min of treatment are shown.

We next sought to determine if these findings generalize to a polyclonal population of T cells beyond a TCR-Tg system. Lymphocytes from B6 mice were stimulated with

PMA, anti-CD3/CD4 crosslinking, or both stimulation conditions and pERK was measured by phospho-flow cytometry (Fig. 3.9). As in the case of TCR-Tg 5C.C7 cells, in both polyclonal CD4 and CD8 T cells, PMA treatment yielded strong ERK phosphorylation within 5 min of treatment that was sustained for >60 min (Fig. 3.9B-C). The kinetics of antibody-mediated ERK signaling were also virtually identical between 5C.C7 and polyclonal CD4 T cells, while CD8 T cells did not show ERK phosphorylation under these conditions, likely due to the absence of anti-CD8 crosslinking. Additionally, when polyclonal CD4 T cells were both treated with PMA and underwent TCR-crosslinking,

ERK remained strongly phosphorylated for 60 min, indicating that PMA-induced ERK phosphorylation kinetics dominate within polyclonal T cells as well as 5C.C7 cells. Finally, as in 5C.C7 cells, total ERK expression remained constant for >60 min under all 3 stimulation conditions within CD4 and CD8 T cells (Fig. 3.9B-C). These data suggest that

PMA-induced ERK phosphorylation is not subject to the negative regulatory pathways responsible for ending TCR signals.

B C

Figure 3.9. Polyclonal T cells also show distinct kinetics between PMA-induced and TCR-mediated pERK. Lymph nodes were isolated from B6 mice and cells were stained with 10 g/mL biotinylated-anti-CD3 and anti-CD4 or left unstained. Unstained cells were treated with 50 ng/mL PMA (“PMA”, red) while stained cells were treated with 5 g/mL streptavidin, with (“-CD3/CD4 + PMA”, purple) or without (“- CD3/CD4”, blue) 50 ng/mL PMA. All treatments occurred at 37C for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy used for analyzing cells. (B-C) Representative histograms of pERK (T202/Y204) (left) and total ERK (right) expression by CD4+ (B) and CD8+ (C) for up to 60 min of treatment.

Because TCR-CD3 complexes are downregulated and degraded following activation174,175, we sought to determine if the observed attenuation in ERK phosphorylation following TCR-crosslinking is due to a reduction in available TCR-CD3 complexes that can be crosslinked and initiate signaling. 5C.C7 TCRs were crosslinked using biotinylated-anti-CD3 and anti-CD4 and streptavidin as above and then surface expression of CD3ε and TCR were measured by flow cytometry (Fig. 3.10). Although there is a minor decrease in the signal intensity between the unstimulated and stimulated cells for both CD3ε and TCR, there is no further decrease in expression of TCR complexes with increased stimulation time, suggesting that CD3 downregulation cannot explain the attenuation of ERK signaling following TCR-crosslinking. Further, the difference between the staining intensity of unstimulated and stimulated cells is likely technical, as these cells were not stained with biotinylated antibodies to ensure that there is no signal induction in the “unstimulated” sample. It is possible that access to CD3ε and

TCR by the fluorescently labeled antibodies was sterically hindered by the previously bound biotinylated antibodies.

Figure 3.10. CD3 is not downregulated by anti-CD3/CD4 stimulation. Lymph nodes were isolated from 5C.C7-Foxp3-GFP mice and cells were stained with 10 g/mL biotinylated- anti-CD3 and anti-CD4. Cells were treated with 5 g/mL streptavidin for the indicated times at 37C before transfer to an ice bath. Cells were stained with antibodies for flow cytometry analysis. Representative histograms of TCR (left) and CD3ε (right) expression by 5C.C7 cells for up to 30 min of treatment.

We next sought to further examine the differences in negative feedback induction by PMA and TCR-crosslinking by comparing the phosphorylation of ERK following a secondary stimulation. We hypothesized that antibody-mediated stimulation would induce negative feedback on ERK signaling such that a second round of TCR-crosslinking would show no pERK, while PMA treatment would not inhibit subsequent ERK phosphorylation triggered by TCR-crosslinking. As such, 5C.C7 cells were treated with PMA for 30 or 120 min or underwent anti-CD3 and anti-CD4 crosslinking, as above, for 30 min, then all treatments were subjected to secondary stimulation with antibody crosslinking. ERK phosphorylation, as well as that of TCR-signaling molecules SLP-76 and Zap70 were examined by phospho-flow cytometry (Fig. 3.11). Additionally, primary PMA and antibody treatments were used as controls.

As predicted, primary stimulation with antibodies resulted in robust negative regulation on signaling as no ERK phosphorylation was observed at any time point following secondary stimulation (Fig. 3.11B). Further, both SLP-76 and Zap70 also showed impaired phosphorylation (Fig. 3.11C-D), indicating that the negative regulation induced by prior antibody-mediated stimulation most likely acts immediately proximal to the TCR as shown in previous studies167. Conversely, prior treatment with PMA resulted in enhanced ERK phosphorylation following secondary stimulation with antibody- crosslinking (Fig. 3.11B). Both 30 and 120 min incubations resulted in increased pERK observed at baseline, though the 30 min treatment showed approximately 10-fold increased pERK gMFI compared to 120 min suggesting that some form of negative regulation on

ERK signaling is imposed by the later time point (Fig. 3.11B). This is supported by the observation that, although 1-10 min antibody-mediated stimulations resulted in increased

pERK expression compared to baseline, 120 min PMA treatment reached a peak intensity approximately 2-fold lower than that of the 30-min treatment group; however, both PMA treatment lengths resulted in enhanced ERK phosphorylation compared to primary TCR- crosslinking (Fig. 3.11B). This suggests that the augmented T cell activation observed following PMA pre-Tx (Fig. 3.1-3.5) is likely due to increased ERK phosphorylation.

Importantly, SLP-76 and Zap70 phosphorylation remain unchanged following PMA pre-

Tx (Fig. 3.11C-D), indicating that PMA synergizes with TCR-signaling downstream of these signaling nodes, likely at the level of PKC. Therefore, the mechanisms induced by

TCR-mediated activation must predominantly act on TCR-proximal signals.

B C D

Figure 3.11. Transient PMA exposure augments subsequent TCR-mediated ERK phosphorylation kinetics. Lymph nodes were isolated from 5C.C7 mice and cells were stained with 10 g/mL biotinylated-anti-CD3 and anti-CD4 or left unstained. Unstained cells were treated with 50 ng/mL PMA for either 30 or 120 min (or left untreated on ice), while stained cells were treated with 5 g/mL streptavidin for 30 min at 37C, constituting a “1 stimulation.” Treated cells as well as a group of previously untreated cells were then labeled with antibodies as above. These cells were then treated with streptavidin as above for the indicated times before fixation with 1.5% PFA. An additional group of previously untreated cells were stimulated with 50 ng/mL PMA for the indicated times before fixation. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy used for analyzing cells. (B-D) Representative histograms of pERK (B), pSLP-76 (C), and pZap70 (D) expression by cells for up to 10 min of 2 treatment.

3.2.4. Mechanism for differential ERK activation is not clear

One explanation for the lack of apparent negative regulation of PMA-induced ERK phosphorylation by negative-regulatory processes activated by the anti-CD3 treatment, could be that the former is not accessible to the latter. This could be because there are distinct intracellular pools of ERK in cells. PMA would be acting to induce a different pool of ERK than the pool that is subject to TCR-mediated negative regulation. The compartmentalization of ERK signaling has been well studied in several cell lines and primary epithelial tissues, establishing that ERK can be activated at many membrane locations including the plasma membrane, early and late endosomes, the endoplasmic reticulum-Golgi intermediate compartment (ERGIC), and mitochondrial membranes as well as at sites of cytoskeletal arrangement and in the cytosol176-178. The utilization of different scaffolding proteins localized to the various membranes that act to aggregate Raf,

MEK, and ERK and propagate signaling underlie the intracellular compartmentalization of

ERK. This phenomenon has not been investigated thoroughly within T cells, however thymocytes exhibit differential pERK localization and kinetics, such that high affinity, negative selecting ligands evoke rapid, plasma membrane localized ERK phosphorylation, while low affinity, positive selecting ligands show delayed ERK phosphorylation localized to the Golgi179,180.

Based on the difference in ERK phosphorylation kinetics, we hypothesized that antibody-mediated ERK phosphorylation would resemble a high affinity ligand and occur at the plasma membrane, while PMA-induced pERK, which is delayed but sustained, would be show Golgi localization. 5C.C7 T cells were activated using antibodies, PMA, or concurrent stimulation and then stained and prepared for fluorescence imaging, using anti-

GM130 to mark the Golgi membrane (Fig. 3.12). While this experiment does not definitively support or refute the hypothesis, it does appear that pERK and GM130 fluorescence shows the greatest overlap following PMA stimulation alone (Fig. 3.12). An improved staining and imaging scheme including confocal microscopy may be necessary to resolve localization differences. Additionally, comparing different stimulation lengths will likely be necessary to fully appreciate the potentially complex pERK localization within the anti-CD3/CD4 and PMA stimulation condition, as a 5 min time point would be predicted to exhibit mixed pERK localization (Fig. 3.7-3.9).

Figure 3.12. PMA- and TCR-induced ERK phosphorylation may show differential localization. Lymph nodes were isolated from 5C.C7 mice and cells were stained with 10 g/mL biotinylated-anti-CD3 and anti-CD4 or left unstained. Unstained cells were treated with 50 ng/mL PMA (“PMA”), while stained cells were treated with 5 g/mL streptavidin, with (“-CD3/CD4 + PMA”) or without (“-CD3/CD4”) 50 ng/mL PMA. All treatments occurred at 37C for 5 min before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with primary antibodies: AF488-anti-pERK and mouse-anti-GM130. The cells were washed the next day and stained with Cy3-donkey-anti-mouse IgG secondary antibodies. After additional washing, cells were adhered to glass slides using cytocentrifugation. Representative images of the three treatment conditions are shown, with pERK in green and GM130 in red. All images were taken using the 100x objective.

3.2.5. PMA exposure may decrease T cell survival in vivo

We next sought to study the functional impact of PMA pre-Tx on T cell activation and differentiation in vivo. 5C.C7 T cells were collected, loaded with either CFSE or CTV, treated with PMA for 30 min at 37C, and then transferred via retro-orbital injection to

B10.A mice expressing both CD45.1 and CD45.2 (B10.A 1x2). Animals then immediately received i.p. injection of MCC peptide and LPS to activate the transferred 5C.C7 cells.

After 4 days, 5C.C7 T cells treated with PMA showed decrease recovery from secondary lymphoid organs (Fig. 3.13; 2.31-fold; p=0.0551). The decrease in cell recovery is not due to differences in proliferation as both treatment groups showed identical patterns of proliferation at 4 days of stimulation (Fig. 3.14D, 3.15D; data from replicate experiments).

The difference in recovery may be due to increased expression of CD69 following PMA treatment (Fig. 3.1-3.3) resulting in altered T cell circulation between blood and lymphatics. Although, it may be predicted that increased CD69 would lead to increased retention in SLO103,105, it is possible that CD69 expression is accelerated by PMA treatment such that it is also downregulated earlier following than non-treated T cells interfering with the normal kinetics of T cell proliferation within the SLO.

In addition to no observed differences in proliferation, PMA- and PBS-treated

5C.C7 cells showed no significant difference in T-bet or RORt expression across two experiments (Fig. 3.14B-C; Fig 3.15B-C). Further, upon ex vivo restimulation of recovered

5C.C7 T cells with MCC, we observed no significant differences in IFN, IL-4, or IL-17A secretion as measured by ELISA analysis of recovered supernatant (Fig. 3.14E-G; 3.15E).

These findings suggest that PMA treatment does not alter the differentiation of CD4 T cells in vivo.

Figure 3.13. Transient PMA exposure alters T cell survival in vivo. Lymph nodes were isolated from 5C.C7-Foxp3-GFP mice, loaded with CTV, treated with 50 ng/mL PMA or PBS for 30 min at 37C. Cells were washed, resuspended in PBS, and 100,000 cells were transferred to B10.A 1x2 mice (N=3) by retro-orbital injection. Animals then immediately received 1 g LPS and 30 g MCC peptide in PBS via i.p. injection. 4 days after transfer, LN and spleens from animals were collected for flow cytometry. (A) Gating strategy for analyzing transferred 5C.C7 cell population. (B) Total cell numbers of recovered 5C.C7 T cells at day 4 post transfer. Displayed as mean ± SD. Groups compared using Student’s T-test.

B C D

RORt

Figure 3.14. Transient PMA exposure does not alter in vivo T cell proliferation of differentiation. T cells were prepared for flow cytometry after 4 d of in vivo activation as described in Figure 3.13. Cell samples from each animal were restimulated ex vivo using 1 M MCC peptide and supernatants were collected 48 hr later for ELISA. (A) Gating strategy for analyzing transferred 5C.C7 cell population. (B-C) Histograms displaying T-bet (B) and RORγt (C) expression by PBS- (“PBS”, blue) and PMA-treated (“PMA”, red) 5C.C7 T cells. Average gMFI displayed in lower panels, respectively. (D) Histograms displaying CFSE dilution among PBS- and PMA-treated 5C.C7 cells. In all histograms, biological replicates are displayed.

E F G

0 . 6 0 . 8 0 . 3

0 . 6

)

) . ) 0 . 2

. 0 . 4

(

( 0 . 4



I I 0 . 2 I 0 . 1

0 . 2

0 . 0 0 . 0 0 . 0

S A S A S A B M B M B M P P P P P P

Figure 3.14. Transient PMA exposure does not alter in vivo T cell proliferation of differentiation (cont.) T cells were prepared for flow cytometry after 4 d of in vivo activation as described in Figure 3.13. Cell samples from each animal were restimulated ex vivo using 1 M MCC peptide and supernatants were collected 48 hr later for ELISA. (E-G) IFNγ (E), IL-4 (F), and IL-17A (G) concentration in supernatants of PBS- and PMA-treated cells. Concentration represented by measured optical density.

B C D

RORt E

3 0 0 2 0 0 0 . 8

1 5 0 0 . 6

) F

2 0 0 .

t (

t 1 0 0 0 . 4



F I

T 1 0 0 R

5 0 0 . 2

0 0 0 . 0

S A S A S A B M B M B M P P P P P P

Figure 3.15. Lack of effect of PMA treatment of T cell proliferation and differentiation is repeated. T cells were prepared for flow cytometry after 4 d of in vivo activation as described in Figure 3.13, except 5C.C7-Foxp3-GFP cells were used and CTV was used in place of CFSE. Cell samples from each animal were restimulated ex vivo using 1 M MCC peptide and supernatants were collected 48 hr later for ELISA. (A) Gating strategy for analyzing transferred 5C.C7 cell population. (B-C) Histograms displaying T-bet (B) and RORγt (C) expression by PBS- (“PBS”, blue) and PMA-treated (“PMA”, red) 5C.C7 T cells. Average gMFI displayed in lower panels, respectively. (D) Histograms displaying CFSE dilution among PBS- and PMA-treated 5C.C7 cells. In all histograms, biological replicates are displayed. (E) IFNγ concentration in supernatants of PBS- and PMA-treated cells, represented by measured optical density.

3.2.6. Ceramide (C6) does not induce ERK phosphorylation in our hands

As discussed in the introduction to this chapter, there is a paucity of reagents to directly activate individual signaling pathways in T cells. So, we attempted to expand our studies using PMA by using an ERK-specific small molecule activator. C6 had been reported as being a relatively specific ERK activator, with minimal off-target effects including induction of protein phosphatase 2A (PP2A)155,156. Accordingly, we first sought to measure CD69 expression among polyclonal T cells from B6 mice following 30 min pre-Tx with C6 and 24 hr stimulation with anti-CD3 and anti-CD28. We observed no change in CD69 expression as measured by percentage of cells positive for CD69 or gMFI of CD69 among CD69+ cells (Fig. 3.16).

Figure 3.16. C6 exposure does not alter CD68 expression in B6 T cells. Lymph nodes and spleens from 3 B6 mice were collected and pooled. Live, mononuclear cells were separated using ficoll density centrifugation. T cells were enriched using 2 rounds of magnetic, negative selection, first with Dynabeads then with Miltenyi AutoMACS. In both rounds of selection, antibodies targeting CD11b, B220, NK1.1, and MHCII were used to enrich for T cells. Enriched T cells were then treated with the indicated doses of C6 or PBS for 30 min before washing and incubation with anti-CD3ε (1 g/mL, plate- bound) and anti-CD28 (2 g/mL, soluble). After 24 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histogram of CD69 expression by CD4+ T cells. (B) Percentage of CD4+ T cells positive for CD69 (left) and gMFI of CD69+ CD4+ T cells (right). Data includes 3 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD.

We next sought to characterize the induction of ERK phosphorylation by C6 in our model systems. First, CD4 T cells from C57BL/6 mice were stimulated with increasing doses of C6, DMSO, or plate-bound anti-CD3 for 5 to 30 min, and then lysed for western blot analysis (Fig. 3.17A-E, replicate blots from same experiment). Although pERK could be detected by western blot, C6 treated samples do not show an increase in ERK phosphorylation over DMSO control (Fig. 3.17A,C,E).

We alternatively used phospho-flow cytometry to examine ERK phosphorylation following C6 treatment of 5C.C7 cells (Fig. 3.17F). Again, pERK signal could not be detected above baseline for up to 30 min of stimulation using a high dose of C6. One explanation for the lack of pERK induction in our T cell systems could be that the dose or

156 timing of C6 treatment is insufficient to elicit an observable response. One prior study observed ERK phosphorylation within 30 s of treatment but the effect tapered off by 2 min of treatment. Another study155 observed ERK phosphorylation following 24 hr culture with

C6. Both studies, however, used doses of 5 or 10 M C6, suggesting that the dose range used in our experiments would likely elicit an effect. Another explanation is that C6 is not effective in primary T cells, as these studies examined effects in leukemia (HL-60) and hepatocyte carcinoma (HepG2) cell lines.

A B

C D

Figure 3.17. C6 does not induce ERK phosphorylation. (A-E) Lymph nodes and spleens from 2 B6 mice were collected and pooled. CD4+ T cells were enriched for using the SCT EasySep mouse CD4+ T cell isolation kit. Cells were then stimulated with anti-CD3ε (10 g/mL, plate-bound), the indicated dose of C6, or DMSO for 5, 10, 15, or 30 min (“T5” through “T30,” respectively) at 37C. Cells were then flash frozen in liquid N2 for 5 min before thawing and applying 2x lysis buffer. Lysates were then prepared for western blot. (A) Western blot analysis of anti-CD3ε treated cells and C6 treated samples for 15 min using anti-pERK (T202/Y204) primary antibody. (B) Densitometry of panel A quantified using ImageJ. (C) All samples were analyzed using anti-pERK primary antibody before stripping and reblotting for total ERK (D).

100

Figure 3.17. C6 does not induce ERK phosphorylation (cont.). (A-E) Lymph nodes and spleens from 2 B6 mice were collected and pooled. CD4+ T cells were enriched for using the SCT EasySep mouse CD4+ T cell isolation kit. Cells were then stimulated with anti-CD3ε (10 g/mL, plate-bound), the indicated dose of C6, or DMSO for 5, 10, 15, or 30 min (“T5” through “T30,” respectively) at 37C. Cells were then flash frozen in liquid N2 for 5 min before thawing and applying 2x lysis buffer. Lysates were then prepared for western blot. (E) Replicate of panel C. (F) Lymph nodes were isolated from 5C.C7 mice and cells were treated with 100 M C6 for the indicated times or 50 ng/mL PMA for 5 min. All treatments occurred at 37C before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. Representative histograms of pERK expression are shown.

101

3.2.7. Early TCR signaling is modulated by prior antigen encounter

After examining how brief, transient alterations of TCR-signaling bias subsequent

T cell activation, we were interested in how more remote or long-term signaling impacts subsequent activation events. As such, we used an in vivo activation scheme modeling acute activation by adoptive transfer of 5C.C7 T cells to either T cell-depleted (Fig. 3.18) or T cell replete (Fig. 3.19) host animals, followed by i.p. injection of MCC peptide and

LPS. In parallel, we modeled a chronic stimulation paradigm by adoptively transferring

5C.C7 T cells to animals that constitutively express PCC under the control of the MHCI promoter (T cell deplete host: PCC-CD3-KO, Fig. 3.18; T cell replete host: PCC-B10.A

1x2, Fig 3.19). We could then compare the early signaling events following ex vivo PMA,

TCR-crosslinking, or combination stimulation between these groups as well as naïve

5C.C7 cells.

In T cell deplete hosts (Fig. 3.18), PMA treatment induced strong ERK phosphorylation across all 3 in vivo activation states (Fig. 3.18B); however, both acutely and chronically activated cells showed diminished pERK signal at 1 min compared to naïve cells, suggesting slightly delayed signaling in these populations (Fig. 3.18B, purple and blue). Additionally, while naïve and acutely activated T cells showed >90% pERK positivity at peak signaling, the chronically activated T cells only reached 75% pERK+, suggesting some inhibition of PMA-induced ERK phosphorylation unique to this activation state (Fig. 3.18B). Further, PMA treatment did not yield phosphorylation of

SLP-76 or Zap70 over the baseline state, as expected, but both acutely and chronically activated T cells showed increased baseline phosphorylation of both molecules (Fig. 3.18B, middle and right columns). Interestingly, among the chronically activated population, 2

102

populations of cells, ~50% each at baseline, could be observed, one expressing pZap70 and pSLP-76, while the other was negative for both markers (Fig. 5.18B, purple). Although increased expression of phosphorylated Zap70 has been observed by western blot analysis following chronic activation, this technique would not allow for the observation of these two discrete populations.

All three populations of T cells stimulated ex vivo by antibody-crosslinking showed rapid SLP-76 and Zap70 phosphorylation that returned to baseline between 10 and 30 min of stimulation (Fig. 5.18C). Among chronically activated cells, while peak Zap70 phosphorylation occurred after 1 min, only 82% of cells were positive, suggesting that a portion of these T cells are refractory to TCR signaling (Fig. 5.18C, purple, right column).

Most strikingly, both the acutely and chronically activated T cells showed no ERK phosphorylation following antibody-mediated stimulation, while naïve cells showed robust phosphorylation peaking at 3 min and diminishing after 10 min of stimulation (Fig. 5.18C, left column).

T cells stimulated with both antibodies and PMA showed ERK phosphorylation kinetics virtually identical to that of PMA alone, with slightly increased pERK signal at 1 min across all 3 groups, while SLP-76 and Zap70 phosphorylation kinetics followed that of anti-CD3 and anti-CD4 alone (Fig. 5.18D). These data suggest that, as in naïve cells, in acutely and chronically activated T cells PMA-induced ERK signaling kinetics dominate pERK expression in the context of combined stimulation, though it may have been predicted that activated T cells would induce different negative feedback mechanisms than naïve cells, potentially resulting in attenuated effects of PMA.

103

Figure 3.18. Acutely and chronically activated T cells show distinct TCR-mediated signaling patterns from naïve T cells. Lymph nodes from 5C.C7 mice were collected, washed, and prepared for adoptive transfer. An “acute” activation state was achieved by transferring 1 million naïve 5C.C7 T cells were to CD3ε-KO mice by retro- orbital injection, which then received 0.5 g LPS and 30 g MCC peptide by i.p. injection 6 hr after transfer. A “chronic” activation state was achieved by transferring 1 million naïve 5C.C7 T cells to PCC-CD3ε-KO mice by retro-orbital injection. After 12 d, LN were collected from these mice as well as a naïve 5C.C7 mouse (“Naïve”) and prepared for ex vivo stimulation and intracellular phospho-flow cytometry as in Figure 3.9. (A) Representative gating strategy for identifying and analyzing recovered 5C.C7 T cells. (B-D, next page) Representative histograms of pERK, pSLP-76, and pZap70 expression by cells for up to 30 min of ex vivo PMA-treatment (B), anti-CD3 and anti-CD4 stimulation (C), or PMA and antibody stimulation (D). Naïve (red), acute (blue), and chronic (purple) in vivo activation as indicated.

104

105

Although recovery of 5C.C7 T cells was greatly diminished in our experiments using T cell-replete hosts (Fig. 3.19), the trends described for the T cell deplete host were largely replicated. ERK phosphorylation was greatly diminished among acutely and chronically activated T cells stimulated ex vivo with anti-CD3 and anti-CD4. Additionally, chronically activated cells showed increased baseline pSLP-76 (Fig. 3.19B) and pZap70

(Fig. 3.19C), with a fraction of cells remaining negative for each despite antibody-mediated stimulation.

Lastly, the T cell-replete hosts allowed for the examination of an internal control group of host-derived, V3+ T cells (“Endogenous V3+ cells”). In response to PMA or antibody stimulation, this population showed phosphorylation kinetics most closely resembling the naïve 5C.C7 T cells, consistent with the fact that these have been largely un-manipulated and should reflect a naïve T cell state (Fig. 5.19B-C, right columns).

However, as ex vivo PMA-treated endogenous cells show increased ERK phosphorylation at 1 min compared to naïve 5C.C7, more closely resembling the respective acutely activated 5C.C7 sample, it is possible that a subpopulation of endogenous V3+ cells had been previously activated. This is bolstered by the observation that endogenous cells shoe a population of pERK- T cells following TCR-crosslinking, again resembling the respective acutely activated 5C.C7 population. Activation of these endogenous cells could be from the MCC and LPS injection, as it would be predicted that a small precursor frequency of MCC specific, V3+ cells would be present in a polyclonal population.

Alternatively, unmanipulated wild type animals show a small population of CD44hi, antigen-experienced T cells, possibly due to encounter with the microbiome. These cells

106

may underlie the observed differences in early signaling between the endogenous V3+ and transferred 5C.C7 populations.

107

Figure 3.19. Distinct TCR-mediated signaling patterns of acutely and chronically activated T cells apparent in T cell replete hosts. Lymph nodes from 5C.C7 mice were collected, washed, and prepared for adoptive transfer. An “acute” activation state was achieved by transferring 100,000 naïve 5C.C7 T cells were to B10.A 1x2 mice by retro- orbital injection, which then received 0.5 g LPS and 30 g MCC peptide by i.p. injection 6 hr after transfer. A “chronic” activation state was achieved by transferring 100,000 naïve 5C.C7 T cells to PCC-B10.A 1x2 mice by retro-orbital injection. After 17 d (B) or 19 d (C), LN were collected from these mice as well as a naïve 5C.C7 mouse (“Naïve”) and CD4+ T cells were enriched for using negative magnetic selection, using Dynabeads targeting B220, CD11b, CD11c, NK1.1, and CD8. Cells were then prepared for ex vivo stimulation and intracellular phospho-flow cytometry as in Figure 3.9. (A) Representative gating strategy for identifying and analyzing recovered 5C.C7 T cells. The “lineage” channel includes anti-B220, anti-CD11b, and anti-MHCII. (B-C, next pages) Representative histograms of pERK, pSLP-76, and pZap70 expression by cells for up to 30 min of ex vivo PMA-treatment (upper), anti-CD3 and anti-CD4 stimulation (middle), or PMA and antibody stimulation (lower). Naïve (red), acute (blue), and chronic (purple) in vivo activation as indicated.

108

109

110

3.2.8. Optimizing a comprehensive panel for further evaluation of signaling interactions

Although conventional phospho-flow cytometry is a powerful tool for interrogating changes in TCR signaling, the limited array of lasers and detectors on most flow cytometers limits the breadth of signaling nodes that can be examined concurrently. Spectral flow cytometry that uses upwards of 20 detectors to measure the entire emission spectrum from each excitation laser allows for the discrimination between similar fluorophores that are considered identical by conventional flow cytometry181,182.

Accordingly, this allows for the concurrent use of dozens of fluorescent markers regardless of peak emission overlap, increasing the power to analyze several signaling molecules within discrete T cell subpopulations of a polyclonal T cell population. We sought to use this technology to more specifically characterize the changes in TCR signaling following prior treatment conditions and activation states. A pilot study examining the phosphorylation of 8 signaling molecules downstream of TCR and CD28 signaling following antibody-mediated activation of polyclonal B6 T cells is shown below

(Fig. 3.20). We were able to resolve ERK phosphorylation (Fig. 3.20B,C), showing similar kinetics to prior experiments in both CD4+ and CD8+ T cells. Additionally, CD44hi, antigen-experienced T cells showed minimal ERK phosphorylation, among CD4+ and

CD8+ cells, consistent with our adoptive transfer studies (Fig. 3.18-3.19). Further, both pSLP-76 and pZap70 (Y292) showed kinetics in line with prior experiments; however, pZap70 (Y319) showed minimal increase in signal intensity above baseline (Fig. 3.20 B,F).

Although we had difficulty detecting shifts in phosphorylation of p38, CREB, and

Akt, phosphorylation of S6, a signaling molecule downstream of Akt, showed robust signal among CD44lo CD4+ and CD8+ T cells (Fig. 3.20B, J). Interestingly, much like ERK

111

phosphorylation, S6 showed diminished phosphorylation among CD44hi T cells. These data suggest that specific signaling nodes are repressed following prior T cell activation.

112

Figure 3.20. Detailed TCR-mediated signaling kinetics can be observed using spectral flow cytometry. Lymph nodes were isolated from B6 mice and cells were stained with 10 g/mL biotinylated-anti-CD3, anti- CD4, anti-CD8, and anti-CD28. Cells were treated with 5 g/mL streptavidin at 37C for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for spectral flow cytometry using the Cytek Aurora. (A) Gating strategy used for analyzing T cells. (B) Representative histograms of indicated phosphorylated signaling protein for the indicated length of stimulation. CD4+ (upper) and CD8+ (lower) are shown separately, as are CD44lo (left) and CD44hi (right) expressing cells. (C-J, next page) gMFI of the indicated signaling protein plotted for all 4 populations shown above.

113

C D

E F

G H

I J

114

3.3. Significant findings & Discussion

The central goal of this chapter was to use a reductionist model to evaluate the hypothesis that triggering subsets of TCR signaling pathway (Signal 0) before the T cell actually engages its antigen and receives a full activating signal from it, biases the kind of activation resulting from the latter. We used PMA, a compound that activates PKC and

MAPK pathways to mimic a signal 0. Interestingly, we find that this pre-biasing selectively affects expression of CD69 and subsequent cell survival, but not cytokine production, proliferation, or differentiation of the T cells. While these data suggest that the basic premise of signal 0 merits further evaluation, the mechanisms by which PMA mediates this long-term consequence is also worth considering. T cell survival in vivo is a complex outcome resulting from both the ability of cells to utilize available trophic factors

(by either upregulating receptors for nutrients, such as Glut1 or by migrating to specific sites of cytokine/nutrient availability) as well as intrinsic changes in cell survival machinery. We showed that PMA pre-Tx yields increased CD69 expression by T cells, which, by driving the internalization and inhibition of S1PR, would be predicted to promote retention in lymph nodes 103,105. Although T cells activate and proliferate rapidly within

SLO, lymphoid stromal cells have mechanisms for restraining T cells including the release of nitric oxide (NO), presumably to provide a brake on an overactive, potentially damaging

T cell responses 183-185; thus, the effect of PMA may have been to sequester pre-treated cells in lymph nodes where they were subject to suppression by these mechanisms.

In the process of evaluating the consequences of PMA on T cells, we also developed sensitive assays to measure intracellular signaling by flow cytometry. These studies revealed an interesting dichotomy between the kinetics of ERK phosphorylation due to

115

PMA vs TCR-proximal (CD3) signaling. While TCR-mediated signaling elicited ERK phosphorylation that was rapidly attenuated within 15 of stimulation indicative of negative feedback, PMA maintained high levels of pERK for at least 2 hours. Further, experiments in which both stimuli were provided concurrently, which also yielded protracted ERK phosphorylation kinetics, suggest that PMA-induced ERK is not subject to the same negative feedback as TCR signaling. This could potentially be due to intracellular compartmentalization between ERK signals elicited by TCR and PMA, preventing negative regulation at nodes of PMA-induced signaling; however, experiments aiming to determine differential intracellular localization of ERK were inconclusive, so further experimentation will be required to test these hypotheses.

Finally, these techniques allowed us to examine the signaling differences in T cells existing in two different functional states in vivo. The fate of T cells which have been recently stimulated by antigen itself can (perhaps) be considered an extreme of the signal

0 state – one in which T cells were recently already stimulated along all pathways. In this context, the changes in T cells resulting from chronic exposure is quite potent. Two populations could be identified among chronically activated cells. The first shows high baseline phosphorylation of proximal signaling molecules Zap70 and SLP-76, approximately 10-fold above that of naïve cells; however, upon TCR-mediated activation, these cells showed an attenuated increase in phosphorylation of both molecules. A second population shows very low expression of pZap70 and pSLP-76 that remained completely unchanged by TCR-stimulation. Interestingly, both populations show hypo-responsiveness to activation but the mechanisms that enforce it would likely be vastly different between the two populations. With the prevalence of studies examining techniques to reverse T cell

116

exhaustion, most notably checkpoint blockade, it seems that further investigation of the heterogeneity within the “exhausted” pool is necessary, as more complex strategies may be needed to reactivate these different states of hypo-responsiveness.

117

Chapter 4: Receptors for the non-immune chemiome on T cells

4.1. Introduction

In our approach to evaluate the Signal 0 hypothesis, the logical question that follows from our studies showing that triggering parts of the TCR signaling pathway does impact subsequent T cell activation is its physiological counterpart. We assume that in vivo, resting steady-state T cells are also continuously exposed to a variety of other environmental ligands – such as hormones, pharmaceuticals, metabolites, neurotransmitters etc. – all of which together we refer to as the “chemiome” (introduced in

Chapter 1). Of course, the key is to understand which subsets of these have receptors on T cell which can indeed signal using parts of the signaling machinery that would lead to differential signal 0 states. Obviously, the receptors which can signal in such a fashion must fulfill three criteria. First, they should be expressed in resting (naïve or memory) T cells, without requiring an initial TCR signal for upregulation. Secondly, they should be able to signal using some (but not all) signaling modules that are subsets of the overall

TCR signaling cascade. Third, the ligands for these receptors should be available in vivo, and signal to resting T cells in order to set up differential TCR biases. The emphasis on

‘non-immune’ also signifies these requirements. We consider cytokine receptors and known cell adhesion molecules to be ‘immune’ receptors – largely because they have been studied in hematopoietic cells. Our focus here is on those functional receptors that are expressed quite broadly and whose signaling properties are not unique to T cells.

These criteria are key, because if the TCR signals are first required to upregulate expression of the receptor, then it is unlikely to impact the initial antigen-activation signal.

In our initial surveys of the literature, we found surprisingly little information published

118

about receptors that meet these criteria. In this chapter, we discuss our efforts to catalog and characterize such receptors.

4.1.1. Defining a strategy for identifying relevant receptors for the Non-immune chemiome (rNIC)

The strategy we used here was to focus on receptors that are known to share signaling modalities. The MAPK ERK1/2 is triggered by several other receptors, including insulin and insulin-like growth factor receptors186,187, while cellular processes, like the response to stress induce p38 activity188,189. In order to define a list of receptors for NIC ligands expressed by T cells, an extensive literature review was performed. First, utilizing a combination of pharmacology databases, we defined a list of the totality of receptors expressed by mammals, broken down by receptor type. Membrane receptors are classified into three types of which all were included for analysis: 1) ligand-gated ion channels, 2)

GPCR, and 3) kinase-linked and related receptors. Among these, the following subclasses were excluded from analysis as they were considered “immune” receptors: chemokine, leukotriene, prostaglandin, S1P, complement receptors. Additionally, among enzyme- linked receptors, receptor-type protein tyrosine phosphatases (PTPR) were not included as they are not ligand binding. Further, orphan receptors with no known ligands also excluded.

Lastly, receptors for neurotransmitters are considered separately in Chapter 5.

Upon compilation of a curated receptor list, each receptor was searched on PubMed using the inclusive search term scheme of “(t cell) and ([receptor]),” repeating the search with the receptor gene name and any common protein names. Resultant publications were considered indicative of receptor expression if data directly measured expression using RT- qPCR or protein analysis in at least one T cell population or if receptor ligation showed a

119

measurable functional effect; however, functional effects must be mediated by direct signaling rather than indirectly through agonism of another cell. This could be shown through purified culture or T cell specific knock out experiments. An inclusive approach was taken with respect to studies with some ambiguity in this regard.

4.2. Results

4.2.1. T cells express receptors for a broad array of environmental ligands.

A comprehensive list of NIC receptors that show expression by T cells based on the above criteria is found in Table 4.1. The list includes 23 receptors of which all but one are of the enzyme-linked receptor class with the remaining a GPCR. In considering these findings, trends can be identified when considering the varying ligands for these receptors as well as the signaling modules elicited upon ligation. Each will be discussed in turn.

Ligand specificity

The majority of receptors (13 of 23) respond to hormones and growth factors, signaling through receptor tyrosine kinases (RTK) or receptor serine/threonine kinases

(RSTK). This group can be subdivided into receptors for ligands that are systemically circulated – insulin, hepatocyte growth factor (HGF), insulin-like growth factor (IGF), fibroblast growth factor (FGF), erythropoietin (EPO), and activin – and those that signal locally – EGF, nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and bone morphogenic protein (BMP). Detailed discussion of the numerous effects mediated by each of these factors is beyond the scope of this section, but, generally, these factors promote proliferation and cell survival, commonly regulating cellular metabolism by Akt (Table 4.1).

120

A second group of receptors bind membrane-bound ligands. Axl and MerTK are

TAM family receptors (named for its constituents Tyro3, Axl, and MerTK) that bind Gas6 and Protein S, which are expressed by a variety of tissues and act, generally, to inhibit cell cycle progression190. Additionally, 6 members of the ephrin receptor (Eph) family have been identified in T cells (Table 4.1). Ephrins are a family of GPI-anchored proteins that have been predominantly described in governing tissue development and patterning during development, especially of neuronal axons191,192. They also act in adulthood to guide angiogenesis as well as the maintenance of most stem cell niches193,194. Despite typically engaging in cell-cell interactions, ephrins can be released by GPI-anchor cleavage and signal at a distance.

Another receptor identified to be expressed by our search is neuropilin-1 (NRP1).

NRP1 is a somewhat enigmatic receptor as it does not have intrinsic kinase ability and instead acts as a co-receptor to stabilize other receptor-ligand interactions. NRP1 necessarily participates in the binding of two independent ligands: class 3 semaphorins

(Sema3s) and VEGF195,196. Sema3s are secreted, unlike other semaphorins, and act similarly to ephrins in that they guide axon development as well as regulate cell migration, organogenesis, and vascular development in many contexts197. NRP1 cooperates with plexins as well as L1 cell adhesion molecule (L1CAM) in order to mediate response to

Sema3s. VEGF, which is critical for guiding angiogenesis, requires NRP1 to bind VEGF- receptor2 in order to generate high affinity ligation 195,198. Additionally, NRP1 has been shown to bind TGF1 during SMAD induction as well as binding c-MET and platelet- derived growth factor, NRP1 is not essential in any of these cases.

121

The final receptor identified by this analysis is the trace amine-associated receptor

(TAAR) 1. The TAAR family was recently discovered in 2001 as an intracellular receptor that binds exogenous amphetamines and ephedrines as well as amino acid metabolites, such as phenylethylamine, tyramine, and tryptamine199,200. TAAR1 signaling plays a variety of roles within the central nervous system but also has been shown to regulate metabolism, both by triggering hormonal release in the pituitary as well as locally in the pancreas and stomach199.

Receptor signaling

RTK are the most common type of receptor identified (18 of 23) with respect to the mechanism of signaling (Table 4.1, blue). As described for cytokine signaling in section

1.2.3, RTK act by receptor dimerization upon ligand binding, leading to activation of their intrinsic kinase domain and initiation of signaling cascades. Of note, the EPO receptor

(EPO-R) does not have intrinsic kinase activity, but instead binds JAK2 to initiate STAT5 signaling. Again, the precise mechanisms by which each receptor elicits downstream signaling modules will not be described but can be found within the cited literature (Table

4.1). However, it is worth noting that RTK most commonly induce 2 signaling modules:

PI3K/Akt and MAPK (typically ERK) signaling. Thus, signaling through these receptors would be predicted most likely to modulate subsequent TCR and co-stimulatory receptor signaling.

The second most common class of receptors are RSTK of the activin-receptor family, of which TGFR is a member. Therefore, these receptors initiate signaling as described in section 1.2.3 to activate SMAD signaling. As a consequence, signal 0 events

122

mediated by these receptors would be predicted to act similarly to TGF, most likely influencing signal 3 but potentially signal 1 as well (see section 1.2.3).

TAAR1 is the lone NIC GPCR identified by our review of the literature. As it is a

Gs-coupled receptor, its ligation activates adenylate cyclase to generate the second messenger cAMP. Additionally, TAAR1 has been shown to elicit effects through -arrestin

199,200 signaling as well . Extensive discussion of Gs-coupled signaling can be found in

Chapter 5 so it will not be discussed further here.

Although NRP1 does not have intrinsic signaling capabilities, we can consider the signaling pathways triggered by each receptor complex it participates in. Sema3 commonly binds plexins A1 or A2 in coordination with NRP1197. Plexins act predominantly by recruiting GTPases including Ras and Rac1201, thus activating MAPK signaling. VEGF binding NRP1 with VEGFR2 induces PLC signaling and thus activates ERK as well as

Ca2+ signaling202,203. VEGFR2 has also been shown to elicit FAK, which plays a critical role in cell migration as part of angiogenesis198. In T cells, FAK has been shown to broadly inhibit TCR signaling by targeting Lck204,205. Although the context of NRP1-mediated signaling will be complex due to its ligand co-receptor complexity, when it is recruited as part of a signal 0 event, signal 1 would be predicted to be most impacted.

123

Table 4.1. T cells express receptors for NIC molecules. List of receptors reported to be expressed by at least 1 T cell subset. Receptor names are colored based on their receptor type: RTK (blue), RSTK (gold), GPCR (purple). Citations for studies that show expression are included under “T cell expressed.” Citations that describe the signaling modality utilized by the receptor appear under “Signaling modality observed.” Among the observed signaling modalities, entries in green boxes indicate that the citation refers to signaling in a cell type other than T cells or the generalized understanding of signaling pathways for the given receptor. Entries in red boxes indicate the receptor was shown to inhibit the indicated pathway.

Receptor Gene T cell Signaling modality observed 2+ symbol expressed ERK p38 JNK Ca NFB Akt JAK- SMAD Gs STAT Insulin-R Insr 187,206-209 187 187,209 c-MET Met 210-212 213,214 213,214 213,214 213,214 215 (HGFR) IGF-1R Igfr1 126,131,216- 131 126,131 126 223 IGF-2R Igfr2 224-229 (M6PR) FGFR1 Fgfr1 124,230-232 130 130 124 130 EGF-R Egfr 233-235 234,235 235 236 TrkA Ntrk1 237,238 239,240 240 240 240 TrkB Ntrk2 241-244 244,245 245 245 245 245 245 TrkC Ntrk3 243,246 247 247 247 247 247 247 247 EPO-R Epor 132,248,249 132 Activin-R1B Acvr1b 250,251 252,253 (ALK4) Activin-R2A Acvr2a 254 253,255 BMP-R1A Bmpr1a 256 257 Axl Axl 258 259,260 259 259 259,261, 262 MerTK Mertk 258 259 259 259 259,261 Neuropilin-1 Nrp1 263-268 EphA1 Epha1 269,270 191 191 EphA3 Epha3 271 191 EphB1 Ephb1 272-275 191,273 273 EphB2 Ephb2 272 EphB3 Ephb3 276 EphB6 Ephb6 277,278 TAAR1 Taar1 279 280

124

4.2.2. EGFR may not participate in signal 0

We chose to examine EGFR as a potential component of signal 0 because, among the NIC receptors identified, its signaling has been best characterized in T cells, specifically (see Table 4.1). Before examining the effects of EGFR signaling, we sought to detect EGFR expression on T cells by flow cytometry. Due to a limited availability of fluorophore conjugated, flow cytometry antibodies, we used an unconjugated anti-EGFR primary antibody followed by anti-isotype secondary conjugated to AF488. This staining strategy was applied to B6 splenocytes, mesenteric LN (mLN), and subcutaneous LN

(scLN) as well as murine fibroblast NIH 3T3, which, although not highly, express

EGFR281. We additionally stained splenocytes with anti-mouse IgG as a technical control.

We could not identify EGFR expression by T cells using this assay, finding no signal above the negative control (Fig. 4.1). It is possible that the primary antibody we used is ineffective in this context as it is unclear whether 3T3 cells showed expression either.

Because EGFR has been shown to be expressed by naïve and activated T cells in the mLN and spleen by flow cytomety233, further analysis using an alternative experimental strategy will be necessary.

125

Figure 4.1. EGFR expression by T cells is not detected. Lymph nodes and spleen were isolated from B6 mice, while P7 3T3 cells were recovered from culture using trypsin. After washing, cells were fixed using 1.5% PFA, permeabilized with MeOH, and rehydrated using FACS buffer. Cells were then incubated with the indicated primary antibody, washed, and incubated with AF488-goat-anti-rabbit IgG secondary antibodies, in preparation of flow cytometry. (A) Representative secondary antibody fluorescence in T cell populations. (B) gMFI of secondary antibody signal among primary mouse cell samples (“negative control” indicates spleen population incubated without primary antibody).

126

Despite being unable to detect EGFR directly, based on prior literature describing

EGF-induced ERK phosphorylation234,235, we sought to replicate these findings using and characterize the ERK-signaling kinetics using the phospho-flow strategy described in chapter 3. In this experiment, we treated LN cells from B6 mice with EGF for up to 1 hr, using PMA treatment as a positive control (Fig. 4.2). EGF treated cells showed no ERK phosphorylation above baseline with up to 60 min of EGF treatment in either CD4 cells as a whole or the CD4+CD44hi population (Fig. 4.2B-C). Although we predicted pERK would not be observable in naïve cells because it is not highly express in this population, it was surprising to see no ERK signaling within the CD44hi population which expresses

EGFR highly233. One explanation may be that, while CD44hi, this population in unmanipulated mice not recapitulate the expression status of T cells activated in the context of parasitic worm infection233. Other explanations are more technical, including the possibility that the dose of EGF used was insufficient or the reagent had degraded.

Further troubleshooting will be necessary, but based on these results, EGF does not effectively transduce a signal among resting T cells, and, thus, cannot contribute to signal

127

Figure 4.2. EGF does not induce ERK phosphorylation in T cells. Lymph nodes were isolated from B6 mice and cells were treated with 1 g/mL EGF for the indicated times or 50 ng/mL PMA for 5 min. All treatments occurred at 37C before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy for analyzing T cells. (B) Representative histograms of pERK expression among all CD4+ T cells (left) and CD4+ CD44hi (right) at the indicated stimulation times are shown. (C) Quantification of pERK gMFI among all CD4+ T cells (left) and CD4+ CD44hi (right) are shown.

128

4.2.3. Insulin treatment does not modulate T cell proliferation

We next examined the role insulin signaling plays in T cell activation, because, like EGFR, INSR signals by inducing ERK phosphorylation and Akt, making it a prime candidate for acting as part of signal 0. Further, the endocrine and cyclic nature of insulin circulation physiologically present opportunities for unique temporal and spatial influence on T cell signaling. To screen for a broad effect on T cell activation as would be predicted based on its associated signaling modalities, we simply stimulated 5C.C7 cells, both naïve and pre-activated, with antibody or peptide and APC while in the presence of a broad range of insulin doses and assayed for T cell proliferation. We first found that naïve and pre-activated T cells show different propensities to proliferate depending on the type of stimulation, with antibody dramatically favoring pre-activated cells while naïve cells respond better to antigen (Fig. 4.3), likely due to different demands for co-stimulation. Despite this difference, both T cell populations showed no change in proliferation in response to insulin treatment across a 108-fold dose range.

We considered the possibility that the antigenic stimulation using the agonist peptide MCC may be too strong and masked any potential effect of insulin. Thus, we then repeated the experiment using a range of antigenic peptides with ranging affinities for TCR binding, using MCC with varying amino acid substitutions. Interestingly, in this experiment, across all peptides and peptide doses, pre-activated T cells showed greater proliferation than naïve cells (Fig. 4.4). Yet, in line with the previous findings, insulin treatment had absolutely no effect on T proliferation (Fig. 4.4). One explanation the observed lack of effect may be that naïve T cells show minimal expression of INSR209; however, this does not account for the absence of effect in the pre-activated cell

129

population. Alternatively, while prior studies have shown a dependence on INSR to generate effector function and proliferation, these studies utilized receptor knock out models206,209. It is possible that insulin is in, fact necessary for these processes, but that only a minimal amount is required to provide sufficient signaling, thus exogenous insulin would not elicit a measurable effect. This may be overcome by employing an experimental paradigm in which naïve cells are rested in culture, thus eliminating on going insulin exposure from the in vivo setting, before activation with insulin supplementation.

130

Figure 4.3. Insulin has no effect on 5C.C7 T cell proliferation. “Pre-activated” 5C.C7 cells were generated using ex vivo culture with APCs from CD3-KO mice and 1 M MCC for 3 d. Cells were then rested in the absence of antigenic stimuli but IL-2 supplementation for 8 d. Pre-activated and freshly isolated naïve 5C.C7 were activated using anti-CD3/28 or with APC (CD3- KO) and 1 M MCC peptide in the presence of indicated dose of insulin (combination of bovine and human in a 1:1 mix). After 48 hr in culture, 1 Ci of 3H-thymidine was added to each well. After an additional 24 hr, cells were harvested onto a nylon wool filter and scintillation was counted. (A-C) Measured counts per million among naïve T cells (open circles) or pre-activated T cells (closed circles) activated using (A) 0.1 g/mL anti-CD3, (B) 1 g/mL anti-CD3, or (C) APC and MCC peptide. Experiment run in biological triplicate.

131

B C

D E

Figure 4.4. Insulin does not alter proliferation mediated by altered peptide ligands. Naïve or pre-activated (as in Figure 4.3) were activated by incubation with APC (CD3-KO) and increasing doses of the indicated altered MCC peptide ligand. Cultures were either supplemented with insulin (mix of human and bovine, 0.5 g/mL each) or left untreated. After 48 hr, 3H-thymidine was added and scintillation subsequently measured 24 hr later as described in Fig. 4.3. (A-E) Measured counts per million for each culture condition. Experiment run in technical triplicate.

132

4.3. Significant findings & Discussion

The objective of this chapter was to evaluate a subset of potential receptors for environmental ligands that can deliver signal 0 to T cells in vivo. Specifically, we focused on receptors that are not also neurotransmitter receptors (see Chapter 5 for discussion of this receptor family). While the study laid the groundwork of receptors for future experimentation, following up on individual receptors proved challenging – especially in terms of reagents available to study them. We were surprised to observe that EGF treatment did not yield ERK signal transduction in B6 T cells given that prior studies using amphiregulin (AREG), another ligand for EGFR, elicits pERK234 and that specific EGFR inhibition reduces pERK signal both at baseline and during antibody-mediated TCR stimulation235. It is possible that the difference in ligand, AREG vs EGF, drives distinct signaling modalities in T cells282,283, but that has yet to be reported. Further, both of these studies used primary human cell cultures, introducing the possibility that species differences underlie the discrepancy in observed signaling.

A second receptor identified is the insulin receptor which we found especially interesting considering that the pancreas is an organ that is targeted by T cells during diabetes284,285. So, any dose-dependent effects of insulin on T cell activation could affect the potential for triggering diabetes quite significantly. Intriguingly, despite an exhaustive analysis of different doses as well as different affinity of peptides, we did not find any modulatory effect of insulin with regards to T cell proliferation. Additional studies examining T cell differentiation and cytokine production may reveal a role for insulin signaling.

133

These findings, while highlighting the challenges in studying this hypothesis, also suggest an alternative. It is possible that at least some intuitive candidates for signal 0 are shielded. Similar to the discussion of compartmentalization with regards to PMA-mediated activation of ERK, so too may the signaling elicited by NIC receptor signaling be isolated from that of the TCR. Noting that ERK signaling occurs in discrete subcellular locations176,178, those of EGFR- and TCR-induced signaling may not be compatible for crosstalk. It is thought that high affinity, agonistic ERK signaling occurs at the plasma membrane179 and that appears to be the same for EGFR236; however, EGFR undergoes endocytosis shortly following activation to attenuate signaling236, potentially inhibiting the opportunity for interaction with the TCR. Additionally, the ability of lipid rafts to compartmentalize ERK signaling nodes could be responsible for the lack of an effect286,287.

Although these possibilities would limit the potential impact of signal 0, signaling compartmentalization would likely be receptor and signaling modality specific, as we know from our understanding of signals 1-3 that signaling from different receptor sources can modulate one another. Further studies will be necessary to characterize these phenomena.

134

Chapter 5: Neurotransmitter receptors constitute a major source of rNIC relevant

to T cells

5.1. Introduction

5.1.1. Neuronal regulation of immunity

Given the vast diversity of the endogenous chemiome, we posited that a system with a range of related endogenous chemicals that are naturally likely to be in a T cell’s microenvironment even without an infectious exposure would provide compelling insight into the impact of signal 0 on T cell behavior. Based on our review of the NIC literature

(as discussed in Chapter 4), we discovered a system that fits these criteria perfectly: neurotransmitters (NT). Further, the effects of neurotransmitters on T cells has been studied in many contexts before, providing firm ground for further investigation. There is even a teleological argument for studying NT signaling to T cells as the immune and nervous systems are both tasked with sensing and responding to threats to the organism’s survival.

While lymphocytes detect and resolve the threats posed by infections, tumors, etc. at a cellular level, the nervous system can be thought of as acting at an organismal level using mechanisms such as reflexive withdrawal from harmful stimuli or the engagement of the sympathetic “fight or flight” response288,289. In this context, it is not surprising that the two systems have evolved to communicate with each other and perhaps even regulate one another290-292. The ability of neural outputs to regulate cells of the innate immune system has been well studied, not only in mammals but also in invertebrates291,293-298. A seminal work in the field is in the context of sepsis where parasympathetic signaling through the vagus nerve was shown to limit TNF release by splenic macrophages, suppressing systemic inflammation and limiting shock299,300.

135

Several studies have found that neurotransmitters and neuropeptides play a critical role in regulating innate lymphoid cell (ILC) function in the lung and gut298,301-308. The neuropeptide neuromedin U (NMU) was found to support ILC2 effector function, including IL-5 and IL-13 release, and protection from parasitic worm infections298,301.

Similarly, NMU synergizes with IL-25 and IL-33 alarmins to promote ILC2 cytokine production and exacerbation of allergic inflammation in the lung307. Conversely, norepinephrine acting on ILC2 expressed β2-adrenergic receptor (2AR) inhibits ILC2 cytokine release, dampening the response to Nippostrongylus in mice302. Calcitonin gene- related peptide (CGRP), similarly dampens ILC2 responses to airway alarmins306, but in the context of worm infection, CGRP skews ILC2 cytokine production in response to alarmins and NMU, selectively promoting IL-5 release303.

CGRP, released by pain sensing neurons, or nociceptors, has further been shown to regulate several other innate immune cells in several contexts. During Staphylococcus aureus pneumonia, lung nociceptors release CGRP, reducing lung neutrophil and  T cell recruitment and limiting bacterial clearance and survival309. Similarly, Streptococcus pyogenes releases streptolysin S toxin during skin infections, triggering CGRP release and suppression of neutrophil recruitment310. Both cases show that CGRP acts to inhibit inflammation and indicate that its blockade can be used therapeutically as it enhances bacterial clearance and survival. Further, these data suggest that bacterial species may have evolved strategies to take advantage of the anti-inflammatory nature of CGRP and promote its release for immune evasion. Interestingly, nociceptor-derived CGRP in the small intestine promotes segmentous filamentous bacteria growth and protects from Salmonella

136

infection311, indicating that neuronal regulation of immunity is complex and that context- specific considerations are necessary to understanding its nuances.

Although relatively less is known about similar interactions between neurons and the adaptive immune system (B and T cells), both functional and molecular evidence for regulation of adaptive immunity by neurotransmitters has accumulated in recent years290,312,313. Indeed, subsequent studies on the effects of vagotomy on controlling innate inflammation implicate that neural activity might directly act on T cells – which then regulate the splenic macrophage response314,315.

5.1.2. NR classes and signaling modalities

By definition, a neurotransmitter is any molecule that is released from a neuron, typically at the axon terminal, which acts on another cell, i.e. a molecule that mediates communication from neurons. The central and peripheral nervous systems (CNS, PNS) employ a broad range of chemical types to carry out this cell-to-cell communication, including amino acids and their derivatives, peptides, lipids, purines, and other small molecules like acetylcholine and nitric oxide. In order to mediate communication, the NT necessarily acts on a receptor expressed on the recipient cell. NT receptors (NR) are defined by the specific NT that they bind, i.e. the NT acetylcholine (ACh) binds acetylcholine receptors (AChR); however, one NT can bind up to dozens of different receptor subtypes (Table 5.1). In the case of ACh, there are approximately 17 different

AChR produced in combination by 21 different genes316,317. The immense diversity of

NT and NR with differences in ligand binding, signaling, and regulation allows for incredible complexity as well as specificity within this communication scheme.

137

Despite the tremendous diversity of NR, the signaling modalities initiated by receptor signaling are surprisingly limited. All NR fall into one of two categories: 7- transmembrane G-protein coupled receptors (GPCR) or ligand-gated ion channels, described as “metabotropic” and “ionotropic,” respectively. Each class will be reviewed briefly.

GPCR

As the name suggests, GPCR associate with heterotrimeric G proteins, which share a common mechanism to initiate signal transduction. G proteins consist of 3 subunits, G, which has GTPase function and when active can dissociate from the trimeric complex, and G and G, which remain bound to one another (G) and are lipidated to remain at the surface membrane. At rest, G is bound to GDP and the trimeric protein is associated with the GPCR. Upon ligand binding, a conformational change triggers exchange of GDP for GTP, allowing G to dissociate from the receptor complex. Then, active G acts to induce or inhibit the generation of secondary messengers dictated by the particular G protein. GTP is eventually hydrolyzed to GDP and the GPCR and trimeric G protein re-associate with G at the membrane.

As just mentioned, different G proteins mediate different signaling based on the class of G subunit. While many G proteins have been described, three are most common, particularly among NR: 1) Gs, which acts to activate adenylate cyclase (AC) and the production of the second messenger cAMP, 2) Gi, which inhibits AC but can also mediate other effects in different cell types, and 3) Gq, which activates PLC to generate IP3 and DAG (Fig. 5.1). The role of G canonically was to promote GTP hydrolysis and reassembly of the G protein trimer; however, the dimer has been shown to

138

have diverse signaling capabilities complicated by the cell type and particular isoform of each subunit318. While these proximal signaling events can generally be predicted for a given GPCR based on the type of G protein coupled to it, the downstream effects of signaling will, of course, be cell type-dependent. This is further complicated by the discovery that complexities in receptor-ligand interactions can lead to biased signaling319.

That being said, within the nervous system, the overriding effect of GPCR signaling is dictated by the stereotypical G signal transduction320 (Fig. 5.1).

Ion channels

Ionotropic receptors act, rather simply as channels that permit the flux of ions following ligation with the neurotransmitter agonist. Functional channels are generated by combining a variable numbers of subunits depending on the receptor class – 3

(purinergic receptors), 4 (glutamatergic receptors), or 5 (cholinergic, serotonergic,

GABAergic, glycinergic receptors) – that form a water filled pore across the lipid bilayer321-326. In the unbound state, the pore is occluded by hydrophobic residues; however, upon ligand binding, a conformational change allows ions to permeate the membrane. Most ionotropic receptors are permissive to the flux of cations Na+, K+, and

Ca2+, allowing for membrane depolarization and excitation of an action potential in neurons or muscle cells325,326. Generally, cationic channels are non-specific for these ion species, but variation in subunit composition can modify permissiveness for a given ion,

2+ 324 typically regulating Ca flux . Ionotropic GABAA receptors as well as a class of glycine receptors, instead, are Cl- channels, which stabilizes membrane potential and are, this, inhibitory receptors (Fig. 5.1).

139

Table 5.1. Overview of neurotransmitters and their receptors. This table provides a summary of the major neurotransmitters, the different classes of receptors they each bind, the proximal signaling modalities triggered, and their role in the CNS. Note: this both an abbreviated list of neurotransmitters and receptors. Missing NT include neuropeptides (substance P, NPY, VIP, enkephalin, somatostatin, etc.), lipids (2-AG, AEA, etc.), and purines (adenosine, ATP, NAD, etc.). Beyond the receptors for the missing NT, subtypes of receptors have been condensed, such as the nicotinic AChR that form 12 different combinations of ion channels from 16 gene products, or the -adrenergic receptors that have 3 subtypes each. Reprinted with permission from: https://basicmedicalkey.com/introduction-to-central-nervous-system-pharmacology-2/

140

Figure 5.1. Summary of NR signaling modalities. (A) Overview of GPCR signaling for each of the stereotypical G subunits. Note the second messenger molecules elicited by each in blue text. Reprinted with permission from Huang, et al. 2009327, license number: 4893291415606. (B) Diagram of a nicotinic AChR (nAChR), providing an example of the ionotropic NR class. The receptor is made up of 5 subunits forming a central pore that, when bound the ACh, allows for the inward flux of Na+ and Ca2+ and the outward flux of K+. Reprinted with permission from Wonnacott, S. “Nicotinic ACh Receptors,” Tocris Scientific Review Series: https://resources.tocris.com/pdfs/literature/reviews/nicotinic-review-2019-web.pdf

141

5.1.3. Proposed functional impacts of NR signaling on T cell biology

Among the assortment of adaptive immune cells available to be regulated by neurons, the helper T cell is especially significant. Helper CD4 T cells typically play a central role in not only initiating an adaptive response (by helping CD8 T cells, B cells etc.) but also in channeling the response after themselves differentiating into different flavors of effector (e.g. TH1, TH2, TH17) or regulatory (Treg) cells. It is also known that

CD4 T cells express NRs and/or respond to pharmacological agonists and neurotransmitters312,313,328-334. A canonical receptor that has been extensively studied in this context is the β2AR which is expressed on T cells and poised to receive sympathetic input from the nervous system335-342. β2AR agonism affects the polarization of CD4 T cells and

330,339 promotes IFNγ release by TH1 primed cells in an IL-12 dependent manner ; however, receptor ligation has, conversely, been shown to inhibit IFNγ and TNF production and cytolytic activity of CD8+ T cells343. In line with a suppressive function for 2AR, norepinephrine signals enhance the activity of regulatory T cells341. 2AR signaling has also been implicated in the regulation of circulation by promoting retention of T cells in

LN344, and is a key intermediate in the vagal nerve inhibition of macrophage TNF secretion discussed above314. Understanding the contexts under which all of these effects can occur as well as the underlying signaling mechanisms that drive them requires further investigation.

5.1.4. Limitations of the prior literature and the necessity of its reexamination

While these data suggest a direct effort by neurons to regulate peripheral T cell differentiation, the same NR has also been implicated in the development and migration of

T cells 345-347. In addition to these pleiotropic effects on different subsets or developmental

142

stages of T cells, adrenergic stimulation can also impact T cells indirectly – because of their effects on innate immune cells. β2AR receptors are expressed on DCs, Innate

Lymphoid cells (ILC) etc. leading to sympathetic control of many innate immune functions302,348,349. This complex interplay of signals acting on different cells of the immune system, just from a single agonist-NR combination, highlights the challenges inherent in deciphering how and when neuronal stimulation can impact acute and chronic immune responses. Innate cells (such as DC) are critical not only for T cell activation but also for the programs of differentiation and migration that follow for days later. Indeed subtle changes in the function of such innate cells can (indirectly) affect T cell fate quite substantially350,351 – even when T cells are not directly regulated by neural stimuli. This distinction can be important in several areas relevant to understanding the mechanics of neuro–immune communication and eventually to the development of therapeutic approaches that have antigen-specific versus broad-acting effects.

The mechanistic distinction between a direct effect of neurotransmitters on T cells vs indirect regulation of T cell biology via effects on innate cells can lead to fundamentally different models of the role of nervous-immune communication. The key difference is whether or not a new signaling paradigm (from the nervous system to the immune cell) is required to understand the regulatory process operating there. If the effect on T cell fate is indirect, then the NR signaling in T cells is not likely relevant. Once we understand the effect on the innate cell, it should be trivial to extrapolate from that onto what would be the impact on T cells. For instance, the amounts of cytokines (e.g. IL-12, TNF, IL-6 etc.) or costimulatory molecules (e.g. B7, CD40 etc.) expressed by DC have reasonably predictable consequences on T cells. A direct effect on T cells, however, raises important questions on

143

how the NR signals affect T cells. Since different NRs (even for related neurotransmitters) can transmit quite distinct signals, the first challenge is to understand which of the 185 possible mammalian NR genes can be expressed by T cells. Ideally, the expression data should come from a pure population of T cells (to categorically rule out contamination by innate cells) and incorporate changes that happen to the T cell transcriptome following activation and differentiation to an antigenic stimulation. Further, many of the studies describing functional consequences of NR signaling through 2AR used whole animal KO or in vivo agonist/antagonist experimental designs, which cannot distinguish between direct and indirect consequences of adrenergic signaling. A careful approach using purified

T cell populations is necessary to truly parse out the role for NR signaling in modulating T cell behavior.

5.2. Results

5.2.1. Peripheral T cells express a limited subset of Neurotransmitter receptors (NR)

There are currently 185 transcripts reported in the NCBI gene database to encode

NRs, which are expressed in a variety of tissues (primarily neuronal). In order to evaluate which ones are expressed in T cells, we took two parallel approaches – one using a bioinformatics strategy to mine publicly available gene expression data for different T cell subpopulations and the other to validate these signatures using a combination of FACS sorting and mRNA isolation. In terms of developing an NR-expression map for T cells, a key concern is the purity of the cells used for analysis of the transcripts, since a small number of other cell types could easily impact our interpretation. In this context, one of the best curated datasets available for bioinformatic analysis come from the Immunological

Genome Project (ImmGen) – a multicenter collaboration that utilizes highly standardized

144

protocols to generate genome-wide microarray-based expression data from FACS-purified mouse immune cell populations across a broad array of developmental stages, functional states, and tissue sources. We first extracted the ImmGen data for naïve (CD44lo) and memory (CD44hi) CD4 and CD8 T cell populations from the spleen and subcutaneous lymph nodes (scLN) of B6 mice, as well as splenic CD25+Foxp3+ T Regulatory cells

(Treg), to probe for expression of the 185 NR transcripts. As with any transcriptomic analysis, the precise signal detection cutoffs used to identify expressed or non-expressed genes is critical. Adapting from the ImmGen recommendations, we used two different cutoff values of 40 and 100 (see Chapter 2). Using a value of 100, we find that mouse T cells, irrespective of lineage or activation status, could express only 26 of the possible 185

NR genes (Summarized in Fig. 5.2A). The expression of these 26 genes was also different in each subset of T cells – i.e. naïve, memory, Treg etc. (Fig. 5.2A). Nearly half of these

(12 of 26) were expressed by all the T cell subsets examined – but the remaining had an intricate pattern of subset specific expression (Fig. 5.2A-C; Tables 5.2-5.3). These include previously reported NRs – albeit whose T-cell-subset-specific quantitative differences were not necessarily appreciated before. Some of these are consistent with published literature 328,341,346,352,353. Among CD4 T cells, naïve cells show high expression of the 2 adrenergic receptor Adrb2. This is downregulated in T cells with an activated phenotype

(~3-fold and 1.67-fold in CD4 memory and Treg) cells. This differential expression is not apparent in CD8 cells as both naïve and memory populations show intermediate expression of this receptor. Another receptor which shows a subset-dependent pattern of transcription is the VIP receptor Vipr1. In this case memory T cells show nearly 4-fold downregulation of the receptor relative to naïve T cells and expression drops to below the cutoff levels

145

among Treg. The expression of the cannabinoid receptor Cnr2 is enriched in CD8 cells compared to CD4 - with greater expression among naïve T cells than memory. Among purinergic receptors, the adenosine receptor Adora2a shows enrichment in both CD4 and

CD8 memory populations as well as Treg. The ATP-gated ion channel P2rx7 also shows upregulation among CD4 memory and Treg but its expression is absent in CD8 cells. The metabotropic lysophosphatidic acid receptor Lpar6 (also known as P2ry5) is highly expressed in both lineages but it is enriched among scLN derived populations compared with splenic cells (Fig. 5.2B).

146

Figure 5.2. T cells express a limited diversity of neurotransmitter receptors in a subset-specific fashion. (A) Overview of NR expression by different T cell subsets, summarized from the analysis of ImmGen datasets discussed below. Numbers in brackets next to each subset label indicates the total number NRs found in each T cell fraction, above the ImmGen threshold of expression. Note, Treg and effector CD8 T cell populations were not included in this diagram. (B-C, next page) ImmGen datasets of mRNA levels in different T cell subsets were retrieved and analyzed for NR expression using recommended statistical criteria. Normalized signal intensity of NR genes expressed by CD4 (B) and CD8 (C) T cell populations that are either naïve (“n” - filled circles and squares) or memory (“m” – open circles and squares) are shown, T cells were either from the subcutaneous lymph nodes (“scLN” - circles) or spleen (“sp” - squares). In addition, T cells from the spleens of a reporter strain expressing GFP in Treg (Foxp3-GFP) was used by ImmGen to analyze gene expression in naïve CD4+ cells (filled diamonds, cells negative for Foxp3 expression) or Foxp3+ CD4 Treg (open diamonds). Dotted line represents expression cutoff of 100 (see Methods). All data points represent average signal intensity from 2-4 biological replicates. Statistical comparisons are included in Tables 5.4 and 5.5. Reprinted with permission from Rosenberg & Singh, 2019354

147

148

Table 5.2. NR expression by T cells using restrictive thresholding (100) of ImmGen data Reprinted with permission from Rosenberg & Singh, 2019354

Lymph node Lymph node Spleen Spleen (subcutaneous) (subcutaneous) CD4+ Gene CD4+ CD4+ CD4+ CD4+ CD8+ CD8+ CD8+ CD8+ naïve T Symbol naïve memory naïve memory reg naïve memory naïve memory (Foxp3-) Adora2a 110.03 158.75 126.99 272.24 172.58 159.51 103.86 200.30 Adra1d 100.71 107.51 104.57 123.56 129.34 145.86 108.46 Adrb1 100.32 123.44 Adrb2 360.41 135.28 398.67 124.81 342.76 207.50 260.13 178.80 225.69 215.17 Agtr1b 106.01 114.36 Avpr2 100.16 104.17 106.93 100.94 113.89 Bdkrb2 105.38 119.38 106.74 Chrm1 109.22 132.66 107.53 104.35 125.88 132.08 116.24 165.39 136.21 113.82 Cnr2 123.21 125.69 136.77 111.80 108.92 307.00 183.39 263.10 198.21 Drd4 111.51 109.14 113.99 132.55 100.75 Gabbr1 131.03 138.63 174.22 195.04 119.20 158.49 173.57 141.99 132.18 172.11 Gabrr2 102.77 Ghrhr 109.29 104.62 113.46 117.36 117.40 110.44 Grik5 102.36 124.18 135.41 Grm2 114.68 134.61 120.06 154.51 137.16 101.17 Hrh2 111.54 107.07 103.38 100.87 113.41 109.70 135.72 109.92 127.37 Lpar6 397.12 442.19 511.35 667.32 313.48 243.32 463.28 336.28 296.69 379.55 Ogfr 552.93 595.06 494.67 525.60 390.57 406.54 544.59 461.23 396.49 400.01 Oprd1 118.49 109.85 P2rx4 139.35 100.71 155.71 128.75 115.35 149.65 117.46 P2rx7 302.39 142.22 495.21 375.72 Ramp1 134.15 124.98 108.01 109.46 152.96 119.81 Sigmar1 259.18 170.52 355.98 163.21 257.99 205.53 282.68 236.98 219.97 228.54 Tspo 456.25 321.36 466.26 352.79 348.52 287.30 404.88 243.49 330.97 311.13 Vipr1 615.95 224.36 508.15 135.55 470.45 326.40 249.52 273.61 296.09

149

Table 5.3. NR expression by T cells using inclusive thresholding (40) of ImmGen data. Reprinted with permission from Rosenberg & Singh, 2019354

Lymph node Lymph node Spleen Spleen (subcutaneous) (subcutaneous) CD4+ Gene CD4+ CD4+ CD4+ CD4+ CD8+ CD8+ CD8+ CD8+ naïve T Symbol naïve memory naïve memory reg naïve memory naïve memory (Foxp3-) Adora1 55.08 52.63 46.89 51.73 57.31 56.90 51.31 63.07 64.07 52.55 Adora2a 110.03 158.75 126.99 272.24 96.26 172.58 97.59 159.51 103.86 200.30 Adora2b 45.89 50.96 47.96 41.57 52.49 54.84 41.63 Adra1a 44.82 41.89 42.39 65.29 46.99 Adra1b 50.14 58.56 48.22 46.82 59.06 59.63 50.92 71.49 63.32 54.85 Adra1d 100.71 107.51 97.59 98.82 104.57 123.56 95.89 129.34 145.86 108.46 Adra2a 59.08 71.48 53.25 58.79 63.73 61.75 57.76 81.35 68.98 58.13 Adra2b 42.40 53.04 41.73 45.25 49.57 60.84 41.87 58.94 55.33 44.06 Adra2c 48.04 66.06 46.22 42.66 53.82 60.46 46.72 74.53 56.15 49.58 Adrb1 88.70 100.32 77.06 89.47 88.68 98.03 89.50 123.44 90.97 90.56 Adrb2 360.41 135.28 398.67 124.81 342.76 207.50 260.13 178.80 225.69 215.17 Agtr1a 40.57 42.53 45.60 42.37 Agtr1b 81.45 85.71 70.25 72.27 95.34 95.88 80.39 106.01 114.36 71.36 Avpr1a 42.95 42.64 42.84 41.06 Avpr2 100.16 88.61 99.78 95.79 104.17 106.93 98.26 85.93 100.94 113.89 Bdkrb1 50.63 57.76 49.83 52.63 48.21 61.04 47.67 66.74 51.50 53.38 Bdkrb2 99.31 105.38 74.81 92.28 80.06 97.31 76.72 119.38 106.74 79.71 Cckar 44.46 40.61 Chrm1 109.22 132.66 107.53 104.35 125.88 132.08 116.24 165.39 136.21 113.82 Chrm4 50.65 76.76 47.66 68.17 46.08 73.47 41.49 69.78 48.95 48.58 Chrm5 42.90 43.70 Chrna1 45.59 41.14 Chrna2 41.41 Chrna3 46.13 Chrna4 60.41 68.70 47.12 56.60 60.74 70.21 57.70 85.68 76.54 62.40 Chrna7 42.50 44.30 44.13 43.53 48.51 47.12 43.69 Chrna9 42.76 42.36 41.43 46.23 42.41 46.36 44.79 Chrnb1 53.47 57.92 47.35 46.87 52.09 60.31 56.50 66.97 72.17 54.27 Chrnb2 63.44 60.90 53.76 50.66 64.24 71.73 58.15 75.75 61.60 58.77 Chrnb4 66.03 75.61 62.76 65.25 73.14 73.85 64.84 87.07 98.86 62.78 Chrnd 52.29 43.94 43.76 46.60 47.98 48.67 41.64 Chrne 44.93 48.37 45.17 47.14 47.46 49.46 44.83 59.05 46.08 51.51 Chrng 59.05 57.41 51.17 53.24 65.70 65.29 58.03 76.94 83.61 59.44 Cnr2 123.21 92.86 125.69 136.77 111.80 108.92 307.00 183.39 263.10 198.21 Crhr1 61.62 71.23 56.80 59.32 71.22 69.91 62.27 72.84 63.04 64.75 Crhr2 40.09 42.63 42.79 44.98 54.73 44.22 Drd1 55.76 67.75 55.56 60.94 68.06 77.99 48.77 79.67 73.87 64.14 Drd2 44.01 48.70 45.79 Drd3 41.07 Drd4 98.06 111.51 85.36 85.39 109.14 113.99 94.86 132.55 100.75 93.86 Gabbr1 131.03 138.63 174.22 195.04 119.20 158.49 173.57 141.99 132.18 172.11 Gabbr2 45.59 51.00 42.19 43.36 48.12 44.37 44.73 62.88 50.28 45.95 Gabrb3 42.89 49.57 41.57 47.43 52.34 55.24 42.14 60.13 55.88 48.69 Gabrd 71.08 76.25 61.93 59.49 70.62 74.10 63.52 95.64 83.61 71.68 Gabrr2 102.77 77.76 90.24 64.77 97.20 84.83 81.47 86.93 98.85 81.75 Galr2 59.26 65.04 58.58 52.34 63.26 60.62 56.11 67.90 64.56 59.43 Gcgr 49.51 51.47 43.53 45.41 46.57 53.66 47.34 45.77 54.54 48.84 Ghrhr 99.12 109.29 96.99 104.62 96.21 113.46 87.89 117.36 117.40 110.44

150

Table 5.3. NR expression by T cells using inclusive thresholding (40) of ImmGen data (cont.).

Glp1r 40.14 43.36 40.39 54.83 44.77 Glp2r 46.31 40.29 Gria3 50.03 Grid1 40.46 40.27 48.59 Grik3 46.41 52.07 42.40 44.72 53.79 50.81 45.99 65.09 60.23 46.59 Grik4 42.98 Grik5 75.36 102.36 78.23 124.18 48.12 135.41 60.31 85.28 68.36 66.33 Grin1 49.70 54.60 43.42 49.14 49.66 44.20 45.12 65.02 58.35 46.19 Grin2b 43.88 Grin2c 47.76 42.11 47.53 46.16 51.99 40.03 57.67 44.23 46.85 Grin2d 52.06 54.23 46.40 47.89 50.62 52.04 48.53 55.97 60.23 50.28 Grin3a 40.53 41.77 44.48 Grin3b 81.01 83.18 74.65 76.54 88.61 84.67 79.23 91.67 91.91 71.59 Grm1 50.08 40.10 50.24 47.02 53.50 50.41 43.80 Grm2 98.96 114.68 84.69 95.54 134.61 120.06 93.88 154.51 137.16 101.17 Grm4 51.16 54.99 46.60 46.62 54.47 53.70 49.74 61.48 61.24 48.53 Grm5 41.68 40.42 46.86 40.52 43.00 Grm6 61.89 65.33 59.43 48.37 65.23 62.76 58.66 76.61 84.70 53.31 Hcrtr1 54.02 55.17 49.31 47.94 56.80 55.83 47.75 68.25 60.60 54.81 Hrh1 90.36 84.18 81.07 71.83 94.99 93.76 81.96 98.56 93.40 90.79 Hrh2 111.54 92.11 107.07 103.38 100.87 113.41 109.70 135.72 109.92 127.37 Hrh3 44.50 Htr1a 43.39 43.53 Htr1b 42.52 47.75 40.12 40.83 48.68 45.35 52.95 48.39 44.87 Htr1f 55.11 64.02 48.22 49.90 60.94 56.84 47.49 71.47 56.77 51.98 Htr2a 50.77 43.78 Htr3b 40.79 Htr4 42.34 48.11 45.45 42.78 46.81 43.78 51.39 53.39 46.51 Htr5a 48.08 45.95 42.35 Htr5b 48.43 44.31 Htr6 55.39 52.81 52.52 54.28 56.29 62.88 45.79 55.09 44.60 60.74 Lpar6 397.12 442.19 511.35 667.32 313.48 243.32 463.28 336.28 296.69 379.55 Nmur1 57.43 65.21 56.54 63.89 64.63 71.93 54.99 74.87 62.36 68.48 Ntsr1 42.08 43.93 47.89 40.64 Ntsr2 76.81 81.96 70.24 76.72 74.59 89.09 77.94 94.80 76.54 74.03 Ogfr 552.93 595.06 494.67 525.60 390.57 406.54 544.59 461.23 396.49 400.01 Oprd1 94.29 99.28 77.24 81.80 88.34 90.89 88.15 118.49 109.85 86.15 Oprm1 40.18 42.98 Oxtr 44.63 40.79 41.83 40.45 54.64 45.47 P2rx2 40.49 47.70 P2rx3 40.20 41.53 P2rx4 139.35 100.71 155.71 128.75 115.35 149.65 117.46 82.71 88.63 91.30 P2rx5 62.35 61.84 49.41 50.81 65.15 67.64 58.64 66.08 67.24 60.95 P2rx6 55.39 51.75 45.11 49.02 53.68 48.03 48.31 69.97 60.86 48.79 P2rx7 96.88 302.39 142.22 495.21 96.79 375.72 62.82 73.96 64.80 72.41 P2ry1 41.58 66.57 41.85 50.84 45.18 41.62 62.26 47.25 40.28 P2ry13 44.09 40.32 P2ry2 47.91 40.47 P2ry6 40.46 45.55 42.48 43.96 46.70 42.36 48.55 41.23 43.44 Prokr1 43.18 40.62 Prokr2 41.06 44.40 Ramp1 134.15 96.83 124.98 108.01 109.46 83.89 152.96 43.86 119.81 52.19 Sctr 70.02 73.36 66.75 58.92 63.86 78.34 63.35 73.44 76.56 65.88 Sigmar1 259.18 170.52 355.98 163.21 257.99 205.53 282.68 236.98 219.97 228.54

151

Table 5.3. NR expression by T cells using inclusive thresholding (40) of ImmGen data (cont.).

Sstr1 40.90 40.77 44.68 46.08 Sstr2 41.04 43.24 40.92 49.19 43.10 Sstr3 66.64 71.12 54.80 63.01 67.13 66.31 54.11 81.69 80.74 61.67 Sstr4 72.53 81.16 65.32 69.81 73.70 72.61 68.98 98.76 75.15 74.26 Sstr5 40.75 44.75 Tacr1 41.61 41.27 40.54 41.62 51.23 Tacr2 48.53 40.16 Trhr2 53.11 54.95 53.90 54.06 49.77 70.85 49.82 67.72 59.83 56.24 Tspo 456.25 321.36 466.26 352.79 348.52 287.30 404.88 243.49 330.97 311.13 Vipr1 615.95 224.36 508.15 135.55 470.45 68.05 326.40 249.52 273.61 296.09

152

Table 5.4. Statistical assessment of NR expression in T cells of ImmGen data. For each population listed, expression of the indicated gene was compared to Adad1 (a testes specific gene not expressed in T cells) using a Student’s T-test with resulting p-value displayed. p-values  0.05 are highlighted in grey. Only genes that met the threshold criteria of 100 (see Chapter 2) are included in the table (see Table 5.1 for normalized expression values). Reprinted with permission from Rosenberg & Singh, 2019354

Lymph node Lymph node Spleen Spleen (subcutaneous) (subcutaneous) CD4+ Gene CD4+ CD4+ CD4+ CD4+ CD8+ CD8+ CD8+ CD8+ naïve T Symbol naïve memory naïve memory reg naïve memory naïve memory (Foxp3-) Adora2a 0.0002 0.0058 0.0001 0.0001 0.1644 0.0264 0.0020 0.0009 Adra1d 1.23E-05 0.0026 0.0066 0.0688 0.0025 0.0615 0.0085 Adrb1 0.0050 0.0004 Adrb2 0.0095 0.0009 0.0018 0.0219 0.0054 0.0388 0.0055 0.0077 0.0071 0.0209 Agtr1b 9.73E-06 0.0632 Avpr2 0.0008 0.0096 0.0214 0.0035 0.0128 Bdkrb2 0.0023 0.0147 0.0314 Chrm1 0.0004 0.0015 0.0010 0.0028 0.0015 0.0204 0.0011 0.0083 0.0071 0.0083 Cnr2 0.0103 0.0017 0.0093 0.0196 0.0231 0.0061 0.0242 0.0031 0.0031 Drd4 0.0048 0.0006 0.1193 0.0006 0.0115 Gabbr1 0.0163 0.0331 0.0001 0.0182 0.0006 0.0264 0.0003 0.0028 0.0220 0.0099 Gabrr2 0.0001 Ghrhr 0.0034 0.0044 3.46E-06 0.0004 0.0007 0.0048 Grik5 0.0002 0.0008 0.0173 Grm2 0.0212 0.0050 0.1435 0.0009 0.0237 0.0013 Hrh2 0.0035 1.05E-05 0.0002 0.0019 0.1363 0.0016 0.0023 0.0111 0.0002 Lpar6 0.0033 0.0079 0.0001 0.0154 0.0415 0.0094 0.0091 0.0043 0.0451 0.0016 Ogfr 0.0004 0.0073 0.0005 0.0006 0.0061 0.0608 0.0117 0.0119 0.0044 0.0014 Oprd1 0.0080 0.0090 P2rx4 0.0325 0.0039 0.0001 0.0022 0.0123 0.0004 0.0002 P2rx7 0.0081 0.0002 0.0029 0.0217 Ramp1 0.0130 0.0003 0.0044 0.0062 0.0064 0.0086 Sigmar1 0.0058 0.0437 0.0003 0.0394 0.0000 0.0020 0.0015 0.0176 0.0061 0.0050 Tspo 0.0001 0.0290 0.0004 0.0290 0.0038 0.0029 0.0037 0.0122 0.0015 0.0177 Vipr1 0.0036 0.0009 0.0001 0.0048 0.0032 0.0027 0.0020 0.0005 0.0021

153

Table 5.5. Statistical comparison of NR expression between T cell subsets of ImmGen data. For the indicated tissue and population comparisons, differential gene expression was determined using a Student’s T-test. p-values < 0.05 are highlighted in grey. Genes for which at least one population met the threshold value of 100 (see Chapter 2 and Table 5.2) are displayed. Reprinted with permission from Rosenberg & Singh, 2019354

scLN Spleen scLN Spleen Gene CD4+ memory vs Foxp3+ CD8+ memory vs Symbol naïve vs Foxp3- naïve Adora2a 0.0371 4.63E-07 0.2977 0.0992 0.0006 Adra1d 0.3033 0.2809 0.0538 0.3502 Adrb1 0.2598 0.0400 Adrb2 0.0196 0.0021 0.0184 0.0262 0.7764 Agtr1b 0.0412 0.2039 Avpr2 0.4098 0.7909 0.3828 Bdkrb2 0.3837 0.0568 0.2265 Chrm1 0.0231 0.7178 0.4320 0.0552 0.2148 Cnr2 0.0953 0.4954 0.8471 0.0229 0.0382 Drd4 0.3430 0.8208 0.0426 0.6124 Gabbr1 0.7872 0.4753 0.0228 0.0283 0.1586 Gabrr2 0.0728 Ghrhr 0.2413 0.5877 0.1135 0.0184 0.4808 Grik5 0.0074 0.0003 0.0011 Grm2 0.3744 0.6214 0.0027 0.1898 Hrh2 0.0712 0.2179 0.6271 0.0349 0.2621 Lpar6 0.3714 0.1910 0.3678 0.0771 0.2946 Ogfr 0.4845 0.3784 0.7618 0.3293 0.9151 Oprd1 0.1101 0.1472 P2rx4 0.2091 0.0152 0.0883 0.0521 P2rx7 0.0145 0.0038 0.0019 Ramp1 0.0849 0.0959 0.1628 0.0047 0.0241 Sigmar1 0.0915 0.0072 4.32E-05 0.2506 0.7298 Tspo 0.1215 0.1773 0.0978 0.0092 0.6735 Vipr1 0.0066 2.70E-05 0.0039 0.0222 0.2487

154

5.2.2. Validation of NR expression in FACS-sorted T cells.

In order to independently validate the expression patterns we observed in the

ImmGen dataset, we then used two independent approaches (RTPCR and NanoString) to measure expression in Flow-Sorted primary T cells. We first FACS-purified CD4 naïve

(CD44lo), CD4 memory (CD44hi), Treg (CD25hi) and CD8 populations to >95% purity

(Fig. 5.3-5.4). RNA isolated from these cells was analyzed using the NanoString nCounter system (see Table 5.6). Comparison of naïve CD4 and CD8 T cells (Fig. 5.5A-B) shows that of the 21 NR transcripts detected, 15 were less than 2-fold different between the two populations. Consistent with ImmGen, there were small but statistically significant increases in Cnr2 expression (2.3-fold, p=7e-5) by naïve CD8 T cells while naïve CD4 T cells had a 1.6-fold (p=5e-5) higher level of Adrb2 and 2.3-fold (p=0.0003) higher Vipr1.

A more substantial difference was observed for the ionotropic glutamate receptor Gria3 among the CD8 population (13.7-fold, p=0.0008) apparent using NanoString analysis (Fig.

5.4A-B). Although CD8 expression of Gria3 has been previously reported355,356 this degree of difference was not initially flagged in the ImmGen data (Fig. 5.2C). So we examined if this was because of the increased sensitivity of NanoString method compared to the microarrays used by ImmGen. Consistent with this hypothesis, expression of Gria3 is found to be highly differential among CD8 naïve cells even in the ImmGen dataset when a more inclusive threshold of 47 is applied (Table 5.1, Chapter 2).

The differences between naïve and memory CD4 T cells in terms of NR expression as assayed by NanoString also validate the ImmGen profile (Fig. 5.5C-D). An upregulation of Adora2a (2.5-fold, p=0.0026) and P2rx7 (7.0-fold, p=0.0003) by CD4 memory cells compared with naïve as well as significant downregulation of Adrb2 (2.4-fold, p=5e-5) and

155

Vipr1 (4.4-fold, p=6e-5) receptors is notable. Interestingly, this analysis identified the histamine receptor Hrh4 and purinergic receptor P2ry1 as being uniquely expressed by the

CD4 memory T cell population. Significant downregulation (5.5-fold, p=0.0049) of

Gabrr2 encoding the GABAA-ρ2 receptor and the significant enrichment of the muscarinic acetylcholine receptor Chrm4 among CD4 memory (5.9-fold, p=0.016) and Treg (3.8-fold, p=0.037; Fig. 5.5E-F) populations was also observed. Also, consistent with ImmGen data,

Treg cells (Fig. 5.5E-F) uniquely downregulated Vipr1 27-fold (p=0.0014). Finally, these trends were further confirmed using RT-qPCR (Fig. 5.6). Utilizing these two techniques allowed us to provide robust validation for the NR expression patterns of naïve, memory, and regulatory T cells determined by ImmGen but also to identify some expressed receptors that may have eluded detection in the microarray analyses (Fig. 5.2).

156

Figure 5.3. FACS gating strategy for isolating T cell subsets. Lymph nodes and spleens from 4 B6 mice were collected and pooled. Live, mononuclear cells were separated using ficoll density centrifugation. T cells were enriched using magnetic, negative selection with Dynabeads targeting CD11b, B220, NK1.1, and MHCII. Enriched T cells were then FACS-purified into the four subsets below. The gating strategy for FACS from a representative sample is shown. Collected fractions were 1) CD8+, 2) CD4+CD25hi, 3) CD4+CD25loCD44hi, and 4) CD4+CD25loCD44lo. Reprinted with permission from Rosenberg & Singh, 2019354

157

Figure 5.4. Post-sort purities of isolated T cell populations. Four T cell populations were FACS-purified from B6 lymph node and spleens as described in Figure 5.3. Post-sort purities from a representative FACS sample are shown. Collected fractions were 1) CD8+, 2) CD4+CD25hi, 3) CD4+CD25loCD44hi, and 4) CD4+CD25loCD44lo. Reprinted with permission from Rosenberg & Singh, 2019354

158

A B

C D

E F

Figure 5.5. Quantitative comparison of NR expression by flow-sorted primary T cell subsets. Cells from four B6 mice were pooled and FACS purified into the indicated populations as described in Figure 5.3. Volcano (A,C,E) and scatterplots (B,D,F) representing transcript levels of NR genes in sorted T cell populations measured by NanoString RNA count analysis. Fold change of CD8 (A,B), CD4 memory (C,D), and Treg (E,F) populations as compared to CD4 naïve. Data includes 3 biological replicates. Red and blue lines (B,D,F) indicate 2-fold enrichment in respective populations. Reprinted with permission from Rosenberg & Singh, 2019354

159

Figure 5.6. NR expression differences between T cell subsets are verified by qPCR. Cells from four B6 mice were pooled and FACS purified into the indicated populations as described in Figure 5.3. Relative transcript levels of NR genes in sorted T cell populations by RT-qPCR using the Qiagen RT2 Profiler PCR Array Mouse Neurotransmitter Receptors kit, normalized to house-keeping genes (Actb, B2m, Gapdh, Gusb, Hsp90ab1; geometric mean Ct for all 5 genes). Displayed as mean ± SD. Data includes 4 biological replicates. * p<0.05 using Student’s T-test. Reprinted with permission from Rosenberg & Singh, 2019354

160

Table 5.6. Complete NanoString gene list and probe details.

Gene Name Accession Target Position Tot Isoforms # Hit Isoforms Not Hit By Probe Off Target 1 Adora1 NM_001008533.3 1604-1703 6 6 2 Adora2a NM_009630.2 2307-2406 5 5 3 Adora2b NM_007413.4 1095-1194 1 1 4 Adora3 NM_009631.4 574-673 1 1 5 Adra1a NM_001271760.1 477-576 13 13 6 Adra1b NM_007416.3 2666-2765 4 4 7 Adra1d NM_013460.4 2002-2101 1 1 8 Adra2a NM_007417.4 3595-3694 1 1 9 Adra2b NM_009633.3 3473-3572 1 1 10 Adra2c NM_007418.3 2539-2638 1 1 11 Adrb1 NM_007419.2 1962-2061 1 1 12 Adrb2 NM_007420.2 681-780 1 1 13 Adrb3 NM_013462.3 977-1076 6 6 14 Agtr1a NM_177322.3 1668-1767 3 3 15 Agtr2 NM_007429.4 986-1085 2 2 16 Avpr1a NM_016847.2 1936-2035 1 1 17 Avpr1b NM_011924.2 3333-3432 1 1 18 Avpr2 NM_001276298.1 1191-1290 5 5 19 Bdkrb1 NM_007539.2 11-110 1 1 20 Bdkrb2 NM_009747.2 165-264 3 3 21 Brs3 NM_009766.3 9-108 1 1 22 Calcr NM_001042725.1 806-905 4 4 23 Cckar NM_009827.2 2266-2365 2 2 24 Cckbr NM_007627.4 1815-1914 3 3 25 Chrm1 NM_007698.3 4146-4245 2 2 26 Chrm2 NM_203491.3 616-715 2 2 27 Chrm3 NM_033269.2 1791-1890 2 2 28 Chrm4 NM_007699.2 733-832 3 3 29 Chrm5 NM_205783.2 1646-1745 2 2 30 Chrna1 NM_007389.4 3771-3870 1 1 31 Chrna10 NM_001081424.1 541-640 3 3 32 Chrna2 NM_144803.2 1575-1674 4 4 33 Chrna3 NM_145129.2 566-665 1 1 34 Chrna4 NM_015730.5 2876-2975 2 2 35 Chrna5 NM_176844.4 1283-1382 2 2 36 Chrna6 NM_021369.2 2161-2260 1 1 37 Chrna7 NM_007390.3 336-435 1 1 38 Chrna9 NM_001081104.1 386-485 7 4 XR_868285.1;XR_376992.2;XR_868283.2 39 Chrnb1 NM_009601.4 657-756 4 4 40 Chrnb2 NM_009602.4 4471-4570 1 1 41 Chrnb3 NM_027454.4 1417-1516 2 2 42 Chrnb4 NM_148944.4 2036-2135 3 3 43 Chrnd NM_021600.2 2341-2440 1 1 44 Chrne NM_009603.1 931-1030 3 3 45 Chrng NM_009604.3 231-330 2 2 46 Cnr1 NM_007726.3 2206-2305 7 7 47 Cnr2 NM_009924.3 2651-2750 2 2 48 Crhr1 NM_007762.4 1749-1848 8 7 XR_001779871.1 49 Crhr2 NM_009953.3 2231-2330 5 5 50 Drd1 NM_010076.3 1786-1885 2 2 51 Drd2 NM_010077.2 631-730 2 2 52 Drd3 NM_007877.1 706-805 2 2 53 Drd4 NM_007878.2 1216-1315 3 3 54 Drd5 NM_013503.2 1761-1860 1 1 55 Gabbr1 NM_019439.3 1367-1466 3 3 56 Gabbr2 NM_001081141.1 2291-2390 1 1 57 Gabra1 NM_010250.4 906-1005 5 5 58 Gabra2 NM_008066.3 1783-1882 2 2 59 Gabra3 NM_008067.4 2146-2245 4 4 60 Gabra4 NM_010251.2 1121-1220 3 3 61 Gabra5 NM_176942.4 2443-2542 6 5 XM_006540557.3 62 Gabra6 NM_001099641.1 1343-1442 4 4 63 Gabrb1 NM_008069.4 1631-1730 2 2 64 Gabrb2 NM_008070.3 4336-4435 4 4 65 Gabrb3 NM_008071.3 4201-4300 2 2 66 Gabrd NM_008072.2 476-575 2 2 67 Gabre NM_017369.2 1621-1720 6 3 XR_878092.1;XM_011247523.1;XM_011247526.2 68 Gabrg1 NM_010252.4 386-485 4 4 69 Gabrg2 NM_177408.5 1614-1713 4 4 70 Gabrg3 NM_008074.2 1445-1544 2 2 71 Gabrp NM_146017.3 485-584 2 2 72 Gabrq NM_020488.1 866-965 4 4 73 Gabrr1 NM_008075.2 1835-1934 3 3 74 Gabrr2 NM_008076.3 771-870 3 3 75 Gabrr3 NM_001081190.1 636-735 1 1 76 Galr1 NM_008082.2 2331-2430 1 1 77 Galr2 NM_010254.4 1721-1820 4 1 XM_011248728.2;XM_011248727.2;XM_011248729.2 78 Galr3 NM_015738.2 839-938 1 1 79 Gcgr NM_008101.2 1311-1410 10 10 80 Ghrhr NM_001003685.3 933-1032 1 1 161

Table 5.6. Complete NanoString gene list and probe details (cont.).

81 Glp1r NM_021332.2 408-507 1 1 82 Glp2r NM_175681.3 1109-1208 3 3 83 Glra1 NM_020492.3 623-722 2 2 84 Glra2 NM_183427.4 2291-2390 5 5 85 Glra3 NM_080438.2 757-856 2 2 86 Glrb NM_010298.5 705-804 17 17 87 Gnrhr NM_010323.2 313-412 4 4 88 Gria1 NM_001252403.1 2477-2576 6 6 89 Gria2 NM_001039195.1 301-400 21 21 90 Gria3 NM_016886.3 391-490 9 9 91 Gria4 NM_001113180.1 1275-1374 16 16 92 Grid1 NM_008166.2 1437-1536 5 5 93 Grid2 NM_008167.2 661-760 2 2 94 Grik1 NM_010348.3 1336-1435 13 13 95 Grik2 NM_010349.2 257-356 13 13 96 Grik3 NM_001081097.2 6365-6464 3 3 97 Grik4 NM_175481.5 959-1058 8 8 98 Grik5 NM_008168.2 2773-2872 4 4 99 Grin1 NM_008169.2 493-592 8 8 100 Grin2a NM_008170.2 1789-1888 3 3 101 Grin2b NM_008171.3 6341-6440 7 7 102 Grin2c NM_010350.2 2409-2508 14 14 103 Grin2d NM_008172.2 1202-1301 10 10 104 Grin3a NM_001033351.1 1333-1432 4 4 105 Grin3b NM_130455.2 2031-2130 5 5 106 Grm1 NM_001114333.2 2126-2225 6 6 107 Grm2 NM_001160353.1 2771-2870 1 1 108 Grm3 NM_181850.2 2526-2625 1 1 109 Grm4 NM_001013385.1 823-922 8 6 XM_006524320.3;XM_017317482.1 110 Grm5 NM_001143834.1 4243-4342 4 4 111 Grm6 NM_173372.2 3451-3550 4 4 112 Grm7 NM_177328.3 1296-1395 13 13 113 Grm8 NM_008174.2 1771-1870 8 8 114 Grpr NM_008177.2 305-404 1 1 115 Hcrtr1 NM_001163027.1 1721-1820 6 6 116 Hcrtr2 NM_198962.3 686-785 1 1 117 Hrh1 NM_001252642.1 1453-1552 7 7 118 Hrh2 NM_001010973.2 1245-1344 6 6 119 Hrh3 NM_133849.3 833-932 3 3 120 Hrh4 NM_153087.2 435-534 3 2 XM_017317874.1 121 Htr1a NM_008308.4 3171-3270 1 1 122 Htr1b NM_010482.1 626-725 1 1 123 Htr1d NM_008309.5 2499-2598 7 7 124 Htr1f NM_008310.3 1006-1105 1 1 125 Htr2a NM_172812.2 743-842 1 1 126 Htr2b NM_008311.2 1057-1156 4 4 127 Htr2c NM_008312.2 1991-2090 2 2 128 Htr3a NM_013561.2 523-622 3 3 129 Htr3b NM_020274.4 1091-1190 1 1 130 Htr4 NM_008313.4 163-262 9 9 131 Htr5a NM_008314.2 1697-1796 2 2 132 Htr6 NM_021358.2 565-664 2 2 133 Htr7 NM_008315.2 2306-2405 5 5 134 Lpar4 NM_175271.4 677-776 1 1 135 Lpar6 NM_175116.4 731-830 1 1 136 Nmbr NM_008703.2 865-964 1 1 137 Nmur1 NM_010341.1 955-1054 2 2 138 Nmur2 NM_153079.4 1533-1632 1 1 139 Npy1r NM_010934.4 311-410 5 5 140 Npy2r NM_001205099.1 591-690 4 4 141 Npy4r NM_008919.4 111-210 1 1 142 Npy5r NM_016708.3 725-824 3 3 143 Ntsr1 NM_018766.2 1123-1222 1 1 144 Ntsr2 NM_008747.2 1258-1357 4 4 145 Ogfr NM_031373.3 2141-2240 3 3 146 Oprd1 NM_013622.3 1407-1506 2 2 147 Oprk1 NM_001204371.1 721-820 4 4 148 Oprm1 NM_001039652.1 1197-1296 15 14 XM_017313829.1 149 Oxtr NM_001081147.1 3195-3294 2 2 150 P2rx1 NM_008771.3 1047-1146 2 2 151 P2rx2 NM_001164833.1 1435-1534 9 9 152 P2rx3 NM_145526.2 2005-2104 6 6 153 P2rx4 NM_011026.2 1656-1755 4 4 154 P2rx5 NM_033321.3 1131-1230 5 5 155 P2rx6 NM_011028.2 465-564 4 4 156 P2rx7 NM_001038839.2 379-478 5 5 157 P2ry1 NM_008772.4 921-1020 2 2 158 P2ry12 NM_027571.3 440-539 4 4 159 P2ry13 NM_028808.3 476-575 1 1 160 P2ry14 NM_001008497.2 493-592 8 8

162

Table 5.6. Complete NanoString gene list and probe details (cont.).

161 P2ry2 NM_008773.3 1051-1150 3 3 162 P2ry4 NM_020621.4 841-940 5 5 163 P2ry6 NM_183168.2 1406-1505 3 3 164 Prokr1 NM_021381.3 1811-1910 4 4 165 Prokr2 NM_144944.3 4187-4286 4 4 166 Sctr NM_001012322.2 1153-1252 12 12 167 Oprs1 NM_011014.3 973-1072 9 9 168 Sstr1 NM_009216.3 2046-2145 2 2 169 Sstr2 NM_009217.3 101-200 2 2 170 Sstr3 NM_009218.3 1829-1928 4 4 171 Sstr4 NM_009219.3 1205-1304 1 1 172 Sstr5 NM_001191008.1 407-506 2 2 173 Tacr1 NM_009313.5 3646-3745 4 4 174 Tacr2 NM_009314.4 19-118 1 1 175 Tacr3 NM_021382.6 931-1030 1 1 176 Trhr NM_013696.2 823-922 3 3 177 Tspo NM_009775.4 242-341 1 1 178 Vipr1 NM_011703.4 2481-2580 2 2 179 Vipr2 NM_009511.2 357-456 6 5 XM_017315044.1 HKG1 Actb NM_007393.1 816-915 1 1 also targets Lrrc58 (NM_177093) @97% HKG2 Cd3e NM_007648.4 381-480 2 2 also targets multiple predicted transcripts HKG3 Gapdh NM_008084.2 216-315 3 3 and pseudogenes at high identity also targets a RIKEN gene, E030024N20Rik (NR_033228) @ 98% HKG4 Ppia NM_008907.1 391-490 1 1 and a predicted pseudogene, Gm9234 (XR_869917) @ 96% HKG5 Rpl13a NM_009438.5 348-447 1 1 HKG6 Ubc NM_019639.4 22-121 1 1

163

The next question is whether these maps reflect protein expression, preferably at a single cell level – given the possibility that not all cells within each subset express all the receptors associated with their population-level signature. This is challenging since the reagents currently available for evaluating NR proteins have not all been extensively validated for quantitative techniques such as flow cytometry. Nevertheless, with the limited reagents currently available we were able to confirm that 7 of the core signature members are also expressed by T cells at the protein level (Figure 5.7).

164

A B C

D E F G

Figure 5.7. T cells NR at the protein level. Lymph nodes and spleens were collected from B6 or B10.A mice and enriched for T cells using two rounds of magnetic, negative selection as described in Figure 3.16. Confluent P6 Neuro-2a and P38 SH-SY5Y were collected from culture using trypsin. Enriched T cells and cell lines were then lysed using 2x lysis buffer. Brain, lung, and small intestine (SI) were collected from unperfused B6 mice, washed with normal saline, and underwent bead homogenization. Lysates were then prepared for western blot analysis. Western blots targeting Adrb2 (A), Adora2a (B), Gabrr2 (C), Vipr1 (D), Gria3 (E), Lpar6 (F), and Sigmar1 (G) encoded NR. Arrows or brackets indicate target band location. β-actin loading controls included below the respective lanes. Reprinted with permission from Rosenberg & Singh, 2019354

165

5.2.3. T cell expression of NR is dynamic with many showing rapid downregulation after activation.

Given that different T cell subsets already showed a bias in expression of different

NR transcripts, we further sought to analyze whether this pattern changes during T cell activation in vivo. Towards this, we used an ImmGen dataset in which ovalbumin (OVA)- specific CD8 OT-1 T cells were sorted from mice infected with OVA-expressing Listeria monocytogenes (Lis-OVA) at several time points post inoculation. In this case, the OT-1 cells are all homogenously naïve to start with and are synchronously activated by the inflammatory signals and antigen from the bacteria. Interestingly this strong activation did not seem to extend the number of NRs expressed by T cells that we saw in the steady state

T cell population – i.e. we did not identify an activation-induced NR pattern in peripheral

CD8 T cells. A subset of the genes we had already identified in CD8 T cells (12 of 26) did exhibit dynamic changes in expression during activation by the bacterial OVA, using a ≥

2-fold increase (Fig. 5.8A) or decrease (Fig. 5.8B) as the criteria. The adenosine receptor

Adora2a reaches a peak expression of 2.9-fold over that of the naïve at 6 days post infection

(dpi) and remains stably elevated through the memory phase. Conversely, Cnr2, Gria3, and Ramp1 show rapid downregulation within the first 24 hour of infection that remains low or absent (Fig. 5.8B). Multiple genes, including Adrb2, Avpr2, Gabbr1, Lpar6, and

Vipr1, show a similar kinetic pattern also, with early downregulation before gradually regaining receptor expression beginning during the effector phase. Interestingly, the endoplasmic reticular receptor Sigmar1 (also known as Oprs1) increases expression acutely during the first 48 hours of infection but shows downregulation during the effector phase 6-15 dpi before recovering to 90% of baseline expression during the memory phase

166

(Fig. 5.8B). Of note, the naïve and 106 dpi memory populations of the single OT-1 clone show remarkably consistent trends as compared to bulk CD8 populations (Fig. 5.2C).

Second, in order to evaluate this over a broad range of contexts, we identified datasets that profiled gene expression in FACS sorted T cells identified with different lineages and activation states. We extracted a dataset published by Doering et al 357 and

Crawford et al 143 where they examined LCMV-specific T cells from a polyclonal repertoire after acute viral infection (Fig. 5.8C-D). C57BL/6J mice were infected with

LCMV Armstrong and H2-Db GP33-specific CD8 T cells were sorted at the indicated times post infection using MHC-I tetramers. We found that the dynamics of NR expression in polyclonal CD8 T cells responding to the virus was largely consistent with expression dynamics exhibited by the transgenic OT-1 cells responding to a bacterium (Fig. 5.8A-B).

Interestingly, Adrb2 expression increased in response to LCMV infection at the day 15 and day 30 time points (Fig. 5.8D). Though, in the absence of later time points during the memory phase, it cannot be determined whether the expression decreases to the naïve baseline in viral specific cells as seen in bulk resting memory cells (Fig. 5.2B, 5.8C). This analysis suggests that the NR expression kinetics in CD4 T cells closely resemble those of

CD8 T cells (Fig. 5.8C-D). However, while CD8 cells downregulate Cnr2 and Gria3 following activation, CD4 cells show constitutively low expression of these receptors.

Conversely, while P2rx7 remains low in CD8 throughout the immune response, CD4 show strong induction of this receptor peaking at day 8 post infection (2.73-fold, p=0.0005) (Fig.

5.8C). These data, comparing different cell states and different pathogens suggest that most of the NR expression dynamics are relatively insensitive to many overarching contexts –

167

and are relatively intrinsic to the main cell types we initially identified – naïve, memory, and Treg.

168

A B

C D

Figure 5.8. Naïve T cells modulate NR expression following activation. (A,B) NR expression kinetics during T cell activation in vivo. ImmGen datasets of mRNA levels in different T cell subsets were retrieved and analyzed for NR expression using recommended statistical criteria. ImmGen analyzes the mRNA levels in the ova-specific OT-1 T cells in mice that are then infected with ova-expressing Listeria monocytogenes. We extracted the NR genes that show ≥2-fold increase (A) or decrease (B) compared to naïve cells (expression on the day of infection, day-0) are plotted. Dotted lines represent expression cutoffs of 100 and 40 (see Methods). Data points represent average signal intensity from 2-4 biological replicates. Statistical comparisons are included in Table 5.7. (C,D) Microarray data from Doering et al, in which tetramer positive CD4+ (C) and CD8+ (D) were collected from LCMV Armstrong infected animals at the indicated time points (N=3-4). BRB-ArrayTools time course analysis identified 8 NR genes that showed significant changes using FDR<0.05. Expression normalized to d0 time point. * p<0.05 using Student’s T-test comparing to d0 time point. Reprinted with permission from Rosenberg & Singh, 2019354

169

Table 5.7. Statistical analysis of NR expression dynamics within ImmGen data. For each time point (Fig. 5.8A-B), gene expression was compared to the naïve population using a Student’s T-test. p-values  0.05 are highlighted in grey. Reprinted with permission from Rosenberg & Singh, 2019354

Time post Adora2a Adrb2 Cnr2 Gabbr1 Gria3 Lpar6 P2rx4 P2ry13 P2ry14 Ramp1 Sigmar1 Vipr1 infection 12h 0.0004 0.0011 0.0059 0.0163 0.0001 0.0002 0.0608 0.0912 0.6664 0.0595 0.0150 0.0181 24hr 0.0153 0.0082 0.0029 0.0097 0.0004 0.0002 0.0168 0.0421 0.9875 0.0681 0.0046 0.0234 48hr 0.0437 0.0024 0.0033 0.0098 0.0001 0.0004 0.0200 0.7146 0.9656 0.0339 0.0021 0.0115 d6 0.0037 0.0012 0.0048 0.1471 0.0003 0.0025 0.0075 0.0224 0.1731 0.0311 0.0157 0.0146 d8 0.0106 0.0031 0.0017 0.2522 3.72E-06 0.0008 0.0023 0.0452 0.0745 0.0348 0.0041 0.0149 d10 0.0256 0.0017 0.0045 0.0380 2.26E-05 0.0079 0.0048 0.0280 0.0814 0.0264 0.0049 0.0139 d15 0.0161 0.0661 0.0052 0.8800 0.0001 0.0031 0.0029 0.0472 0.0597 0.0370 0.0059 0.0125 d45 0.2102 0.1803 0.0088 0.5636 0.0013 0.3071 0.0123 0.1037 0.1583 0.0278 0.3594 0.0229 d100 0.0629 0.1258 0.0031 0.7381 5.54E-06 0.1550 0.0056 0.0578 0.1506 0.0372 0.3559 0.0548

170

Typically, after naïve T cells are activated by antigen, they differentiate into effector cells which participate in the clearance of the pathogen358,359. Most of these are short-lived effector cells (SLEC) which die off quickly360. A few of these can give rise to long-term memory and are known as memory precursor effector cells (MPEC). Yu et al361 used the markers KLRG1 and IL-7R to identify these subsets in OT-1 T cells transferred to C57BL/6J mice infected with Lis-OVA 1 d following transfer. 8 days after infection, they FACS-sorted the OT-1 T cells into SLECs (KLRG1hi IL-7Rlo) and MPEC (KLRG1lo

IL-7Rhi). We extracted the data to analyze the differential abundance of NR in these subsets. Only three NR showed any differences between SLECs and MPECs (Fig. 5.9).

The authors’ primary analysis similarly identified few changes between the day 8 precursors. Although Vipr1 is downregulated in day 8 effector cells (Fig. 5.8B,D), CD8

MPEC show increased receptor expression (p=0.0288) compared to SLECs. Conversely

CD4 T cells, memory cells show significantly downregulated expression of Vipr1 (Fig.

5.2B, 5.5C,D), highlighting a potential distinction between the two lineages.

171

1 5 0

y S L E C

i s

n M P E C e

t 1 0 0

e t

u 5 0

b A

r 1 1 f r r g a p O i m V ig S

Figure 5.9. Effector T cells show virtually identical NR profiles. Microarray data from Yu et al361, in which OT-1 cells were collected on day 8 post-infection with Lis-OVA and sorted into short-lived effector cells (SLEC, KLRG1hi IL-7Rlo, N=2) and memory precursor effector cells (MPEC, KLRG1lo IL-7Rhi, N=3). BRB-ArrayTools class comparison analysis using a significance cutoff of p<0.05 identified 3 NR genes that differed between groups. * p<0.05 using Student’s T-test. Reprinted with permission from Rosenberg & Singh, 2019354

172

5.2.4. Tissue residency has minimal effect on the NR signature of activated T cells

In recent years, the important role of T cells that home to and perhaps even reside in many tissue sites has been appreciated362. These cells not only play key roles in protective immunity – but are also involved in homeostatic functions. We wondered whether such T cells would be subject to or influenced by localized NTs as a result of differential NR expression. To determine the role tissue residence in controlling NR expression in T cells we again mined existing data. In particular, we were interested in the regulation of NR expression within the brain parenchyma in which T cell exposure to NTs is increased. Microarray data is available from Wakim et al363, in which C57BL/6J mice containing transferred OT-1 CD8 T cells were sorted from the spleen and brain 20 days following infection with OVA-expressing vesicular stomatitis virus (VSV-OVA). CD103 expression was used to separate resident memory T cells (TRM, CD103+) from other

(CD103-) effector T cell populations 146. Despite minor, yet statistically significant differences in the expression of 12 NR genes, the NR profiles of splenic and brain-derived

T cell populations appeared to be overall similar (Fig. 5.10A) – in that most transcripts were within a 2-fold expression range in these tissues. The most differences observed were

(less than 2-fold) in the 2 adrenergic receptor Adrb2 and the opioid growth factor receptor

Ogfr that exhibited increased expression in both populations isolated from the brain, compared to the splenic T cells (Fig. 5.10A).

Our analysis was next expanded to several other tissue sites using datasets from

Mackay et al146, in which multiple mouse models were used to profile gene expression in lung, gut, and skin resident populations. The T cell populations are defined as follows:

“spleen TN” were collected from the spleens of naïve gBT-1 mice; “spleen TCM”

173

hi lo (CD8+CD62L ), “spleen TEM” (CD8+CD62L ), and “skin TRM” (CD103+) were gBT-1 cells collected from the indicated tissues 30 d following infection of C57BL/6J mice adoptively transferred with the transgenic cells with HSV (KOS strain); “lung TRM” were gBT-1 cells collected from the lung 30 d following infection of C57BL/6J mice adoptively transferred with the transgenic cells with influenza A/WSN expressing HSV glycoprotein

B; “gut TRM” were P14 cells collected from the small intestine 60 d following infection of

C57BL/6J mice adoptively transferred with the transgenic cells with LCMV Armstrong;

“skin γδ DETC” were γδ-TCR expressing dendritic epidermal T cells (DETC) collected from the skin of C57BL/6J mice 30 d following infection with HSV KOS; “αβ DETC” were αβ -TCR expressing DETC cells from uninfected Tcrd-/- mice. Consistent with brain- residence data, only 11 NR genes showed statistically significant changes between these resident populations, including splenic naïve (TN), central memory (TCM), and effector memory (TEM) for comparison (Fig. 5.10B). Among these, Adora2a and P2rx7 show strong enrichment in gut TRM, while P2ry14 is specific for skin resident T cells. Further, as Gria3 and Ramp1 are uniquely expressed in splenic TN, these are likely markers of antigen inexperience rather than tissue residence (Fig. 5.10B). Although there are a few notable

NR that show tissue preference, these data suggest that basic NR expression profiles in T cells are surprisingly resilient and not heavily dependent on tissue localization of the cells themselves.

We additionally sought to validate these trends using NanoString-based quantitation of NR transcript levels among FACS-purified T cell populations from different secondary lymphoid organs reflecting differences in transient tissue localization by T cells.

T cells were prepared and sorted as in Figures 5.3-5.4, except subcutaneous (sqLN),

174

mesenteric (mLN), and cervical lymph nodes (cLN) from C57BL/6 mice were kept separated, yielding CD4 naïve (CD4+CD44lo), CD4 memory (CD4+CD44hi), Treg

(CD4+CD25+), and CD8 T cell populations from each tissue. These were then analyzed using the same NanoString nCounter probe set (Table 5.6) as previously described (Fig.

5.10 C-F). The NR expression is predominantly unchanged when comparing the same T cell population between LN draining the skin, gastrointestinal tract, and brain. This is especially true among CD4 naïve T cells among which no NR transcript shows even a 2- fold difference between LN groups (Fig. 5.10C). Among CD4 memory T cells, NR expression is similarly consistent across secondary lymphoid organs with the exception of the purinergic receptors P2rx7, which shows 3.85- and 3.69-fold upregulation within the mLN and cLN compared to sqLN, respectively, and P2rx4, which is upregulated 2.57-fold in the cLN (Fig. 5.10D). Similarly, CD8 T cells in the mLN and cLN show 6.86- and 3.82- fold upregulation of P2rx7 compared to sqLN (Fig. 5.10F). These data are consistent with the findings from Mackay et al146 (Fig. 5.10B) that indicate that, among NR, purinergic receptors uniquely show tissue-specific expression changes by T cells.

175

A B

C D

E F

Figure 5.10. T cell NR expression is largely intrinsic, rather than dominantly modulated by tissue residence. (A) Microarray data from Wakim et al363, in which OT-1 cells were collected on d20 post-infection with VSV-Ova and sorted from the spleen (CD103-, N=5) or brain (CD103+, N=5; CD103-, N=3). BRB- ArrayTools class comparison analysis using a significance cutoff of p<0.01 identified 12 NR genes that differed between groups. * p<0.05 compared to spleen population, + p<0.05 between brain populations; using Student’s T-test. (B) Microarray data from Mackay et al146, in which the indicated T cell population was sorted from varying mouse models (see primary text). BRB-ArrayTools analysis as in A identified 11 NR genes. * p<0.05 compared to spleen TN population using Student’s T-test. (C-F) Cervical (cLN), mesenteric (mLN), and subcutaneous (sqLN) lymph nodes collected from 4 C57BL/6 mice were sorted as described in Figure 5.3 and underwent NanoString RNA count analysis as in Figure 5.5. Scatterplots comparing the indicated LN source on the y-axis compared to sqLN on the x-axis within CD4+CD44lo (C), CD4+CD44hi (D), CD4+CD25+ (E), and CD8+ (F) T cell populations are shown. Plots represent 2 biological replicates. Reprinted with permission from Rosenberg & Singh, 2019354 176

5.2.5. T cell NR expression patterns confirmed by RNA-Seq analysis

We next sought to validate the NR expression trends we had found using the more sensitive transcriptomic approach, RNA-Seq. We identified multiple available datasets housed in the NCBI GEO database that analyzed FACS-purified T cell populations from multiple contexts. We then compared NR expression profiles within these populations using analysis methods as above (Fig. 5.11). Two datasets147,148 compared memory precursor (MP) and terminal effector (TE) populations within P14 T cells in response to LCMV Armstrong infection, providing identical experimental conditions to the microarray dataset shown in Fig. 5.9. Both experiments recapitulated the finding that at days 7-8 of the T cell effector response, MP and TE cells show nearly identical NR expression profiles to one another (Fig. 5.11A-B). In line with Yu et al.142,

MP phenotype cells showed increased Vipr1 expression compared to TE at day 8, with expression nearly at the level of naïve T cells (Fig. 5.11A). These data also allowed us to compare NR expression between gut TRM (Fig. 5.10B) to small intestine intraepithelial cells (IEL) (Fig. 5.11B), revealing that both populations upregulate P2rx7 compared to naïve cells.

A third dataset compared naïve CD5hi and CD5lo populations to virtual memory

T cells (TVM), cells that have not been antigen exposed yet express CD44, in unmanipulated B6 mice. All three antigen inexperienced populations show nearly identical NR expression profiles, with the exception Adora2a and Ramp1 showing increased and decreased expression, respectively in TVM cells compared to both naïve populations (Fig. 5.11C). Beyond confirming the expression of many of the core signature NR, this dataset makes clear that NR expression is quite uniform among antigen

177

naïve T cells and that T cell activation drives dramatic modulation of the expression of this receptor class.

178

A B

Figure 5.11. NR core signature is validated by RNA-Seq analysis. RNA-Seq datasets from FACS-purified T cell populations were identified and analyzed for the expression of NR. Graphs show all NR that were detected. (A) FPKM RNA-seq data from Wang et al147, in which naïve CD8+ P14 T cells (N=2) were transferred to C57BL/6 mice, infected with LCMV-Armstrong, and splenic terminal effector T cells (TE, KLRG1+CD127-, N=3) and memory precursor T cells (MP, CD127+KLRG1- , N=3) were collected 8 dpi. (B) FPKM RNA-seq data from Milner et al148, in which naïve CD8+ P14 T cells were transferred to C57BL/6 mice, infected with LCMV-Armstrong, and splenic terminal effector T cells (TE, KLRG1hi,CD127lo), memory precursor T cells (MP, CD127hi,KLRG1lo), and small intestinal intraepithelial lymphocytes (IEL) were collected 7 dpi. N=2 for all populations. (C) FPKM RNA-seq data from White et al149, in which splenocytes from unmanipulated B6 mice were sorted into naïve (CD44lo, CD49dlo) CD5hi and lo populations, as well as virtual memory T cells (TVM, CD44hi, CD49dlo). N=3 for all populations. Reprinted with permission from Rosenberg & Singh, 2019354

179

5.2.6. Similar NR expression trends are found in human T cell populations.

Having thoroughly examined NR expression by murine T cells, we were interested to determine whether these findings translated to human populations. As above, we identified datasets through the GEO database in which human T cells were purified and their transcriptome measured by RNA-Seq. All three datasets l150,151 we analyzed examined NR expression by naïve T cells among healthy individuals, allowing us to first consider the core NR signature within this population. We found that all core signature NR were identified by these experiments, with all but two (Cnr2 and Chrnb2) detected in at least 2 experiments (Fig. 5.12). Interestingly, while Adrb2 was the most highly expressed NR within murine populations, it, while detectable in all three datasets, showed minimal expression in human T cells. Notably, however, the relative expression of these genes showed great variability between experiments. For, example Lpar6 was the most highly expressed NR in one dataset (Fig. 5.12C), but only barely reached the limit of detection in a second (Fig. 5.12B) and not at all in the third (Fig. 5.12A). This suggests that conclusions regarding absolute expression of NR may be difficult based on these datasets.

In addition to naïve T cell populations, two datasets150,151 examined NR profiles within TCM and TEM populations. The predominant trend observed is that naïve and memory T cell populations showed nearly identical NR expression profiles (Fig. 11 B-C); though, both datasets showed that Vipr1 is downregulated in TEM compared to naïve T cells, consistent with our murine data (Fig. 5.2, 5.5). Interestingly, both datasets show that Adrb2 expression was increased in TEM compared to naïve cells, suggesting a potential difference between human and murine T cell populations. However, due to the

180

variability within and between datasets, further experiments will likely be needed before drawing such specific conclusions regarding interspecies NR expression differences.

181

A B

Figure 5.12. NR core signature is validated by RNA-Seq analysis. RNA-Seq datasets from FACS-purified T cell populations were identified and analyzed for the expression of NR. Graphs show all NR that were detected. (A) FPKM RNA-seq data unpublished from GEO data series GSE112706, in which naïve CD4+ T cells were sorted from PBMC of healthy patients. Each sample was run in technical duplicate, N=5. (B) TPM RNA-seq data from Tian et al151, in which PBMC from dengue virus seronegative patients were sorted into CD3+CD4+ naïve (CCR7+CD45RA+, N=6), TCM (CCR7+CD45RA- , N=6), TEM (CCR7-CD45RA-, N=6), and TEMRA (CCR7-CD45RA+, N=9). (C) FPKM RNA-seq data from Spurlock et al150, in which PBMC were collected from healthy patients and sorted into CD4+ naïve (CD45RAhi, CCR7hi), TCM (CD45RAlo, CCR7hi), and TEM (CD45RAlo, CCD7lo). N=3 for all populations. Reprinted with permission from Rosenberg & Singh, 2019354

182

5.2.7. A working map of NR expression by T cells

The summary of the analysis presented so far in this chapter allows us to generate virtual maps of NR expression by different mouse T cell subsets (Fig. 5.13). Future studies using better refined antibody reagents, reporter mice and T cell-specific knockouts can further clarify the role of individual and clusters of NR expression in each T cell subset.

We expect that the maps we have generated will be a valuable resource for the design of such studies.

183

Figure 5.13. Maps of neurotransmitter receptor expression by T cells. Summary of NR expression across the T cell subsets studied. The size of each receptor icon approximately represents the relative expression of the receptor. Receptor labels in italics are not expressed in the indicated T cell subset. Plasma membrane seven transmembrane domain GPCR are shown on the left-hand side of the cell diagrams while ion channels are displayed on the right-hand side. OGFr, TSPO-PBR, and σ1-receptor are displayed on the nuclear, outer mitochondrial, and endoplasmic reticular membranes, respectively. (A) Full map of all T cell subsets studied. (B) Map of CD4+ T cell subsets separating the core NR profile from NR uniquely expressed by the indicated populations. Receptor in red indicates a uniquely downregulated receptor in the indicated population. Reprinted with permission from Rosenberg & Singh, 2019354

184

5.2.8. 2-adrenergic receptor signaling dampens T cell activation

Although the 2AR is one of the most heavily studied GPCR and has received significant attention within the context of immunity, the ability of transient signaling to modify subsequent TCR signaling and thus T cell activation has yet to be described. In order to understand this process, we used the pre-Tx experimental paradigm described in

Chapter 3, in which T cells are treated with the 2AR-specific agonist salbutamol for 30 min before being washed with fresh media, and incubated with anti-CD3 and anti-CD28 to induce activation. After 24 hr, T cell activation is quantified by measuring CD69 expression via flow cytometry. Figure 5.14 depicts the gating strategy used for identifying live CD4+ and CD8+ T cells and the gates used to define CD69 expression.

185

Figure 5.14. NR agonism analysis gating strategy. Lymph nodes and spleens from 3 B6 mice were collected and pooled. Live, mononuclear cells were separated using ficoll density centrifugation. T cells were enriched using 2 rounds of magnetic, negative selection, first with Dynabeads then with Miltenyi AutoMACS. In both rounds of selection, antibodies targeting CD11b, B220, NK1.1, and MHCII were used to enrich for T cells. Enriched T cells were then treated with the agonists for NR or PBS for 30 min before washing and incubation with anti-CD3ε (1 g/mL, plate-bound) and anti- CD28 (2 g/mL, soluble). After 24 hr of stimulation, T cells were prepared for flow cytometry. (A) Gating strategy for analyzing live CD4+ and CD8+ T cells. (B) Representative histograms for CD69 expression among CD4+ (left) and CD8+ (right) T cells treated with PBS.

186

Representative histograms of CD69 expression show that, in a dose-responsive manner, 30 min exposure to salbutamol decreases both the percentage of T cells expressing the marker and the per cell expression level as measured by gMFI (Fig. 5.15A). Summary of CD69 expression across 4 experiments confirms statistically significant decreases in both of these metrics among CD4+ T cells (Fig. 5.15B). CD8+ T cells show a similar trend in expression but significant variation between experiments prevented statistical significance (Fig. 5.15C). This variance is likely partially explained by the observation that the anti-CD3 and anti-CD28 stimulation conditions induced an approximately 4-fold lower percentage of CD8+ T cells to become CD69+ than CD4+ cells, thus allowing for a smaller window for observing decreases in expression. Additionally, CD8+ T cells express the 2AR between 1.5- and 2.5-fold less than CD4+ T cells (Fig. 5.1, 5.5-5.6), such that the effect of salbutamol pre-Tx could be diminished within this population.

187

Figure 5.15. 2 adrenergic receptor signaling inhibits early T cell activation. T cells were prepared, treated with salbutamol, a specific agonist for the 2 adrenergic receptor (Adrb2), and analyzed by flow cytometry as described in Figure 6.1. (A) Representative histograms of CD69 expression by CD4+ (left) and CD8+ (right) T cells following the indicated stimulation and treatment conditions. (B-C) Percentage of CD4+ (B) and CD8+ (C) T cells positive for CD69 (left) and gMFI of CD69+ T cells (right). Data includes 4 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

188

Having established that brief 2AR signaling dampens subsequent T cell activation, we sought to understand how 2AR signaling integrates with that of the TCR to elicit this effect. The 2AR is a Gs-coupled GPCR such that receptor agonism yields activation of adenylate cyclase (AC), production of cAMP, and activation of the PKA signaling pathway. cAMP signaling has been long established as inhibitory to cytolytic function of

T cells364-367, but its role in CD4+ T cells may be more complicated as AC signaling promotes IL-2 production368. We first used the AC activator forskolin in the same pre-Tx paradigm as described above to determine if transient AC activation yields the same functional inhibition as 2AR agonism (Fig. 5.16). Forskolin produced similar, but stronger decreases in T cell activation as measured by CD69 expression, both in CD4+

(Fig. 5.16A-C) and CD8+ T cells (5.16A,C). The increased effect of forskolin treatment is likely due to the potency of the drug treatment and that proper dose titration would yield a similar effect size to that of salbutamol. Alternatively, forskolin may bypass the negative regulatory mechanisms that 2AR signaling is subject to, such as -arrestin mediated receptor downregulation369-371. Additionally, the observation that CD8+ T cells are also affected is, again, likely due to bypassing the receptor that shows lower expression among this population. Although these data only provide correlative evidence, cAMP mediated signaling downstream of 2AR agonism likely underlies its dampening effect on T cell activation.

189

Figure 5.16. 2AR-mediated inhibition of T cell activation is likely mediated by cAMP. T cells were prepared, treated with forskolin, a specific inducer of adenylyl cyclase, and analyzed by flow cytometry as described in Figure 6.1. (A) Representative histograms of CD69 expression by CD4+ (left) and CD8+ (right) T cells following the indicated stimulation and treatment conditions. (B-C) Percentage of CD4+ (B) and CD8+ (C) T cells positive for CD69 (left) and gMFI of CD69+ T cells (right). Data includes 3 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

190

5.2.9. Transient agonism of other core signature NR do not inhibit T cell activation

We next sought to understand how brief signaling from other NR core signature constituents integrates with TCR signaling to impact T cell activation. We focused on the histamine H2 receptor (H2R), encoded by Hrh2, because it shares Gs signaling with the

2AR. Thus, we predicted that H2R engagement with the specific, small-molecule agonist amthamine dihydrobromide would similarly inhibit T cell activation as measured by CD69 expression 24 hr following stimulation. Interestingly, pre-Tx with the H2R agonist did not yield a significant decrease in percentage of CD4+ or CD8+ T cells expressing CD69; however, both T cell populations did show significantly lower CD69 gMFI (~70% of PBS group) suggesting that CD69 was downregulated on a per cell basis (Fig. 5.17). The difference in effect between H2R and 2AR engagement may be explained by different expression levels of the two receptors with CD4+CD44lo and CD8+ populations expressing the 2AR 12.0- and 5.8-fold higher, respectively (Fig. 5.5); however it would be predicted that the effect on CD69 gMFI would similarly be blunted by comparison, yet both receptors show similar dampening of CD69 gMFI (Fig. 5.15, 5.17). It is possible that intrapopulation expression variability may explain this discrepancy, but this would need to be explored with a more refined technique such as single cell RNA-sequencing.

191

Figure 5.17. Histamine H2R signaling may limit T cell activation. T cells were prepared, treated with amthamine dihydrobromide, a specific inducer of the histamine H2 receptor (Hrh2), and analyzed by flow cytometry as described in Figure 6.1. (A) Representative histograms of CD69 expression by CD4+ (left) and CD8+ (right) T cells following the indicated stimulation and treatment conditions. (B-C) Percentage of CD4+ (B) and CD8+ (C) T cells positive for CD69 (left) and gMFI of CD69+ T cells (right). Data includes 3 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

192

We next considered the role nicotinic acetylcholine receptors (nAChR) play on T cell activation within the pre-Tx paradigm. Our analysis found that the 2 subunit (encoded by Chrnb2) is expressed by all T cell subsets, while the 1 subunit (encoded by Chrnb1) is expressed by CD8 T cells (Fig. 5.5, 5.12). To study both of these receptors, we employed the prototypical agonist for the receptor class, nicotine, within our pre-Tx model. Following

24 hr of antibody stimulation, we observed no significant difference in CD69 expression by any metric in both CD4+ and CD8+ populations (Fig. 5.18). One explanation may be that the receptor is only expressed at low levels and effects comparable to the highly expressed 2AR should not be expected. An alternative explanation may be that, among the receptors tested in this paradigm, these are uniquely ligand-gated ion channels as opposed to GPCR. First, as cationic channels, these receptors would most likely interact with the TCR-driven Ca2+ signaling pathway and potentially have functional impacts not captured by this assay. Additionally, the duration of ion flux and its resultant signaling may be short-lived by comparison to GPCR signaling, such that the 30 min time frame may not capture the functional effects.

193

Figure 5.18. Nicotinic acetylcholine receptor signaling does not alter T cell activation. T cells were prepared, treated with nicotinic, a specific inducer of the nicotinic acetylcholine receptor (nAChR) class, and analyzed by flow cytometry as described in Figure 6.1. (A) Representative histograms of CD69 expression by CD4+ (left) and CD8+ (right) T cells following the indicated stimulation and treatment conditions. (B-C) Percentage of CD4+ (B) and CD8+ (C) T cells positive for CD69 (left) and gMFI of CD69+ T cells (right). Data includes 3 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

194

5.2.10. NR agonism does not induce ERK phosphorylation

In order to further understand the mechanism by which NR signaling integrates with TCR signaling, we used phospho-flow cytometry to measure ERK phosphorylation in the context of NR agonism (Fig. 5.18). We focused on ERK because the literature is mixed regarding the impact of cAMP signaling on ERK as some report that it is activated by cAMP mediated activation of PKA372,373, while other studies suggest that ERK

374,375 phosphorylation is reduced . We found that specific agonism of the 2AR, H2R, histamine H4 receptor (H4R), and adenosine A2A receptor (A2AR) as well as induction of

AC by forskolin did not yield ERK phosphorylation from 1 to 30 min of agonist incubation

(Fig. 5.19). Additionally, agonism of the sigma-1 non-opioid receptor (1R), which modulate IP3 mediated Ca2+ release, potentially indirectly altering ERK signaling, did not induce ERK phosphorylation. These data refute the hypothesis that cAMP signaling drives activation of the ERK pathway, suggesting that, in primary T cells, cAMP signaling may potentially inhibit ERK signaling. Additional studies in which T cells are treated with these

NR agonists in the context of TCR-mediated activation would allow for further support of this latter hypothesis.

195

Figure 5.19. ERK phosphorylation is not modified by NR agonism alone. Lymph nodes were isolated from B6 mice and cells were treated with specific agonists at the indicated doses for the indicated times or 50 ng/mL PMA for 10 min. All treatments occurred at 37C before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy for analyzing T cells. (B-G, next page) Representative histograms of pERK expression among all CD4+ T cells (left) and CD4+ CD44hi (right) treated with salbutamol (B), forskolin (C), amthamine dihydrobromide (D), VUF 10460 (E), LUF 5834 (F), or PRE 084 (G) at the indicated stimulation times are shown. (H, next page) Quantification of pERK gMFI among all CD4+ T cells (left) and CD4+ CD44hi (right) are shown, labeled by the agonist target.

196

B C

D E

F G

197

5.3 Significant findings & Discussion

An extensive body of literature suggests that neurotransmitters can affect activation, differentiation, migration and effector functions of T cells – at least some of them by directly impinging on T cells 333. The analysis presented in this paper now allows us to compile for the first time a fairly detailed map of NR expression by different mouse

T cell subsets (Fig. 5.12) – illustrating both the quantitative and qualitative patterns we find. Importantly, we find that a relatively stable subset (26) of all currently annotated (185)

NR genes are found in T cells. Of these, about half (14) are relatively constitutively expressed in all T cell subsets, while very few (often 1 or 2 genes) show expression that is specific to CD4 memory, CD8 memory or regulatory T cells. While some of these NRs have previously been reported to be expressed at the RNA, protein or functional level, others have not. Overall, we expect this data to be a valuable resource towards the design and interpretation of future studies on understanding how such communication influences

T cell fate.

As noted at the end of the results section, evaluating the protein dynamics resulting from this transcriptomic map may require new reagents or the refinement of existing ones.

Nevertheless, additional studies looking at this transcriptomic signature in different disease, behavior and other pathophysiological states can provide new information about the underlying circuitry in T cells that listens to neuronal outputs – and how those change in vivo. Towards this end, our data also allows us to compare the robustness of different techniques for measuring NR transcripts with adequate sensitivity. There is certainly some variation in the results based on the methods. For instance, based on the ImmGen analysis, the nicotinic acetylcholine receptor Chrnb1 was not detected in any subsets, but the

198

NanoString nCounter approach did detect expression uniquely in CD8 T cells. Similarly, the GABAA-ρ2 receptor Gabrr2 was defined by ImmGen as being high in naïve CD4 and

CD8 T cells, but NanoString was able to find it across all subsets. Based on these, we suggest that the NanoString platform provides a rapid, sensitive and fairly straightforward workflow for analyzing NR transcript signature in T cells. The protocols detailed in

Chapter 2 should help most laboratories to investigate NR expression from RNA preps or even cell lysates of approximately 5000 T cells.

The biological significance of the expression of some of the NRs in the core signature we describe have been previously reported in individual studies312,328,329,346,376-

379. The most well studied among these have been the roles of agonists for Cnr2, Hrh2,

Adrb2 and Chrnb2 to affect T cell differentiation – each favoring the differentiation towards TH2 cytokines as opposed to the TH1 ones328,380-384. While the first three are G- protein coupled receptors that can influence T cell activation via cAMP and PKA, the last is the subunit of the nicotinic acetylcholine receptor which signals through alternate mechanisms. A more direct impact of TCR signaling has been suggested by the activity of receptors such as Tspo, Sigmar1, and Ogfr which may alter the intracellular levels of Ca2+ and other ions385-392. These two mechanisms could perhaps be unified, since the strength of TCR signals can strongly influence the differentiation of T cells towards a variety of effector fates79,393-395. The challenge is to understand how these disparate signals are integrated to tweak such T cells fates in significant ways. Indeed there is already some evidence of the consequences of NR expression in protective immunity to pathogens as well as the formation of immunological memory396,397.

199

Circadian rhythm constitutes a compelling context in which neuronal signaling may regulate immunity. Indeed, the numbers many immune cells, including T cells, in circulation show diurnal changes398. In fact, 2AR signaling has been shown to control the circadian patterns in CD4+ T cell and B cell circulation and trafficking, contributing to autoimmune susceptibility347,399. Further, gut ILC function shows circadian patterns with feeding behavior and clock genes contributing to disease susceptibility304,305,400. While it is clear that diurnal cycles contribute to immune responses, the precise mechanisms by which circadian rhythms can regulate T cell function require further investigation.

Given this preponderance of literature288,312,328,329,346,376-378,401-404 it is perhaps not surprising that T cells express a definitive NR signature. But it is intriguing that there are relatively few which are expressed in specific lineages. Of course, further sub-setting as well as single cell transcriptomic analyses may reveal additional complexities, but our data already suggest some candidates for functional studies. We expect such future studies to shed light on how a rigid framework of neural-immune communication operates on specific subsets of T cells to modulate immune responses.

200

Chapter 6: VIP as a model NIC for subset specific modulation of T cell biology

6.1. Introduction

So far in this thesis, we have started with a broad concept of signal 0, focused on signaling modalities that can deliver signal 0 and finally receptors that fulfil the requirements for those capable of delivering signal 0 to T cells. Among this the most prominent, as discussed in chapter 4, are neurotransmitter receptors (NR). The global exploration of NRs also identified some unique players which are interesting in their own right. Some of these have already been examined extensively in the context of T cell activation There is a considerable body of previous literature suggesting that neurotransmitters influence T cell function312,313,329,331,333,340,356. In the case of a few receptors such as the 2 adrenergic receptor335,339,341,347,353,405, characterization using precise reagents and cell-specific knockout models has indeed led to a clear definition of how these receptors directly affect T cells. But this is not the case for many other neurotransmitters.

Here we decided to focus on one, namely VPAC1 (Vipr1), owing to its unique pattern of expression observed in our transcriptomic analysis. Vipr1 is among the most highly expressed NR within the naïve T cell population, yet shows rapid downregulation upon activation (Fig. 5.8). Most intriguingly, Vipr1 expression is almost entirely lost among Treg (Fig. 5.5). Both observations evoke questions as to the functional impact

VPAC1 plays across the T cell subsets and during the response to infection.

6.1.1. VPAC1 or the VIP receptor, is known to have immunomodulatory roles.

Vipr1 is unique, since it is expressed on all T cells, but robustly downregulated by

406-413 Treg. Previous literature on the impact of VIP on T cells lacks clarity , as most studies

201

have found a Treg-promoting role for VIP signaling, yet VIP and VPAC1 expression exacerbates varying disease models412-414. Additionally, recent reports on the role of VIP in regulating ILC3 and small intestine epithelial cells in the context of colitis and bacterial infection highlight its critical significance in regulating immunity in a physiological context; however, two studies showed contradictory results, emphasizing the necessity for further investigation of the biology of VIP within the immune system304,305.

Adding further intrigue is discrepancies in observations regarding the expression of the VIP receptors. As discussed above, while T cells show high expression of VPAC1, the other VIP receptor VPAC2, is not expressed in most T cells – but previous studies have suggested that it might be upregulated in activated T cells. However, in our preliminary analysis, there was no significant upregulation of VPAC2 in the initial phase of clonal expansion – suggesting there may be contextual variations in VPAC2. Nevertheless, our data that Vipr1 is the primary VIP receptor on resting T cells is consistent with others in the field65,75. In previous studies, VIP has been implicated in multiple T cell phenotypes32-

50. Although the precise molecular mechanisms were not clear in many of these studies, there is precedent for suspecting an influence on the TCR-signaling cascade. Using approaches such as forced overexpression of VPAC2 or cultures with agonistic antibodies multiple groups have reported alterations in c-maf, JunB and NFAT activation in the presence of VIP signaling (mostly via VPAC2 in activated T cells)65,72.

Administering VIP as a clinical therapeutic is already under consideration in different contexts and therefore several candidate compounds with lipid modifications that stabilize VIP in the serum are under development412,415-421. Importantly, given the pleiotropic effects on VIP on multiple cell types in vivo 422,423, precisely pinpointing

202

mechanistic basis for VIP action on T cells is expected to be critical in understanding observed clinical outcomes.

6.2. Results

6.2.1. T cell activation is inhibited by transient VIP exposure

To first examine the effect of brief VIP exposure on T cell activation, we used our pre-Tx paradigm described in Chapters 3 and 5 in order to measure expression of CD69.

Polyclonal T cells from B6 were treated for 30 min with VIP before washing with fresh media and activation with anti-CD3 and anti-CD28. After 24 hr, CD69 expression, measured by percentage of cells positive for the marker as well as CD69 gMFI among positive cells, was significantly decreased within the CD4+ population treated with the highest dose of VIP (Fig. 6.1B-C). Similarly to 2AR pre-Tx (Fig. 5.15), CD8+ cells did not show significant decreases in CD69 expression (Fig. 6.1B,D), again, likely due to the narrow window for observable decrease. Additionally, when stratified by CD44 expression, CD44lo, naïve CD4 T cells showed significant decreases in activation, while the effect was diminished among CD44hi, antigen-experienced CD4 T cells, not reaching statistical significance (Fig. 6.1E). This may be due to decreased expression of the VIP receptor (Vipr1) among CD44hi cells by 4.4-fold (Fig. 5.5-5.6). These results are remarkably consistent with the effect caused by agonism of the 2AR, which may be expected as both are Gs-coupled GPCR; however, the H2R did not yield this effect so it is likely that signal integration between NR and TCR is complex.

203

C D E

Figure 6.1. VIP inhibits early T cell activation in a polyclonal population. T cells were prepared, treated with the indicated doses of VIP, and analyzed by flow cytometry as described in Figure 5.13. (A) Gating strategy for analyzing live CD4+ and CD8+ T cells. (B) Representative histograms of CD69 expression by CD4+ (left) and CD8+ (right) T cells following the indicated stimulation and treatment conditions. (C-D) Percentage of CD4+ (C) and CD8+ (D) T cells positive for CD69 (upper) and gMFI of CD69+ T cells (lower). (E) Percentage of CD4+ T cells positive for CD69, stratified by CD44 expression. Data includes 3 independent experiments, each normalized to the PBS treated group within the experiment. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

204

We next used a similar pre-Tx experimental design to probe the effect of VIP signaling on T cell proliferation. B6 T cells were loaded with CTV, treated with VIP for

30 min, washed with fresh media, and activated using anti-CD3 and anti-CD28. After 72 hr, proliferation as measured by dilution of CTV was assayed (Fig. 6.2). CD4 T cells exposed to VIP for a brief period, prior to TCR-mediated activation show a dose-dependent decrease in proliferation, indicated by an increased percentage of undivided T cells (Fig.

6.2B) and a decreased mean cell division number (Fig. 6.2C). These data are consistent with a decreased expression of CD69 at 24 hr after stimulation, suggesting that even transient VIP signaling dampens T cell activation.

205

A B

Figure 6.2. VIP decreases T cell proliferation. Lymph nodes and spleens from B6 mice were collected and live, mononuclear cells were separated using ficoll density centrifugation. T cells were enriched using 2 rounds of magnetic, negative selection, first with Dynabeads then with Miltenyi MACS. In both rounds of selection, antibodies targeting CD11b, B220, NK1.1, and MHCII were used to enrich for T cells. Enriched T cells were then loaded with CTV and treated with the indicated dose of VIP or PBS for 30 min before washing and incubation with anti-CD3ε (3 g/mL, plate-bound) and anti-CD28 (2 g/mL, soluble). After 72 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histograms of CTV dilution among CD4+ T cells treated with the indicated dose of VIP or PBS. (B-C) Quantitation of the percentage of undivided cells, corrected for divisions (B) and mean number of divisions (C) among CD4+ T cells. Data represents biological triplicates. Displayed as mean ± SD. * p<0.05 compared to PBS-treated group using Student’s T-test.

206

Although we had shown a dampening effect of VIP within a polyclonal CD4 T cell population, this may mask intrapopulation variability. As such, we next aimed to determine whether this effect is also observed within monoclonal T cell populations, using the TCR-

Tg 5C.C7 and SMARTA mouse lines. T cells from both lines underwent VIP pre-Tx before antibody-mediated activation with anti-CD3 and ani-CD28, measuring CD69 expression by flow cytometry 4 and 24 hours later. After 4 hr, both VIP-treated 5C.C7 (Fig. 6.3A-B) and SMARTA (Fig. 6.4A-B) T cells showed strong decreases in percentage of CD69 expressing T cells and CD69 expression on a per cell basis as measured by gMFI. At 24 hr, VIP treatment showed less pronounced decreases in percentage of CD69 expressing T cells in both transgenic lines, but CD69 gMFI was still dramatically reduced (Fig. 6.3A,C;

Fig. 6.4A,C). These data suggest that VIP acts similarly on individual T cell clones to reduce early activation as it does on the entire population.

207

B C

Figure 6.3. VIP inhibits early antibody-mediated activation of 5C.C7 T cells. Lymph nodes from 5C.C7 mice were collected and treated with 100 nM VIP or PBS for 30 min before washing and incubation with anti-CD3ε (3 g/mL, plate-bound) and anti-CD28 (2 g/mL, soluble). After 4 or 24 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histograms of CD69 expression by 5C.C7 T cells following 4 (left) or 24 hr (right) of stimulation. (B-C) Percentage of 5C.C7 T cells positive for CD69 (upper) and gMFI of CD69+ T cells (lower) after 4 (B) and 24 hr (C) of stimulation. Data represents technical replicates. Displayed as mean ± SD.

208

B C

Figure 6.4. The effect of VIP on early antibody-mediated activation is replicated in SMARTA T cells. Lymph nodes from SMARTA mice were collected and treated with 100 nM VIP or PBS for 30 min before washing and incubation with anti-CD3ε (3 g/mL, plate-bound) and anti-CD28 (2 g/mL, soluble). After 4 or 24 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histograms of CD69 expression by SMARTA T cells following 4 (left) or 24 hr (right) of stimulation. (B-C) Percentage of SMARTA T cells positive for CD69 (upper) and gMFI of CD69+ T cells (lower) after 4 (B) and 24 hr (C) of stimulation. Data represents technical replicates. Displayed as mean ± SD.

209

Although antibody-mediated activation is a useful tool for studying T cells, we next sought to expand our experimental pre-Tx paradigm to examine T cell activation under more physiological circumstances using APC and antigen to activate the TCR-Tg T cells.

5C.C7 and SMARTA cells were similarly treated with VIP for 30 min and then washed with fresh media but were then added to culture with increasing doses of antigenic peptide

(MCC and LCMV GP, respectively) and splenocytes from T cell deficient mouse lines with the same MHC background (CD3-KO and TCR-KO, respectively). After 4 and

24 hr of culture, T cell expression of CD69 was assessed (Fig. 6.5). Similar to antibody- mediated activation, at 4 hr, both T cell lines show decreased CD69 expression by percentage and gMFI (Fig. 6.5 B,D). Interestingly, CD69 gMFI seems to plateau at antigen doses 30 nM. Importantly, at 24 hr, both T cell lines show high frequencies of cells expressing CD69 across all peptide doses, with VIP showing no effect; however, CD69 gMFI is still attenuated at this time point (Fig. 6.5C,E). These data suggest that VIP also dampens early T cell activation within TCR-Tg T cell populations activated with antigenic peptide, but that either the intensity of stimulation or kinetics of CD69 expression under this context is different than that of antibody-mediated activation (Fig. 6.3-6.4).

210

B C

D E

SMARTA

Figure 6.5. VIP inhibits early peptide-mediated activation of 5C.C7 and SMARTA T cells. Lymph nodes from 5C.C7 and SMARTA mice were collected and treated with 100 nM VIP or PBS for 30 min before washing and incubation with CD3ε-KO or TCR-KO splenocytes, respectively, as well as the indicated dose of agonist peptide. After 4 or 24 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histograms of CD69 expression by 5C.C7 T cells following 4 hr of stimulation with the indicated dose of peptide. (B-E) Percentage of 5C.C7 (B-C) and SMARTA (D-E) T cells positive for CD69 (upper) and gMFI of CD69+ T cells (lower) after 4 (B,D) and 24 hr (C,E) of stimulation. Data represents technical replicates. Displayed as mean ± SD.

211

To assess the effects of transient VIP exposure on proliferation of monoclonal T cell populations activated using antigenic peptides, we again employed our pre-Tx experimental paradigm. 5C.C7 T cells were loaded with CTV, treated with VIP, washed with fresh media, and incubated with CD3-KO splenocytes and MCC peptide. After 72 hr of culture, proliferation was measured using flow cytometry (Fig. 6.6). Interestingly,

VIP did not decrease T cell proliferation under these conditions, and may show slight enhancement of proliferation as measured by mean division number and percentage of undivided cells. This is consistent with CD69 data that indicate that VIP has a limited effect in the context of antigen-mediated T cell activation compared to antibody-mediated stimulation (Fig. 6.3-6.5). Alternatively, the dose of antigenic peptide used may induce much stronger activation than that of the antibody cocktail, thus masking any inhibitory effects of VIP; however, lower doses of antigen also showed equivalent CD69 expression at 24 hr, suggesting a more complex explanation than stimulation dose underlies this discrepancy.

212

A B

Figure 6.6. Proliferation of 5C.C7 T cells is unaffected by VIP. Lymph nodes from 5C.C7 mice were collected and were loaded with CTV. Cells were then treated with the indicated dose of VIP or PBS for 30 min before washing and incubation with CD3ε-KO and 1 M MCC peptide. After 72 hr of stimulation, T cells were prepared for flow cytometry. (A) Representative histograms of CTV dilution among CD4+ T cells treated with the indicated dose of VIP or PBS. (B-C) Quantitation of the percentage of undivided cells, corrected for divisions (B) and mean number of divisions (C) among CD4+ T cells.

213

6.2.2. VIP signaling inhibits phosphorylation of ERK in T cells

We next sought to understand the mechanism by which VIP and VPAC1 signaling integrates with that of the TCR to dampen T cell activation. As discussed in Chapter 6, the interaction between cAMP signaling and MAPK signaling in T cells is poorly understood, but, based on our previous data (Fig. 5.19), we predict that cAMP induced by VPAC1 signaling would decrease ERK phosphorylation downstream of TCR-mediated activation.

To test this hypothesis, we employed phospho-flow cytometry to monitor ERK phosphorylation following antibody-mediated TCR-crosslinking using anti-CD3, anti-

CD4, anti-CD8, and anti-CD28 with or without VIP treatment (Fig. 6.7-6.8). Both CD4+ and CD8+ T cells show robust decreases in ERK phosphorylation when VIP is added to antibody stimulation compared to the stimulation only control (Fig. 6.7B-C, 6.8 A-B). Both the percentage of T cells that express pERK and the peak pERK gMFI are attenuated by the addition of VIP (Fig. 6.8 A-B). Interestingly, VIP pre-Tx for 30 min before antibody- stimulation also shows dampening of ERK phosphorylation, albeit not to the same degree as concurrent VIP treatment (Fig. 6.7B-C, 6.8 A-B). Additionally, pSLP-76 and pZap70 expression appear unaltered by either VIP treatment paradigm (Fig. 6.7B-C 6.8C-F), suggesting that integration between the two signaling pathways occurs downstream of the

LAT complex. Lastly, consistent with other NR treatments (Fig. 5.19), VIP treatment in the absence of antibody-crosslinking did not induce ERK phosphorylation.

214

Figure 6.7. VIP inhibits ERK induction in T cells. Lymph nodes from B6 mice were collected and cells were stained with 10 g/mL biotinylated-anti-CD3, anti-CD4, anti-CD8, and anti-CD28. Cells were then either treated on ice with 100 nM VIP (“VIP pre-Tx + stimulation”) or left untreated on ice (remaining samples). Cells were then moved to 37C and incubated with 100 nM VIP (“VIP”), 5 g/mL streptavidin (“Stimulation” and “VIP pre-Tx + stimulation”), or both (“Stimulation + VIP”) for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (A) Gating strategy used for analyzing T cells. (B-C, next page) Representative histograms of pERK, pSLP-76, and pZap70 expression among CD4+CD44lo (B) and CD8+CD44lo (C) T cells following up to 10 min of the indicated stimulation condition.

215

B C

Figure 6.7. VIP inhibits ERK induction in T cells (cont.). Lymph nodes from B6 mice were collected and cells were stained with 10 g/mL biotinylated-anti-CD3, anti-CD4, anti-CD8, and anti-CD28. Cells were then either treated on ice with 100 nM VIP (“VIP pre-Tx + stimulation”) or left untreated on ice (remaining samples). Cells were then moved to 37C and incubated with 100 nM VIP (“VIP”), 5 g/mL streptavidin (“Stimulation” and “VIP pre-Tx + stimulation”), or both (“Stimulation + VIP”) for the indicated times before fixation with 1.5% PFA. Cell membranes were permeabilized with MeOH and rehydrated with FACS buffer before being stained overnight with antibodies in preparation for flow cytometry. (B-C) Representative histograms of pERK, pSLP-76, and pZap70 expression among CD4+CD44lo (B) and CD8+CD44lo (C) T cells following up to 10 min of the indicated stimulation condition.

216

A B

C D

E F

Figure 6.8. Quantitation of TCR signaling following transient exposure to VIP. T cells were collected, stimulated, and prepared for flow cytometry as described in Figure 6.7. (A-B) Percentage of pERK+ cells (upper) and gMFI of pERK signal within these cells (lower) among CD4+ (A) and CD8+ (B) T cells over 60 min of stimulation. (C-D) gMFI of pSLP-76 among CD4+ (C) and CD8+ (D) T cells over 60 min of stimulation. (E-F)) gMFI of pZap70 among CD4+ (E) and CD8+ (F) T cells over 60 min of stimulation.

217

6.2.3. VIP enhances TH22 and TH17 differentiation

In addition to effects on T cell activation, we sought to understand the role VIP plays in skewing CD4 T cell differentiation. A recent study showed that VIP released by enteric neurons induces release of IL-22 from small intestine lamina propria ILC3304; however, a similar effect has not been shown in T cells. In order to test this effect in T cells, we activated lymphocytes from B6 mice using anti-CD3 and anti-CD28, in the presence or absence of a TH22 skewing cytokine cocktail, with or without the addition of

VIP. After 3 d, cytokine production by T cells following reactivation with PMA and ionomycin was assessed using intracellular cytokine staining (ICS) (Fig. 6.9-6.11).

Although the addition of VIP alone did not increase IL-22 output, in the context of the skewing cytokines, VIP yielded increased production of IL-22 (Fig. 6.9). Interestingly, IL-

17A was similarly induced by VIP in the context of TH22 skewing (Fig. 6.10). This is consistent with prior studies that show that VIP increases IL-17 release by T cells; however, concurrent addition of TGF was necessary to elicit the effect424. Conversely, T cells differentiated under TH17 conditions treated with VIP showed decreased IL-17 production425. Additionally, while production of IFN was decreased by the addition of the

TH22 cocktail, VIP neither increased nor decreased IFN (Fig. 6.11).

218

Figure 6.9. VIP enhances IL-22 production by T cells. Lymph nodes from B6 mice were collected and activated using anti-CD3 (3 g/mL, plate-bound) and anti- CD28 (2 g/mL, soluble) in the presence or absence of 100 nM VIP as well as the presence or absence of a TH22 skewing cocktail (10 ng/mL IL-1, 30 ng/mL IL-6, 20 ng/mL IL-23, 400 nM FICZ, 10 g/mL anti- IFN, 10 g/mL anti-IL-4). After 3 days in culture, cells received fresh media and were reactivated using 50 ng/mL PMA and 1 g/mL ionomycin for 5 hours. After 1 hour, brefeldin A was added to culture. Cells were then prepared for flow cytometry. (A) Gating strategy for identifying CD4+ T cells. (B) IL-22 production by CD4+ T cells following treatment with or without VIP and the indicated skewing condition.

219

Figure 6.10. VIP enhances IL-17 production by T cells. B6 lymph node cells underwent in vitro activation as described in figure 6.9. IL-17A production by CD4+ T cells following treatment with or without VIP and the indicated skewing condition is shown.

220

Figure 6.11. IFN production largely unaffected by VIP. B6 lymph node cells underwent in vitro activation as described in figure 6.9. IFN production by CD4+ T cells following treatment with or without VIP and the indicated skewing condition is shown.

221

Previous studies have shown that IL-17 production by T cells is restrained by

STAT5-mediated IL-2 signaling426. Because we have shown that VIP dampens T cell activation, including CD69 expression (Fig. 6.1, 6.3-6.5) and proliferation (Fig. 6.2), we hypothesized that IL-2 production is also attenuated, blocking the inhibition on IL-17 production it provides. In order to test this hypothesis, we again activated B6 T cells using antibody-crosslinking under TH17 or TH22 skewing conditions, with the addition of VIP.

Exogenous IL-2 was then added to culture, which we predicted would compensate for any reduction in IL-2 release elicited by VIP treatment. After 4 d of in vitro culture and reactivation, ICS was used to measure cytokine production. While IL-17 production was again enhanced by the addition of VIP under TH17 skewing conditions, exogenous IL-2 decreased the percentage of IL-17+ cells regardless of the presence of VIP (Fig. 6.12B,

3.0-fold decrease without VIP, 2.1-fold decrease with VIP). Additionally, a similar trend is shown for IL-22 production (Fig. 6.12C), while IFN actually showed slight increases to expression when IL-2 was added to culture (Fig. 6.12D). Because exogenous IL-2 showed similar effects on cytokine production irrespective of VIP treatment, it is unlikely that differences in IL-2 production explain the IL-17 and IL-22 enhancing effects of VIP treatment.

222

B C

Figure 6.12. IL-2 dampens IL-17 and IL-22 production. Lymph nodes from B6 mice were collected and activated using anti-CD3 (1 g/mL, plate-bound) and anti- CD28 (2 g/mL, soluble) in a TH17 skewing cocktail (10 ng/mL IL-1, 30 ng/mL IL-6, 20 ng/mL IL-23, 10 g/mL anti-IFN, 10 g/mL anti-IL-4), a TH22 skewing cocktail (10 ng/mL IL-1, 30 ng/mL IL-6, 20 ng/mL IL-23, 400 nM FICZ, 10M galunisertib, 10 g/mL anti-IFN, 10 g/mL anti-IL-4) or no skewing control. Cultures were then supplemented with 100 nM VIP, 10 U/mL IL-2, or both. After 4 days in culture, cells received fresh media and were reactivated using 50 ng/mL PMA and 1 g/mL ionomycin for 5 hours. After 1 hour, brefeldin A was added to culture. Cells were then prepared for flow cytometry. (A) Gating strategy for identifying CD4+ T cells. (B) IL-22 production by CD4+ T cells following treatment with or without VIP and the indicated skewing condition.

223

6.2.4. VPAC1 is most highly expressed in naïve CD4+ T cells

Considering the subset-specific and dynamic expression of Vipr1 by T cells observed in previous studies (Fig. 5.1, 5.5-5.6, 5.8), we next sought to study the differential effects of VPAC1 signaling within these varied contexts. We first validated the early, rapid downregulation of Vipr1 following activation by stimulating 5C.C7 with APCs and MCC peptide, and measuring Vipr1 expression using RT-qPCR (Fig. 6.13A). Our data confirms that Vipr1 is downregulated 8.0-fold within 48 hr of activation, which is consistent, though more robust than L. monocytogenes infection (Fig. 5.8A, 2.68-fold) or LCMV Armstrong infection (Fig. 5.8C, 4.24-fold).

Although we had previously shown that VPAC1 is expressed by T cells at the protein level (Fig. 5.7), we sought to compare protein expression between T cell populations. We used FACS-purified T cell populations as described in Chapter 5 to prepare lysates for western blot analysis (Fig. 6.13B-C). We were able to detect VPAC1 in all subsets studied, and, when normalized to -actin loading control, confirmed that

CD4+CD44lo T cells show the highest receptor expression (Fig. 6.13C). Variation in signal intensity between replicate blots did not allow for precise ordering of receptor expression between populations, but CD4+CD25+ Treg express lower levels of VPAC1, though not as comparatively low, as predicted by transcript quantification.

224

B C

Figure 6.13. Vipr1 is downregulated after activation of naïve T cells. (A) 5C.C7 lymph nodes were collected and cells were activated through culture with splenocytes from CD3- KO mice, 1 M MCC peptide, and 10 U/mL IL-2. After the indicated number of days in culture, cells were recovered and mRNA was isolated for reverse transcription. qPCR was then used to quantify Vipr1 expression (normalized to Actb and day 0 time point). (B-C) T cells were FACS-purified into the indicated populations as described in Figure 5.3, and then lysed using 2x lysis buffer. (B) Anti-VPAC1 western blot, representative of two independent blots. Anti--actin used as loading control. (C) Absolute (upper) and normalized to actin (lower) densitometry quantified for both blots using ImageJ.

225

6.2.5. Overexpression of VPAC1 alters T cell effector function and survival

Because Vipr1 is robustly downregulated in activated T cells as well as Treg, we sought to understand the importance of this downregulation and the impact it has on T cell function. As such, we aimed to use a retroviral expression system to force constitutive expression of VPAC1 in T cells. We generated Vipr1 cDNA from purified naïve CD4 T cell cDNA samples and used In-Fusion cloning to insert the gene segment into a retroviral expression vector plasmid (Fig. 6.14, insertion confirmed by gene sequencing). This system allows for puromycin selection as well as identification of successful protein translation using an E2a-linked BFP reporter system. We confirmed that Vipr1-expressing

(pV-RV) and vector control retroviruses (pQ-RV) can be successfully transduced into T cells with expression confirmed by measuring BFP (Fig. 6.15).

226

Figure 6.14. Vipr1 overexpression plasmid diagram. cDNA from FACS-purified CD4+CD44lo T cells was used as a template to amplify Vipr1 cDNA which was then inserted into the pQ2aB retroviral expression vector using In-Fusion cloning. Translation yields mature VPAC1 connected to BFP by the cleavable E2a linker. Puromycin resistance is included as a selection method.

227

B C

Figure 6.15. Retroviral expression of Vipr1. Lymph nodes from 5C.C7-Foxp3-GFP mice were collected and activated in vitro using splenocytes from PCC-CD3-KO mice supplemented with 3 M MCC peptide. After 30 hours of activation, cells were incubated with retroviral expression vectors expressing Vipr1 (pV-RV, 1:10 final dilution) or empty vector (pQ-RV, 1:100 final dilution), using centrifugation to improve cell to virus contact (2000 rpm x 2 hr, 32C). A second round of transduction was repeated 10 hours later. 24 hr later, puromycin was added to culture (1 g/mL). After an additional 24 hr, cells were collected for flow cytometric analysis. (A) Gating strategy for analyzing 5C.C7 cells. (B-C) Transduction efficiency as measure by BFP expression among vector (B) and Vipr1-expressing (C) retroviruses. Puromycin-selected cultures displayed in blue, unselected cultures in red.

228

Employing our retroviral, constitutive expression approach, we next sought to determine the functional consequences of continual VPAC1 expression in acutely activated

T cells. pV-RV and pQ-RV transduced T cells were adoptively transferred to T cell deficient mice (CD3-KO) and then acutely activated using i.p. injection of MCC peptide and LPS or left unchallenged. After 4 d, T cells were recovered from animals and reactivated ex vivo to measure IFN production by cytokine capture assay (Fig. 6.16).

Interestingly, pV-RV transduced cells in both, challenged and unchallenged mice showed increased IFN production over pQ-RV vector control cells (Fig. 6.16A), with the unchallenged mice showing slightly more IFN release than challenged mice. That both in vivo challenged and unchallenged mice show IFN production can most likely be explained by the fact that T cells in both conditions had been previously activated ex vivo to allow for efficient retroviral transduction. More intriguingly, however, is the observation that

BFP-, untransduced 5C.C7 cells that were challenged in vivo alongside pV-RV transduced

(BFP+) cells showed increased IFN production compared to pQ-RV vector transduced and untransduced cells (Fig. 6.16B). This indicates that constitutive VPAC1 expression allows for increased IFN release in both intrinsic and extrinsic mechanisms.

229

Figure 6.16. Overexpression of Vipr1 increases IFN production by effector cells. 5C.C7-Foxp3-GFP lymph node cells were transduced and puromycin-selected as described in Figure 6.15. 24 hours after addition of puromycin, 100,000 vector (pQ-RV) and Vipr1-expressing (pV-RV) transduced cells were adoptively transferred, separately, by retro-orbital injection to CD3-KO mice. After 24 hr, animals were injected i.p. with 30 g MCC peptide and 2.5 g LPS to activate transferred cells or left unchallenged. After 4 d, LN and splenocytes were recovered from the animals and reactivated using supplementation with 1 M MCC peptide. 24 hours later, IFN production was measured by flow cytometric capture assay. (A) IFN production by the indicated retrovirally transduced CD4+ T cells in challenged or unchallenged mice. Histogram (left), plotted gMFI (right). (B) IFN production by transduced and untransduced CD4+ T cells within challenged mice. Histogram (left), plotted gMFI (right).

230

We next sought to understand whether constitutive expression of VPAC1 played a role in survival of effector T cells. To test this, we transduced T cells from two different

TCR-Tg mice on the same genetic background (B10.A), 5C.C7 and A1M, with both pQ-

RV and pV-RV. We adoptively transferred reciprocal pairings of TCR-Tg and transduced retrovirus (pQ-RV-transduced 5C.C7 + PV-RV-transduced A1M, and vice versa) to T cell replete B10.A 1x2 mice, before activation using i.p. injection of MCC peptide, DbY peptide, and LPS, thus allowing us to track the survival of T cell populations with both retroviral constructs within the same animal. After 14 d, we assessed T cell survival using flow cytometry. We found that the pV-RV transduced population showed predominance among the transferred T cell population in 3 animals, constituting 78.9%, 77.4%, and

50.92%, while transferred T cells could not be identified in the fourth animal (Fig. 6.17).

This suggests that constitutive expression of VPAC1 is not detrimental to survival of activated T cell populations and may, in fact, provide a slight advantage over control T cells.

231

Figure 6.17. Overexpression of Vipr1 may increase T cell survival. Lymph nodes from 5C.C7 and A1M mice were transduced as described in Figure 6.15. 24 hr after the second round of infection, 250,000 of each pQ-RV-transduced 5C.C7 T cells were combined with pV-RV- transduced A1M T cells and adoptively transferred by retro-orbital injection to B10.A 1x2 mice. The reciprocal pairing of pV-RV-transduced 5C.C7 T cells and pQ-RV-transduced A1M cells were combined and adoptively transferred to a separate set of B10.A 1x2 mice. After 24 hr, mice were injected with 30 g MCC peptide, 30 g DbY peptide, and 2.5 g LPS. After 14 d, spleens were recovered from these mice and analyzed by flow cytometry. (A) Gating strategy for identifying transferred TCR-Tg cells. (B) Frequencies of transferred pQ-RV transduced (yellow, “vector TDx”) and pV-RV transduced (black, “Vipr1 TDx) populations recovered from each animal. Transferred cells could not be identified in “Mouse #4.”

232

6.2.6. Overexpression of VPAC1 in bone marrow may promote peripheral Treg development.

407,417,427 Noting the VIP has been repeatedly shown to promote Treg induction , we found our observation that Vipr1 is robustly downregulated among Treg (Fig. 5.1, 5.5) to be quite striking. Using pQ-RV and pV-RV transduced BM, we generated BM chimeras to study the effects of constitutive Vipr1 expression on the development of Treg. After allowing sufficient time for reconstitution, we analyzed T cell populations within peripheral LN by flow cytometry (Fig. 6.18). We first observed that that, although a range of 19.6-61.0% of lymph node cells were derived from transferred bone marrow, only a small fraction (8.18% at most) were retrovirally transduced, as indicated by BFP positivity (Fig. 6.18A). However, comparing transduced, BFP+ peripheral cell populations, we found that pV-RV transduced BM generated more CD4+CD25+ Treg than pQ-RV transduced cells, although this trend did not reach statistical significance

(Fig. 6.18B-C). We found that, among the CD25+ population, pV-RV transduced cells showed a significant increase in CD44 gMFI (Fig. 6.18D). CD44 expression by Treg has been previously reported to correlate with Foxp3 expression and suppressive

428,429 function , suggesting that VPAC1 expression promotes a greater frequency of Treg as well as induces improved Treg effector function. However, the observation that pQ-RV

BFP+ cells also showed increased CD44 gMFI compared with BFP- populations (Fig.

6.18D) suggests that retroviral transduction itself is in some way contributing to CD44 expression, either by intrinsically modulating CD44 or promoting transduced T cells ability to become antigen experienced. Alternatively, the shift in mean CD44 between

233

transduced and untransduced cell populations may simply be a technical artifact due to differences in cell numbers.

234

Figure 6.18. Vipr1 BM chimeras mice show increased peripheral CD4+CD25+ T cell frequency. Bone marrow from B6 mice was collected and cultured for 24 hr in BMM supplemented with IL-2, IL-6, and SCF. BM was transduced with either pQ-RV or pV-RV using 2 rounds of spinoculation separated by 10 hours. Cells were cultured with puromycin (1 g/mL) for 24 hr. Cells were then transferred by retro-orbital injection to irradiated (600 rad) TCR-KO mice. Bone marrow was allowed 68 days to reconstitute before LN were collected for flow cytometric analysis. (A) Gating strategy for identifying progeny of transduced (BFP+) and untransduced (BFP-) BM.

235

C B

Figure. 6.18. Vipr1 BM chimeras mice show increased peripheral CD4+CD25+ T cell frequency (cont.). Bone marrow from B6 mice was collected and cultured for 24 hr in BMM supplemented with IL-2, IL-6, and SCF. BM was transduced with either pQ-RV or pV-RV using 2 rounds of spinoculation separated by 10 hours. Cells were cultured with puromycin (1 g/mL) for 24 hr. Cells were then transferred by retro-orbital injection to irradiated (600 rad) TCR-KO mice. Bone marrow was allowed 68 days to reconstitute before LN were collected for flow cytometric analysis. (B) Representative flow plots of CD44 and CD25 expression by CD4+ T cells derived from pQ-RV (upper) and pV-RV (lower) transduced BM. (C) Frequency of CD25 expressing cells among CD4 T cells. (D) CD44 expression by CD4+CD25+ T cells. N=3. Displayed as mean ± SD. * p<0.05 using Student’s T-test.

236

6.3. Significant findings & Discussion

The aim of these studies was to clarify some of the discrepancies regarding the role of VIP signaling to T cells that have been reported. We first used in vitro approaches using isolated T cell populations to study the direct effects of VIP on naïve T cell activation. In line with previous studies, we found that even brief VIP exposure prior to TCR stimulation inhibits T cell activation as measured by CD69 expression and T cell proliferation.

Interestingly, using TCR-Tg T cells, we observed differential effects of VIP on activation between antibody- and antigen-mediated stimulation, with APC/antigen activation showing minimal sensitivity to VIP compared to antibody stimulation. As primary APC express several co-stimulatory molecules, it is likely that the richer array of co-stimulatory signals limited the efficacy of VIP treatment; though this may suggest that VIP acts via modulating the activation threshold for T cells. Using phospho-flow cytometry, we determined that VIP signaling leads to attenuation of ERK signaling in the context of TCR stimulation, suggesting a mechanistic explanation for the observed suppression of activation. We also showed that VIP is able to skew the in vitro differentiation of T cells, with VIP administration yielding increased production of IL-17 and IL-22.

We also developed a retroviral Vipr1 constitutive expression system to investigate the consequences of decreased receptor expression among effector T cells and Treg.

Unexpectedly, we found that effector T cells expressing VPAC1 produced more IFN than controls, rather than providing a suppressive effect as predicted. Further, untransduced bystander T cells also showed increased IFN release, suggesting a potential indirect mechanism underlying the phenomenon. Using our retroviral system in the context of bone marrow chimeras allowed us to investigate the role Vipr1 plays in Treg development in

237

vivo. In line with several studies that report induction of Treg by VIP signaling, we found that constitutive expression of Vipr1 also enhanced the frequency of peripheral Treg. It still remains an open question, however, why Treg downregulate VPAC1 given that its signaling promotes Treg development. In light of our previous finding, showing that Vipr1 expression enhanced IFN release, receptor downregulation may be necessary to maintain the Treg suppressive function, however this hypothesis needs to be further explored.

The role of VIP as a neurotransmitter is well documented. But like many other neurotransmitters, VIP may also be made by other tissues. VIP has a very short half-life in the blood and recent studies on ILC3s find that they potentially derive VIP from contacting neuronal termini directly15. Similarly, the secretion of VIP at different lymph nodes by direct innervation has been documented430-433. In order to trace the cellular events that follow from neuronal activation to immunological activity, it is important show which cell- types are making the neurotransmitter. Furthermore, new approaches90-97 are required for specifically identifying and tracing cells which interacted with the transmitter – and their subsequent fate.

238

Chapter 7: Conclusions and future directions

7.1. Pre-stimulating specific subsets of the TCR signaling pathway alters T cell activation

7.1.1. Significant findings

We used PMA, which activates a subset of the TCR-signaling network, to model

‘Signal 0’ a pre-exposure to environmental stimuli that we hypothesized to alter baseline signaling states and subsequent antigen-driven T cell activating signals. We found that

PMA pre-treatment enhanced T cell activation as measured by expression of CD69 in vitro.

T cell proliferation and cytokine production were largely unaffected, but a brief pre- exposure to the PKC activator had a long-term tangible impact. It led to decreased T cell survival or retention in secondary lymphoid organs. Mechanistically, we found PMA induced robust and prolonged ERK phosphorylation compared with canonical TCR- proximal stimulation (which is rapidly attenuated). Concurrent stimulation favored PMA- mediated ERK phosphorylation kinetics, suggesting that the functional consequences observed are due to unique PMA signaling dynamics that appear to be resistant to TCR- mediated negative feedback. Further, we showed that chronic T cell activation, another form of baseline TCR-signaling modulation, yields two separate hypo-responsive T cell phonotypes: 1) T cells that express high levels of phosphorylated TCR-proximal signaling molecules at baseline that show attenuated increases in phosphorylation with TCR stimulation and 2) T cells that show no TCR-proximal signaling phosphorylation at baseline or in response to TCR engagement.

239

7.1.2. Future directions

The biochemical mechanism by which PKC pre-activation selectively affects pathways relevant to cell survival in vivo need further investigation. Based on the upregulation of CD69, one hypothesis is that longer retention of T cells in the lymph nodes may deprive them of pro-survival signals either due to enhanced competition in this organ or loss of tissue-specific ones. These can be tested using FTY720 or supplementing IL-7,

IL-15 etc. Similarly, it is not clear how PMA-mediated activation signals ERK differently.

Localization of ERK complexes to sequester away from potential negative regulators need to be evaluated using biochemical fractionation or staining experiments.

7.2. Receptors for the non-immune chemiome on T cells

7.2.1. Significant findings

We surveyed for a comprehensive list of receptors on T cells for non-immune chemiome constituents, which would be predicted to contribute to baseline altering signal

0. More than half of the receptors bind growth factors, both hormonally-acting and locally released, while the remaining tended to bind membrane bound ligands, suggesting a role for direct cell-to-cell contact. Most receptors are receptor tyrosine kinases, most commonly coupling to MAPK and PI3K/Akt pathways. Additional signaling modalities triggered by the other rNIC include SMAD signaling and cAMP/PKA via Gs-coupled receptor signaling.

7.2.2. Future directions

Functional consequences of rNIC signaling may be predicted based on knowledge of the coupled signaling pathways, but a careful examination of signaling events triggered by each rNIC within T cells will be critical. Similarly, while we were unable to observe a

240

role for two rNICs we tested (EGFR and INSR) in modulating naïve T cell signaling or activation, it is not clear if other receptors may have roles in specific in vivo contexts. The challenge here will be to develop assays which can measure the impact of rNICs in a high throughput fashion. Genetic studies using CRISPR-libraries may be helpful in this regard.

7.3. Neurotransmitter receptors constitute a major source of rNIC relevant to T cells

7.3.1. Significant findings

Of the 50000 NRs, we identified a core signature of just 15 that all mouse T cells express. These receptors levels are modulated in major T cell lineages and states, suggesting that each subset can be differentially regulated by neurotransmitter signaling.

Intriguingly T cell activation generally led to a rapid downregulation of most core signature

NR, indicating that T cells are less receptive to neuronal regulation in the immediate period following activation. We further found that the core signature was largely unaffected by T cell tissue localization, suggesting that NR expression is intrinsically regulated rather than dictated by tissue cues. We then showed that signaling by specific NR can act as signal 0 to modulate subsequent T cell activation as both 2-adrenergic and histamine H2 receptors resulted in attenuation of early T cell activation markers.

7.3.2. Future directions

The T cell NR expression profile we have established allows now for careful examination of the subset-specific functional consequences of direct signaling of neurotransmitters to T cells. Although many studies have observed the effects of neurotransmitters on a variety of T cell functions, a select few are able to delineate direct action on T cells from indirect effects by modulating other cells, such as DC. Beyond

241

examining the outcomes of NR signaling, a complete understanding of the contexts under which T cells are exposed to neurotransmitters, including the timing and localization of release as well as the cellular source, will be critical for determining the effects on in vivo

T cell responses. This understanding would then allow for the ability to predict the effects of neurological responses and psychoactive therapies on T cell organization of the immune response.

7.4. VIP as a model NIC for subset specific modulation of T cell biology

7.4.1. Significant findings

We found that VIP acting through its receptor VPAC1, similarly to 2AR, inhibits

T cell activation measured both by CD69 expression and proliferation. This effect was observed both in polyclonal T cell populations as well as clonal TCR-Tg lines, though VIP was more effective when TCR-Tg T cells were stimulated with antibodies rather than antigenic peptide, suggesting that the impact of signal 0 provided by VIP depends on the quality of the subsequent activating stimuli. VIP treatment dampened ERK phosphorylation upon TCR-mediated stimulation suggesting an underlying mechanism for the inhibitory effect of VPAC1 signaling. We further examined the effect of VIP on CD4

T cell differentiation, finding that IL-17 and IL-22 release was increased, implicating VIP as a skewing cytokine. Paradoxically, we also found that VPAC1 overexpression in bone marrow resulted in increased frequencies of Treg in peripheral lymph nodes, indicating that

VIP promotes Treg development.

7.4.2. Future directions

The functional effects of VIP on T cells has received much investigation yet conflicting reports have led to consternation in the field. One potential source of this

242

confusion is the fact that many cell types can both produce and respond to VIP, making it difficult to isolate direct and indirect effects on T cells. Further, disease phenotypes that require the collaboration between varying immune cell types may show seemingly opposing impact of VIP signaling, again muddying the literature. Our in vitro studies utilized highly T cell enriched populations, ensuring that the outcomes observed were mediated by direct ligation of T cell expressed VPAC1; however, our in vivo work did not necessarily overcome this limitation, requiring further investigation to understand the precise mechanisms of Treg induction by VIP. Additionally, because VPAC1 signaling acts to dampen T cell activation, it is possible that its expression among naïve T cells is variable, showing expression proportional to TCR reactivity. We plan to use single cell RNA-Seq analysis to examine Vipr1 variability among this population, as well as its correlation with other negative regulators such as CD5434-437.

A major hole in our understanding of VPAC1 its downregulation within the Treg population, as VIP induces Treg and is not detrimental to their survival in vivo. Additionally, cAMP signaling downstream of VPAC1 would be predicted to enforce Foxp3 expression

438-440 and improve Treg suppressive function . It is possible that downregulation of VPAC1 is the result of negative feedback mechanisms, preventing Treg from being overly suppressive, but this hypothesis will require further testing.

7.5. Implications of the NIC as signal 0

7.5.1. Pre-conditioning of T cells by non-immune encounters likely contributes to cell-to- cell response heterogeneity

Aside from differences in TCR specificity, we routinely think of naïve T cells as a uniform population of cells with equal functional potential. However, individual T cells

243

commonly show heterogeneous responses to stimuli, even among a clonal population441.

While these differences are often considered stochastic, the signal 0 model allows us to appreciate a source of variability stemming from the different environmental contexts the

T cells experienced before activation.

7.5.2 Our mechanistic studies offer a rational strategy to anticipate (and perhaps treat) variation in human immune responses

A significant challenge in translating basic and pre-clinical findings to the human immune system has been the confounding variation between groups and even individuals based on genetics, microbiome and environment 442-452. We propose that knowing the signaling pathways used by the non-immune chemiome can help predict their impact on T cell responses. Exposure to EGF, for instance, may dispose T cells to MAPK biasing, whose effects on subsequent T cell activation can be predicted based on our data. On completion of our studies, we can build a framework whereby, knowing which subsets of

T cell pathways are triggered by specific chemiome members can be used to gauge T cell function. In the short term, this has many significant implications. For example, if a drug such as Ritalin (or similar ones which can signal T cells to pre-bias their responses) is likely to affect the nature of an immune response from a vaccine by modulating T cell effector function, then future immunizations may require a period of Ritalin cessation. Importantly, such effects have not been studied and our experiments using defined mouse models are expected to fuel future interest in the broader community.

7.5.3. The insights gained from this work can lead to new pharmacological approaches

At this time, most immunomodulatory agents administered in the clinic are coupled with antigen-exposure (e.g. during vaccination, checkpoint blockade, etc.). Our studies on

244

how specific ligands (e.g. neurotransmitters, drugs) modify T cell responses explore a parameter space before immune stimulation, which can be used to condition the patient.

For example, prior to allogeneic transplant, the graft recipient’s T cells could be conditioned to be hypo-responsive using an understanding of signal 0. Similarly, responses to vaccination could potentially be boosted by conditioning T cells to have a lower threshold for activation.

245

References

1 Takeuchi, O. & Akira, S. Pattern recognition receptors and inflammation. Cell 140, 805-820, doi:10.1016/j.cell.2010.01.022 (2010).

2 Bhat, R. & Steinman, L. Innate and adaptive autoimmunity directed to the central nervous system. Neuron 64, 123-132, doi:10.1016/j.neuron.2009.09.015 (2009).

3 Lanzavecchia, A. & Sallusto, F. Progressive differentiation and selection of the fittest in the immune response. Nature reviews. Immunology 2, 982-987, doi:10.1038/nri959 (2002).

4 Marchingo, J. M. et al. Antigen affinity, costimulation, and cytokine inputs sum linearly to amplify T cell expansion. Science 346, 1123-1127, doi:10.1126/science.1260044 (2014).

5 Curtsinger, J. M., Johnson, C. M. & Mescher, M. F. CD8 T cell clonal expansion and development of effector function require prolonged exposure to antigen, costimulation, and signal 3 cytokine. J.Immunol. 171, 5165-5171 (2003).

6 Curtsinger, J. M., Lins, D. C. & Mescher, M. F. Signal 3 determines tolerance versus full activation of naive CD8 T cells: Dissociating proliferation and development of effector function. Journal of Experimental Medicine 197, 1141- 1151, doi:10.1084/jem.20021910 (2003).

7 Courtney, A. H., Lo, W. L. & Weiss, A. TCR Signaling: Mechanisms of Initiation and Propagation. Trends Biochem Sci 43, 108-123, doi:10.1016/j.tibs.2017.11.008 (2018).

8 Jenkins, M. K. & Schwartz, R. H. Antigen presentation by chemically modified splenocytes induces antigen-specific T cell unresponsiveness in vitro and in vivo. J Exp Med 165, 302-319, doi:10.1084/jem.165.2.302 (1987).

9 Chappert, P. & Schwartz, R. H. Induction of T cell anergy: integration of environmental cues and infectious tolerance. Curr Opin Immunol 22, 552-559, doi:10.1016/j.coi.2010.08.005 (2010).

10 June, C. H., Ledbetter, J. A., Gillespie, M. M., Lindsten, T. & Thompson, C. B. T- cell proliferation involving the CD28 pathway is associated with cyclosporine-

246

resistant interleukin 2 gene expression. Mol Cell Biol 7, 4472-4481, doi:10.1128/mcb.7.12.4472 (1987).

11 Mueller, D. L., Jenkins, M. K. & Schwartz, R. H. Clonal expansion versus functional clonal inactivation: a costimulatory signalling pathway determines the outcome of T cell antigen receptor occupancy. Annu Rev Immunol 7, 445-480, doi:10.1146/annurev.iy.07.040189.002305 (1989).

12 Baxter, A. G. & Hodgkin, P. D. Activation rules: the two-signal theories of immune activation. Nat.Rev.Immunol. 2, 439-446 (2002).

13 Bretscher, P. & Cohn, M. A theory of self-nonself discrimination. Science 169, 1042-1049, doi:10.1126/science.169.3950.1042 (1970).

14 Aruffo, A. & Seed, B. Molecular cloning of a CD28 cDNA by a high-efficiency COS cell expression system. Proc Natl Acad Sci U S A 84, 8573-8577, doi:10.1073/pnas.84.23.8573 (1987).

15 Harding, F. A., McArthur, J. G., Gross, J. A., Raulet, D. H. & Allison, J. P. CD28- mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature 356, 607-609, doi:10.1038/356607a0 (1992).

16 Chen, L. & Flies, D. B. Molecular mechanisms of T cell co-stimulation and co- inhibition. Nat Rev Immunol 13, 227-242, doi:10.1038/nri3405 (2013).

17 Lederberg, J. Genes and antibodies. Science 129, 1649-1653, doi:10.1126/science.129.3364.1649 (1959).

18 Dresser, D. W. Specific inhibition of antibody production. II. Paralysis induced in adult mice by small quantities of protein antigen. Immunology 5, 378-388 (1962).

19 Landsteiner, K. & van der Scheer, J. On the Influence of Acid Groups on the Serological Specificity of Azoproteins. J Exp Med 45, 1045-1056, doi:10.1084/jem.45.6.1045 (1927).

20 Landsteiner, K. & van der Scheer, J. On the Specificity of Serological Reactions with Simple Chemical Compounds (Inhibition Reactions). J Exp Med 54, 295-305, doi:10.1084/jem.54.3.295 (1931).

247

21 Landsteiner, K. & van der Scheer, J. On the Serological Specificity of Peptides. J Exp Med 55, 781-796, doi:10.1084/jem.55.5.781 (1932).

22 Landsteiner, K. & van der Scheer, J. On the Serological Specificity of Peptides. Ii. J Exp Med 59, 769-780, doi:10.1084/jem.59.6.769 (1934).

23 Landsteiner, K. & van der Scheer, J. On the Serological Specificity of Peptides. Iii. J Exp Med 69, 705-719, doi:10.1084/jem.69.5.705 (1939).

24 Talmage, D. W. & Pearlman, D. S. The antibody response: a model based on antagonistic actions of antigen. J Theor Biol 5, 321-339, doi:10.1016/0022- 5193(63)90067-5 (1963).

25 Mitchison, N. A. Antigen recognition responsible for the induction in vitro of the secondary response. Cold Spring Harb Symp Quant Biol 32, 431-439, doi:10.1101/SQB.1967.032.01.055 (1967).

26 Rajewsky, K. R., E. Tolerance specificity and the immune response to lactic dehydrogenase isoenzymes. Cold Spring Harb Symp Quant Biol 32, 547-554, doi:10.1101/SQB.1967.032.01.066 (1967).

27 Bretscher, P. A. & Cohn, M. Minimal model for the mechanism of antibody induction and paralysis by antigen. Nature 220, 444-448, doi:10.1038/220444a0 (1968).

28 Burnet, F. M. The clonal selection theory of acquired immunity. (Vanderbilt University Press, 1959).

29 Claman, H. N., Chaperon, E. A. & Triplett, R. F. Thymus-marrow cell combinations. Synergism in antibody production. Proc Soc Exp Biol Med 122, 1167-1171, doi:10.3181/00379727-122-31353 (1966).

30 Mitchell, G. F. & Miller, J. F. Immunological activity of thymus and thoracic-duct lymphocytes. Proc Natl Acad Sci U S A 59, 296-303, doi:10.1073/pnas.59.1.296 (1968).

31 Mitchison, N. A. T-cell-B-cell cooperation. Nat Rev Immunol 4, 308-312, doi:10.1038/nri1334 (2004).

248

32 Lafferty, K. J. & Jones, M. A. Reactions of the graft versus host (GVH) type. Aust J Exp Biol Med Sci 47, 17-54, doi:10.1038/icb.1969.3 (1969).

33 Lafferty, K. J., Misko, I. S. & Cooley, M. A. Allogeneic stimulation modulates the in vitro response of T cells to transplantation antigen. Nature 249, 275-276, doi:10.1038/249275a0 (1974).

34 Lafferty, K. J. & Cunningham, A. J. A new analysis of allogeneic interactions. Aust J Exp Biol Med Sci 53, 27-42, doi:10.1038/icb.1975.3 (1975).

35 Zinkernagel, R. M. & Doherty, P. C. Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system. Nature 248, 701-702, doi:10.1038/248701a0 (1974).

36 Janeway, C. A., Jr. Approaching the asymptote? Evolution and revolution in immunology. Cold Spring Harb Symp Quant Biol 54 Pt 1, 1-13, doi:10.1101/sqb.1989.054.01.003 (1989).

37 Matzinger, P. Tolerance, danger, and the extended family. Annu Rev Immunol 12, 991-1045, doi:10.1146/annurev.iy.12.040194.005015 (1994).

38 Sellge, G. & Kufer, T. A. PRR-signaling pathways: Learning from microbial tactics. Semin Immunol 27, 75-84, doi:10.1016/j.smim.2015.03.009 (2015).

39 Reis e Sousa, C. Activation of dendritic cells: translating innate into adaptive immunity. Curr Opin Immunol 16, 21-25, doi:10.1016/j.coi.2003.11.007 (2004).

40 Crotty, S. T follicular helper cell differentiation, function, and roles in disease. Immunity 41, 529-542, doi:10.1016/j.immuni.2014.10.004 (2014).

41 Leung, S. et al. The cytokine milieu in the interplay of pathogenic Th1/Th17 cells and regulatory T cells in autoimmune disease. Cell Mol Immunol 7, 182-189, doi:10.1038/cmi.2010.22 (2010).

42 Luckheeram, R. V., Zhou, R., Verma, A. D. & Xia, B. CD4(+)T cells: differentiation and functions. Clin Dev Immunol 2012, 925135, doi:10.1155/2012/925135 (2012).

249

43 Rochman, Y., Spolski, R. & Leonard, W. J. New insights into the regulation of T cells by gamma(c) family cytokines. Nat Rev Immunol 9, 480-490, doi:10.1038/nri2580 (2009).

44 Schluns, K. S. & Lefrancois, L. Cytokine control of memory T-cell development and survival. Nat Rev Immunol 3, 269-279, doi:10.1038/nri1052 (2003).

45 Rathmell, J. C., Farkash, E. A., Gao, W. & Thompson, C. B. IL-7 enhances the survival and maintains the size of naive T cells. J Immunol 167, 6869-6876, doi:10.4049/jimmunol.167.12.6869 (2001).

46 Surh, C. D. & Sprent, J. Homeostasis of naive and memory T cells. Immunity 29, 848-862, doi:10.1016/j.immuni.2008.11.002 (2008).

47 Rubinstein, M. P. et al. IL-7 and IL-15 differentially regulate CD8+ T-cell subsets during contraction of the immune response. Blood 112, 3704-3712, doi:10.1182/blood-2008-06-160945 (2008).

48 Nurieva, R. I. & Chung, Y. Understanding the development and function of T follicular helper cells. Cell Mol Immunol 7, 190-197, doi:10.1038/cmi.2010.24 (2010).

49 Curtsinger, J. M. et al. Inflammatory cytokines provide a third signal for activation of naive CD4+ and CD8+ T cells. J Immunol 162, 3256-3262 (1999).

50 Valenzuela, J., Schmidt, C. & Mescher, M. The roles of IL-12 in providing a third signal for clonal expansion of naive CD8 T cells. J Immunol 169, 6842-6849, doi:10.4049/jimmunol.169.12.6842 (2002).

51 Curtsinger, J. M., Valenzuela, J. O., Agarwal, P., Lins, D. & Mescher, M. F. Type I IFNs provide a third signal to CD8 T cells to stimulate clonal expansion and differentiation. J Immunol 174, 4465-4469, doi:10.4049/jimmunol.174.8.4465 (2005).

52 Murphy, K. & Weaver, C. Janeway's immunobiology. 9th edition. edn, (Garland Science/Taylor & Francis Group, LLC, 2016).

53 Vignali, D. A. & Kuchroo, V. K. IL-12 family cytokines: immunological playmakers. Nat Immunol 13, 722-728, doi:10.1038/ni.2366 (2012).

250

54 Isaksen, D. E. et al. Requirement for stat5 in thymic stromal lymphopoietin- mediated signal transduction. J Immunol 163, 5971-5977 (1999).

55 Gouilleux, F. et al. Prolactin, growth hormone, erythropoietin and granulocyte- macrophage colony stimulating factor induce MGF-Stat5 DNA binding activity. EMBO J 14, 2005-2013 (1995).

56 Barahmand-pour, F. et al. Colony-stimulating factors and interferon-gamma activate a protein related to MGF-Stat 5 to cause formation of the differentiation- induced factor in myeloid cells. FEBS Lett 360, 29-33, doi:10.1016/0014- 5793(95)00072-h (1995).

57 Mui, A. L., Wakao, H., O'Farrell, A. M., Harada, N. & Miyajima, A. Interleukin-3, granulocyte-macrophage colony stimulating factor and interleukin-5 transduce signals through two STAT5 homologs. EMBO J 14, 1166-1175 (1995).

58 Egwuagu, C. E. STAT3 in CD4+ T helper cell differentiation and inflammatory diseases. Cytokine 47, 149-156, doi:10.1016/j.cyto.2009.07.003 (2009).

59 Oh, S. A. & Li, M. O. TGF-beta: guardian of T cell function. J Immunol 191, 3973- 3979, doi:10.4049/jimmunol.1301843 (2013).

60 Massague, J. TGFbeta signalling in context. Nat Rev Mol Cell Biol 13, 616-630, doi:10.1038/nrm3434 (2012).

61 Gorelik, L., Constant, S. & Flavell, R. A. Mechanism of transforming growth factor beta-induced inhibition of T helper type 1 differentiation. J Exp Med 195, 1499- 1505, doi:10.1084/jem.20012076 (2002).

62 Lin, J. T., Martin, S. L., Xia, L. & Gorham, J. D. TGF-beta 1 uses distinct mechanisms to inhibit IFN-gamma expression in CD4+ T cells at priming and at recall: differential involvement of Stat4 and T-bet. J Immunol 174, 5950-5958, doi:10.4049/jimmunol.174.10.5950 (2005).

63 Park, I. K., Shultz, L. D., Letterio, J. J. & Gorham, J. D. TGF-beta1 inhibits T-bet induction by IFN-gamma in murine CD4+ T cells through the protein tyrosine phosphatase Src homology region 2 domain-containing phosphatase-1. J Immunol 175, 5666-5674, doi:10.4049/jimmunol.175.9.5666 (2005).

251

64 Gorelik, L., Fields, P. E. & Flavell, R. A. Cutting edge: TGF-beta inhibits Th type 2 development through inhibition of GATA-3 expression. J Immunol 165, 4773- 4777, doi:10.4049/jimmunol.165.9.4773 (2000).

65 Chen, W. et al. Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J Exp Med 198, 1875-1886, doi:10.1084/jem.20030152 (2003).

66 Fantini, M. C. et al. Cutting edge: TGF-beta induces a regulatory phenotype in CD4+CD25- T cells through Foxp3 induction and down-regulation of Smad7. J Immunol 172, 5149-5153, doi:10.4049/jimmunol.172.9.5149 (2004).

67 McKarns, S. C., Schwartz, R. H. & Kaminski, N. E. Smad3 is essential for TGF- beta 1 to suppress IL-2 production and TCR-induced proliferation, but not IL-2- induced proliferation. J Immunol 172, 4275-4284, doi:10.4049/jimmunol.172.7.4275 (2004).

68 Boussiotis, V. A. et al. Altered T-cell receptor + CD28-mediated signaling and blocked cell cycle progression in interleukin 10 and transforming growth factor- beta-treated alloreactive T cells that do not induce graft-versus-host disease. Blood 97, 565-571, doi:10.1182/blood.v97.2.565 (2001).

69 Chen, C. H. et al. Transforming growth factor beta blocks Tec kinase phosphorylation, Ca2+ influx, and NFATc translocation causing inhibition of T cell differentiation. J Exp Med 197, 1689-1699, doi:10.1084/jem.20021170 (2003).

70 Mani, V. et al. Migratory DCs activate TGF-beta to precondition naive CD8(+) T cells for tissue-resident memory fate. Science 366, doi:10.1126/science.aav5728 (2019).

71 Morris, R., Kershaw, N. J. & Babon, J. J. The molecular details of cytokine signaling via the JAK/STAT pathway. Protein Sci 27, 1984-2009, doi:10.1002/pro.3519 (2018).

72 Valitutti, S., Muller, S., Dessing, M. & Lanzavecchia, A. Different responses are elicited in cytotoxic T lymphocytes by different levels of T cell receptor occupancy. J Exp Med 183, 1917-1921, doi:10.1084/jem.183.4.1917 (1996).

73 Zehn, D., Lee, S. Y. & Bevan, M. J. Complete but curtailed T-cell response to very low-affinity antigen. Nature 458, 211-214, doi:10.1038/nature07657 (2009).

252

74 Prlic, M., Hernandez-Hoyos, G. & Bevan, M. J. Duration of the initial TCR stimulus controls the magnitude but not functionality of the CD8+ T cell response. J Exp Med 203, 2135-2143, doi:10.1084/jem.20060928 (2006).

75 Knudson, K. M., Goplen, N. P., Cunningham, C. A., Daniels, M. A. & Teixeiro, E. Low-affinity T cells are programmed to maintain normal primary responses but are impaired in their recall to low-affinity ligands. Cell Rep 4, 554-565, doi:10.1016/j.celrep.2013.07.008 (2013).

76 Pham, N. L., Badovinac, V. P. & Harty, J. T. A default pathway of memory CD8 T cell differentiation after dendritic cell immunization is deflected by encounter with inflammatory cytokines during antigen-driven proliferation. J Immunol 183, 2337- 2348, doi:10.4049/jimmunol.0901203 (2009).

77 Restifo, N. P. & Gattinoni, L. Lineage relationship of effector and memory T cells. Curr Opin Immunol 25, 556-563, doi:10.1016/j.coi.2013.09.003 (2013).

78 Yamane, H., Zhu, J. & Paul, W. E. Independent roles for IL-2 and GATA-3 in stimulating naive CD4+ T cells to generate a Th2-inducing cytokine environment. J Exp Med 202, 793-804, doi:10.1084/jem.20051304 (2005).

79 Yamane, H. & Paul, W. E. Early signaling events that underlie fate decisions of naive CD4(+) T cells toward distinct T-helper cell subsets. Immunol Rev 252, 12- 23, doi:10.1111/imr.12032 (2013).

80 Molinero, L. L., Miller, M. L., Evaristo, C. & Alegre, M. L. High TCR stimuli prevent induced regulatory T cell differentiation in a NF-kappaB-dependent manner. J Immunol 186, 4609-4617, doi:10.4049/jimmunol.1002361 (2011).

81 Turner, M. S., Kane, L. P. & Morel, P. A. Dominant role of antigen dose in CD4+Foxp3+ regulatory T cell induction and expansion. J Immunol 183, 4895- 4903, doi:10.4049/jimmunol.0901459 (2009).

82 Navarro, M. N. & Cantrell, D. A. Serine-threonine kinases in TCR signaling. Nat Immunol 15, 808-814, doi:10.1038/ni.2941 (2014).

83 Acuto, O. & Cantrell, D. T cell activation and the cytoskeleton. Annu Rev Immunol 18, 165-184, doi:10.1146/annurev.immunol.18.1.165 (2000).

253

84 Sherman, E., Barr, V. & Samelson, L. E. Super-resolution characterization of TCR- dependent signaling clusters. Immunol Rev 251, 21-35, doi:10.1111/imr.12010 (2013).

85 Bhattacharyya, N. D. & Feng, C. G. Regulation of T Helper Cell Fate by TCR Signal Strength. Frontiers in immunology 11, 624, doi:10.3389/fimmu.2020.00624 (2020).

86 Wu, L., Wei, Q., Brzostek, J. & Gascoigne, N. R. J. Signaling from T cell receptors (TCRs) and chimeric antigen receptors (CARs) on T cells. Cell Mol Immunol 17, 600-612, doi:10.1038/s41423-020-0470-3 (2020).

87 Hwang, J. R., Byeon, Y., Kim, D. & Park, S. G. Recent insights of T cell receptor- mediated signaling pathways for T cell activation and development. Exp Mol Med 52, 750-761, doi:10.1038/s12276-020-0435-8 (2020).

88 Smith-Garvin, J. E., Koretzky, G. A. & Jordan, M. S. T cell activation. Annu Rev Immunol 27, 591-619, doi:10.1146/annurev.immunol.021908.132706 (2009).

89 Peng, S. L., Gerth, A. J., Ranger, A. M. & Glimcher, L. H. NFATc1 and NFATc2 together control both T and B cell activation and differentiation. Immunity 14, 13- 20 (2001).

90 Lee, J. U., Kim, L. K. & Choi, J. M. Revisiting the Concept of Targeting NFAT to Control T Cell Immunity and Autoimmune Diseases. Front Immunol 9, 2747, doi:10.3389/fimmu.2018.02747 (2018).

91 Macian, F. et al. Transcriptional mechanisms underlying lymphocyte tolerance. Cell 109, 719-731 (2002).

92 Manicassamy, S., Gupta, S., Huang, Z. & Sun, Z. Protein Kinase C- -Mediated Signals Enhance CD4+ T Cell Survival by Up-Regulating Bcl-xL. The Journal of Immunology 176, 6709-6716, doi:10.4049/jimmunol.176.11.6709 (2006).

93 Hayden, M. S. & Ghosh, S. NF-kappaB in immunobiology. Cell Res 21, 223-244, doi:10.1038/cr.2011.13 (2011).

94 Oh, H. & Ghosh, S. NF-kappaB: roles and regulation in different CD4(+) T-cell subsets. Immunol Rev 252, 41-51, doi:10.1111/imr.12033 (2013).

254

95 Karin, M. The regulation of AP-1 activity by mitogen-activated protein kinases. J Biol Chem 270, 16483-16486, doi:10.1074/jbc.270.28.16483 (1995).

96 Shaulian, E. & Karin, M. AP-1 as a regulator of cell life and death. Nat Cell Biol 4, E131-136, doi:10.1038/ncb0502-e131 (2002).

97 D'Souza, W. N., Chang, C. F., Fischer, A. M., Li, M. & Hedrick, S. M. The Erk2 MAPK Regulates CD8 T Cell Proliferation and Survival. The Journal of Immunology 181, 7617-7629, doi:10.4049/jimmunol.181.11.7617 (2008).

98 Yang, D. D. et al. Differentiation of CD4+ T cells to Th1 cells requires MAP kinase JNK2. Immunity 9, 575-585 (1998).

99 Dong, C. et al. Defective T cell differentiation in the absence of Jnk1. Science 282, 2092-2095 (1998).

100 Chang, C. F. et al. Polar opposites: Erk direction of CD4 T cell subsets. J Immunol 189, 721-731, doi:10.4049/jimmunol.1103015 (2012).

101 Castellanos, M. C. et al. Expression of the leukocyte early activation antigen CD69 is regulated by the transcription factor AP-1. J Immunol 159, 5463-5473 (1997).

102 Genot, E. M., Parker, P. J. & Cantrell, D. A. Analysis of the role of protein kinase C-alpha, -epsilon, and -zeta in T cell activation. J Biol Chem 270, 9833-9839, doi:10.1074/jbc.270.17.9833 (1995).

103 Bankovich, A. J., Shiow, L. R. & Cyster, J. G. CD69 suppresses sphingosine 1- phosophate receptor-1 (S1P1) function through interaction with membrane helix 4. J Biol Chem 285, 22328-22337, doi:10.1074/jbc.M110.123299 (2010).

104 Hunter, M. C., Teijeira, A. & Halin, C. T Cell Trafficking through Lymphatic Vessels. Front Immunol 7, 613, doi:10.3389/fimmu.2016.00613 (2016).

105 Shiow, L. R. et al. CD69 acts downstream of interferon-alpha/beta to inhibit S1P1 and lymphocyte egress from lymphoid organs. Nature 440, 540-544, doi:10.1038/nature04606 (2006).

255

106 Gaud, G., Lesourne, R. & Love, P. E. Regulatory mechanisms in T cell receptor signalling. Nat Rev Immunol 18, 485-497, doi:10.1038/s41577-018-0020-8 (2018).

107 Rothenberg, E. V. & Ward, S. B. A dynamic assembly of diverse transcription factors integrates activation and cell-type information for interleukin 2 gene regulation. Proc Natl Acad Sci U S A 93, 9358-9365, doi:10.1073/pnas.93.18.9358 (1996).

108 Boomer, J. S. & Green, J. M. An enigmatic tail of CD28 signaling. Cold Spring Harb Perspect Biol 2, a002436, doi:10.1101/cshperspect.a002436 (2010).

109 Holdorf, A. D. et al. Proline residues in CD28 and the Src homology (SH)3 domain of Lck are required for T cell costimulation. J Exp Med 190, 375-384, doi:10.1084/jem.190.3.375 (1999).

110 Raab, M. et al. p56Lck and p59Fyn regulate CD28 binding to phosphatidylinositol 3-kinase, growth factor receptor-bound protein GRB-2, and T cell-specific protein- tyrosine kinase ITK: implications for T-cell costimulation. Proc Natl Acad Sci U S A 92, 8891-8895, doi:10.1073/pnas.92.19.8891 (1995).

111 Cai, Y. C. et al. Selective CD28pYMNM mutations implicate phosphatidylinositol 3-kinase in CD86-CD28-mediated costimulation. Immunity 3, 417-426, doi:10.1016/1074-7613(95)90171-x (1995).

112 Han, J. M., Patterson, S. J. & Levings, M. K. The Role of the PI3K Signaling Pathway in CD4(+) T Cell Differentiation and Function. Front Immunol 3, 245, doi:10.3389/fimmu.2012.00245 (2012).

113 Rogel, A. et al. Akt signaling is critical for memory CD8(+) T-cell development and tumor immune surveillance. Proc Natl Acad Sci U S A 114, E1178-E1187, doi:10.1073/pnas.1611299114 (2017).

114 Esensten, J. H., Helou, Y. A., Chopra, G., Weiss, A. & Bluestone, J. A. CD28 Costimulation: From Mechanism to Therapy. Immunity 44, 973-988, doi:10.1016/j.immuni.2016.04.020 (2016).

115 Tacke, M., Hanke, G., Hanke, T. & Hunig, T. CD28-mediated induction of proliferation in resting T cells in vitro and in vivo without engagement of the T cell receptor: evidence for functionally distinct forms of CD28. Eur J Immunol 27, 239- 247, doi:10.1002/eji.1830270136 (1997).

256

116 Dennehy, K. M. et al. Mitogenic CD28 signals require the exchange factor Vav1 to enhance TCR signaling at the SLP-76-Vav-Itk signalosome. J Immunol 178, 1363- 1371, doi:10.4049/jimmunol.178.3.1363 (2007).

117 Levin, S. E., Zhang, C., Kadlecek, T. A., Shokat, K. M. & Weiss, A. Inhibition of ZAP-70 kinase activity via an analog-sensitive allele blocks T cell receptor and CD28 superagonist signaling. J Biol Chem 283, 15419-15430, doi:10.1074/jbc.M709000200 (2008).

118 Pardigon, N. et al. Delayed and separate costimulation in vitro supports the evidence of a transient "excited" state of CD8+ T cells during activation. J Immunol 164, 4493-4499, doi:10.4049/jimmunol.164.9.4493 (2000).

119 Sckisel, G. D. et al. Out-of-Sequence Signal 3 Paralyzes Primary CD4(+) T-Cell- Dependent Immunity. Immunity 43, 240-250, doi:10.1016/j.immuni.2015.06.023 (2015).

120 Perona-Wright, G., Mohrs, K. & Mohrs, M. Sustained signaling by canonical helper T cell cytokines throughout the reactive lymph node. Nat Immunol 11, 520-526, doi:10.1038/ni.1866 (2010).

121 Mohammadi, M. et al. A tyrosine-phosphorylated carboxy-terminal peptide of the fibroblast growth factor receptor (Flg) is a binding site for the SH2 domain of phospholipase C-gamma 1. Mol Cell Biol 11, 5068-5078, doi:10.1128/mcb.11.10.5068 (1991).

122 Peters, K. G. et al. Point mutation of an FGF receptor abolishes phosphatidylinositol turnover and Ca2+ flux but not mitogenesis. Nature 358, 678- 681, doi:10.1038/358678a0 (1992).

123 Tsuda, T., Kaibuchi, K., Kawahara, Y., Fukuzaki, H. & Takai, Y. Induction of protein kinase C activation and Ca2+ mobilization by fibroblast growth factor in Swiss 3T3 cells. FEBS Lett 191, 205-210, doi:10.1016/0014-5793(85)80009-0 (1985).

124 Byrd, V. M., Ballard, D. W., Miller, G. G. & Thomas, J. W. Fibroblast growth factor-1 (FGF-1) enhances IL-2 production and nuclear translocation of NF- kappaB in FGF receptor-bearing Jurkat T cells. J Immunol 162, 5853-5859 (1999).

257

125 Meij, J. T. et al. Exacerbation of myocardial injury in transgenic mice overexpressing FGF-2 is T cell dependent. Am J Physiol Heart Circ Physiol 282, H547-555, doi:10.1152/ajpheart.01019.2000 (2002).

126 DiToro, D. et al. Insulin-Like Growth Factors Are Key Regulators of T Helper 17 Regulatory T Cell Balance in Autoimmunity. Immunity 52, 650-667 e610, doi:10.1016/j.immuni.2020.03.013 (2020).

127 Cretenet, G. et al. Cell surface Glut1 levels distinguish human CD4 and CD8 T lymphocyte subsets with distinct effector functions. Sci Rep 6, 24129, doi:10.1038/srep24129 (2016).

128 Wu, C. et al. Induction of pathogenic TH17 cells by inducible salt-sensing kinase SGK1. Nature 496, 513-517, doi:10.1038/nature11984 (2013).

129 Yang, Y. H. et al. Salt Sensing by Serum/Glucocorticoid-Regulated Kinase 1 Promotes Th17-like Inflammatory Adaptation of Foxp3(+) Regulatory T Cells. Cell Rep 30, 1515-1529 e1514, doi:10.1016/j.celrep.2020.01.002 (2020).

130 Ornitz, D. M. & Itoh, N. The Fibroblast Growth Factor signaling pathway. Wiley Interdiscip Rev Dev Biol 4, 215-266, doi:10.1002/wdev.176 (2015).

131 Walsh, P. T., Smith, L. M. & O'Connor, R. Insulin-like growth factor-1 activates Akt and Jun N-terminal kinases (JNKs) in promoting the survival of T lymphocytes. Immunology 107, 461-471, doi:10.1046/j.1365-2567.2002.01525.x (2002).

132 Lisowska, K. A. et al. Flow cytometric analysis of STAT5 phosphorylation and CD95 expression in CD4+ T lymphocytes treated with recombinant human erythropoietin. J Recept Signal Transduct Res 31, 241-246, doi:10.3109/10799893.2011.578646 (2011).

133 Sloan-Lancaster, J., Evavold, B. D. & Allen, P. M. Induction of T-cell anergy by altered T-cell-receptor ligand on live antigen-presenting cells. Nature 363, 156- 159, doi:10.1038/363156a0 (1993).

134 Evavold, B. D., Sloan-Lancaster, J. & Allen, P. M. Tickling the TCR: selective T- cell functions stimulated by altered peptide ligands. Immunol Today 14, 602-609, doi:10.1016/0167-5699(93)90200-5 (1993).

258

135 Stefanova, I., Dorfman, J. R. & Germain, R. N. Self-recognition promotes the foreign antigen sensitivity of naive T lymphocytes. Nature 420, 429-434, doi:10.1038/nature01146 (2002).

136 Borovsky, Z., Mishan-Eisenberg, G., Yaniv, E. & Rachmilewitz, J. Serial triggering of T cell receptors results in incremental accumulation of signaling intermediates. J Biol Chem 277, 21529-21536, doi:10.1074/jbc.M201613200 (2002).

137 Zelenika, D. et al. Rejection of H-Y disparate skin grafts by monospecific CD4+ Th1 and Th2 cells: no requirement for CD8+ T cells or B cells. J Immunol 161, 1868-1874 (1998).

138 Oehen, S., Feng, L., Xia, Y., Surh, C. D. & Hedrick, S. M. Antigen compartmentation and T helper cell tolerance induction. J Exp Med 183, 2617- 2626, doi:10.1084/jem.183.6.2617 (1996).

139 Tanchot, C., Barber, D. L., Chiodetti, L. & Schwartz, R. H. Adaptive tolerance of CD4+ T cells in vivo: multiple thresholds in response to a constant level of antigen presentation. J Immunol 167, 2030-2039, doi:10.4049/jimmunol.167.4.2030 (2001).

140 Singh, N. J. & Schwartz, R. H. The strength of persistent antigenic stimulation modulates adaptive tolerance in peripheral CD4+ T cells. J Exp Med 198, 1107- 1117, doi:10.1084/jem.20030913 (2003).

141 Iwama, H. & Gojobori, T. Identification of neurotransmitter receptor genes under significantly relaxed selective constraint by orthologous gene comparisons between humans and rodents. Mol Biol Evol 19, 1891-1901, doi:10.1093/oxfordjournals.molbev.a004013 (2002).

142 Yu, B. et al. Epigenetic landscapes reveal transcription factors that regulate CD8(+) T cell differentiation. Nat Immunol 18, 573-582, doi:10.1038/ni.3706 (2017).

143 Crawford, A. et al. Molecular and Transcriptional Basis of CD4(+) T Cell Dysfunction during Chronic Infection. Immunity 40, 289-302, doi:10.1016/j.immuni.2014.01.005 (2014).

144 Doering, T. A. et al. Network analysis reveals centrally connected genes and pathways involved in CD8+ T cell exhaustion versus memory. Immunity 37, 1130- 1144, doi:10.1016/j.immuni.2012.08.021 (2012).

259

145 Wakim, L. M. et al. The molecular signature of tissue resident memory CD8 T cells isolated from the brain. J Immunol 189, 3462-3471, doi:10.4049/jimmunol.1201305 (2012).

146 Mackay, L. K. et al. The developmental pathway for CD103(+)CD8(+) tissue- resident memory T cells of skin. Nat Immunol 14, 1294-1301, doi:10.1038/ni.2744 (2013).

147 Wang, D. et al. The Transcription Factor Runx3 Establishes Chromatin Accessibility of cis-Regulatory Landscapes that Drive Memory Cytotoxic T Lymphocyte Formation. Immunity 48, 659-674 e656, doi:10.1016/j.immuni.2018.03.028 (2018).

148 Milner, J. J. et al. Runx3 programs CD8(+) T cell residency in non-lymphoid tissues and tumours. Nature 552, 253-257, doi:10.1038/nature24993 (2017).

149 White, J. T. et al. Virtual memory T cells develop and mediate bystander protective immunity in an IL-15-dependent manner. Nat Commun 7, 11291, doi:10.1038/ncomms11291 (2016).

150 Spurlock, C. F., 3rd et al. Profiles of Long Noncoding RNAs in Human Naive and Memory T Cells. J Immunol 199, 547-558, doi:10.4049/jimmunol.1700232 (2017).

151 Tian, Y. et al. Unique phenotypes and clonal expansions of human CD4 effector memory T cells re-expressing CD45RA. Nat Commun 8, 1473, doi:10.1038/s41467-017-01728-5 (2017).

152 Kok, B. P. et al. Discovery of small-molecule enzyme activators by activity-based protein profiling. Nat Chem Biol, doi:10.1038/s41589-020-0555-4 (2020).

153 Albeituni, S. & Stiban, J. Roles of Ceramides and Other Sphingolipids in Immune Cell Function and Inflammation. Adv Exp Med Biol 1161, 169-191, doi:10.1007/978-3-030-21735-8_15 (2019).

154 Bieberich, E. Ceramide signaling in cancer and stem cells. Future Lipidol 3, 273- 300, doi:10.2217/17460875.3.3.273 (2008).

260

155 Bi, L. et al. Saturated fatty acids activate ERK signaling to downregulate hepatic sortilin 1 in obese and diabetic mice. J Lipid Res 54, 2754-2762, doi:10.1194/jlr.M039347 (2013).

156 Raines, M. A., Kolesnick, R. N. & Golde, D. W. Sphingomyelinase and ceramide activate mitogen-activated protein kinase in myeloid HL-60 cells. J Biol Chem 268, 14572-14575 (1993).

157 Dobrowsky, R. T., Kamibayashi, C., Mumby, M. C. & Hannun, Y. A. Ceramide activates heterotrimeric protein phosphatase 2A. J Biol Chem 268, 15523-15530 (1993).

158 Molano, A. et al. Age-dependent changes in the sphingolipid composition of mouse CD4+ T cell membranes and immune synapses implicate glucosylceramides in age- related T cell dysfunction. PLoS One 7, e47650, doi:10.1371/journal.pone.0047650 (2012).

159 Holden, N. S. et al. Phorbol ester-stimulated NF-kappaB-dependent transcription: roles for isoforms of novel protein kinase C. Cell Signal 20, 1338-1348, doi:10.1016/j.cellsig.2008.03.001 (2008).

160 Downward, J., Graves, J. D., Warne, P. H., Rayter, S. & Cantrell, D. A. Stimulation of p21ras upon T-cell activation. Nature 346, 719-723, doi:10.1038/346719a0 (1990).

161 Coffey, R. G. & Hadden, J. W. Phorbol myristate acetate stimulation of lymphocyte guanylate cyclase and cyclic guanosine 3':5'-monophosphate phosphodiesterase and reduction of adenylate cyclase. Cancer Res 43, 150-158 (1983).

162 Chatila, T., Silverman, L., Miller, R. & Geha, R. Mechanisms of T cell activation by the calcium ionophore ionomycin. J Immunol 143, 1283-1289 (1989).

163 Kaldjian, E. et al. Nonequivalent effects of PKC activation by PMA on murine CD4 and CD8 cell-surface expression. FASEB J 2, 2801-2806, doi:10.1096/fasebj.2.12.3261700 (1988).

164 Li, Y. Q. et al. Regulation of lymphotoxin production by the p21ras-raf-MEK-ERK cascade in PHA/PMA-stimulated Jurkat cells. J Immunol 162, 3316-3320 (1999).

261

165 Li, Y. Q., Hii, C. S., Der, C. J. & Ferrante, A. Direct evidence that ERK regulates the production/secretion of interleukin-2 in PHA/PMA-stimulated T lymphocytes. Immunology 96, 524-528, doi:10.1046/j.1365-2567.1999.00724.x (1999).

166 Schnell, F. J., Alberts-Grill, N. & Evavold, B. D. CD8+ T cell responses to a viral escape mutant epitope: active suppression via altered SHP-1 activity. J Immunol 182, 1829-1835, doi:10.4049/jimmunol.0801798 (2009).

167 Stefanova, I. et al. TCR ligand discrimination is enforced by competing ERK positive and SHP-1 negative feedback pathways. Nat Immunol 4, 248-254, doi:10.1038/ni895 (2003).

168 Keyse, S. M. Protein phosphatases and the regulation of mitogen-activated protein kinase signalling. Curr Opin Cell Biol 12, 186-192, doi:10.1016/s0955- 0674(99)00075-7 (2000).

169 Maillet, M. et al. DUSP6 (MKP3) null mice show enhanced ERK1/2 phosphorylation at baseline and increased myocyte proliferation in the heart affecting disease susceptibility. J Biol Chem 283, 31246-31255, doi:10.1074/jbc.M806085200 (2008).

170 Muda, M. et al. The dual specificity phosphatases M3/6 and MKP-3 are highly selective for inactivation of distinct mitogen-activated protein kinases. J Biol Chem 271, 27205-27208, doi:10.1074/jbc.271.44.27205 (1996).

171 Bachmaier, K. et al. Negative regulation of lymphocyte activation and autoimmunity by the molecular adaptor Cbl-b. Nature 403, 211-216, doi:10.1038/35003228 (2000).

172 Jeon, M. S. et al. Essential role of the E3 ubiquitin ligase Cbl-b in T cell anergy induction. Immunity 21, 167-177, doi:10.1016/j.immuni.2004.07.013 (2004).

173 Paolino, M. & Penninger, J. M. Cbl-b in T-cell activation. Semin Immunopathol 32, 137-148, doi:10.1007/s00281-010-0197-9 (2010).

174 San Jose, E., Borroto, A., Niedergang, F., Alcover, A. & Alarcon, B. Triggering the TCR complex causes the downregulation of nonengaged receptors by a signal transduction-dependent mechanism. Immunity 12, 161-170, doi:10.1016/s1074- 7613(00)80169-7 (2000).

262

175 Valitutti, S., Muller, S., Salio, M. & Lanzavecchia, A. Degradation of T cell receptor (TCR)-CD3-zeta complexes after antigenic stimulation. J Exp Med 185, 1859-1864, doi:10.1084/jem.185.10.1859 (1997).

176 Bivona, T. G. & Philips, M. R. Ras pathway signaling on endomembranes. Curr Opin Cell Biol 15, 136-142, doi:10.1016/s0955-0674(03)00016-4 (2003).

177 Mor, A. & Philips, M. R. Compartmentalized Ras/MAPK signaling. Annu Rev Immunol 24, 771-800, doi:10.1146/annurev.immunol.24.021605.090723 (2006).

178 Yao, Z. & Seger, R. The ERK signaling cascade--views from different subcellular compartments. Biofactors 35, 407-416, doi:10.1002/biof.52 (2009).

179 Daniels, M. A. et al. Thymic selection threshold defined by compartmentalization of Ras/MAPK signalling. Nature 444, 724-729, doi:10.1038/nature05269 (2006).

180 Teixeiro, E. & Daniels, M. A. ERK and cell death: ERK location and T cell selection. FEBS J 277, 30-38, doi:10.1111/j.1742-4658.2009.07368.x (2010).

181 Nolan, J. P. & Condello, D. Spectral flow cytometry. Curr Protoc Cytom Chapter 1, Unit1 27, doi:10.1002/0471142956.cy0127s63 (2013).

182 Schmutz, S., Valente, M., Cumano, A. & Novault, S. Spectral Cytometry Has Unique Properties Allowing Multicolor Analysis of Cell Suspensions Isolated from Solid Tissues. PLoS One 11, e0159961, doi:10.1371/journal.pone.0159961 (2016).

183 Khan, O. et al. Regulation of T cell priming by lymphoid stroma. PLoS One 6, e26138, doi:10.1371/journal.pone.0026138 (2011).

184 Lukacs-Kornek, V. et al. Regulated release of nitric oxide by nonhematopoietic stroma controls expansion of the activated T cell pool in lymph nodes. Nat Immunol 12, 1096-1104, doi:10.1038/ni.2112 (2011).

185 Siegert, S. et al. Fibroblastic reticular cells from lymph nodes attenuate T cell expansion by producing nitric oxide. PLoS One 6, e27618, doi:10.1371/journal.pone.0027618 (2011).

263

186 Zhou, G. X., Meier, K. E. & Buse, M. G. Sequential activation of two mitogen activated protein (MAP) kinase isoforms in rat skeletal muscle following insulin injection. Biochem Biophys Res Commun 197, 578-584, doi:10.1006/bbrc.1993.2518 (1993).

187 Boucher, J., Kleinridders, A. & Kahn, C. R. Insulin receptor signaling in normal and insulin-resistant states. Cold Spring Harb Perspect Biol 6, doi:10.1101/cshperspect.a009191 (2014).

188 Junger, W. G. et al. Hypertonic saline activates protein tyrosine kinases and mitogen-activated protein kinase p38 in T-cells. J Trauma 42, 437-443; discussion 443-435 (1997).

189 Rincon, M. & Davis, R. J. Regulation of the immune response by stress-activated protein kinases. Immunol Rev 228, 212-224, doi:10.1111/j.1600- 065X.2008.00744.x (2009).

190 Graham, D. K., DeRyckere, D., Davies, K. D. & Earp, H. S. The TAM family: phosphatidylserine sensing receptor tyrosine kinases gone awry in cancer. Nat Rev Cancer 14, 769-785, doi:10.1038/nrc3847 (2014).

191 Lisabeth, E. M., Falivelli, G. & Pasquale, E. B. Eph receptor signaling and ephrins. Cold Spring Harb Perspect Biol 5, doi:10.1101/cshperspect.a009159 (2013).

192 Egea, J. & Klein, R. Bidirectional Eph-ephrin signaling during axon guidance. Trends Cell Biol 17, 230-238, doi:10.1016/j.tcb.2007.03.004 (2007).

193 Genander, M. & Frisen, J. Ephrins and Eph receptors in stem cells and cancer. Curr Opin Cell Biol 22, 611-616, doi:10.1016/j.ceb.2010.08.005 (2010).

194 Genander, M. & Frisen, J. Eph receptors tangled up in two: Independent control of cell positioning and proliferation. Cell Cycle 9, 1865-1866, doi:10.4161/cc.9.10.11677 (2010).

195 Gelfand, M. V. et al. Neuropilin-1 functions as a VEGFR2 co-receptor to guide developmental angiogenesis independent of ligand binding. Elife 3, e03720, doi:10.7554/eLife.03720 (2014).

264

196 Roy, S. et al. Multifaceted Role of Neuropilins in the Immune System: Potential Targets for Immunotherapy. Front Immunol 8, 1228, doi:10.3389/fimmu.2017.01228 (2017).

197 Sharma, A., Verhaagen, J. & Harvey, A. R. Receptor complexes for each of the Class 3 Semaphorins. Front Cell Neurosci 6, 28, doi:10.3389/fncel.2012.00028 (2012).

198 Herzog, B., Pellet-Many, C., Britton, G., Hartzoulakis, B. & Zachary, I. C. VEGF binding to NRP1 is essential for VEGF stimulation of endothelial cell migration, complex formation between NRP1 and VEGFR2, and signaling via FAK Tyr407 phosphorylation. Mol Biol Cell 22, 2766-2776, doi:10.1091/mbc.E09-12-1061 (2011).

199 Berry, M. D., Gainetdinov, R. R., Hoener, M. C. & Shahid, M. Pharmacology of human trace amine-associated receptors: Therapeutic opportunities and challenges. Pharmacol Ther 180, 161-180, doi:10.1016/j.pharmthera.2017.07.002 (2017).

200 Lam, V. M., Espinoza, S., Gerasimov, A. S., Gainetdinov, R. R. & Salahpour, A. In-vivo pharmacology of Trace-Amine Associated Receptor 1. Eur J Pharmacol 763, 136-142, doi:10.1016/j.ejphar.2015.06.026 (2015).

201 Alto, L. T. & Terman, J. R. Semaphorins and their Signaling Mechanisms. Methods Mol Biol 1493, 1-25, doi:10.1007/978-1-4939-6448-2_1 (2017).

202 Bazzazi, H., Isenberg, J. S. & Popel, A. S. Inhibition of VEGFR2 Activation and Its Downstream Signaling to ERK1/2 and Calcium by Thrombospondin-1 (TSP1): In silico Investigation. Front Physiol 8, 48, doi:10.3389/fphys.2017.00048 (2017).

203 Shibuya, M. Vascular Endothelial Growth Factor (VEGF) and Its Receptor (VEGFR) Signaling in Angiogenesis: A Crucial Target for Anti- and Pro- Angiogenic Therapies. Genes Cancer 2, 1097-1105, doi:10.1177/1947601911423031 (2011).

204 Chapman, N. M., Connolly, S. F., Reinl, E. L. & Houtman, J. C. Focal adhesion kinase negatively regulates Lck function downstream of the T cell antigen receptor. J Immunol 191, 6208-6221, doi:10.4049/jimmunol.1301587 (2013).

265

205 Chapman, N. M. & Houtman, J. C. Functions of the FAK family kinases in T cells: beyond actin cytoskeletal rearrangement. Immunol Res 59, 23-34, doi:10.1007/s12026-014-8527-y (2014).

206 Fischer, H. J. et al. The Insulin Receptor Plays a Critical Role in T Cell Function and Adaptive Immunity. J Immunol 198, 1910-1920, doi:10.4049/jimmunol.1601011 (2017).

207 Helderman, J. H. & Strom, T. B. Emergence of insulin receptors upon alloimmune T cells in the rat. J Clin Invest 59, 338-344, doi:10.1172/JCI108646 (1977).

208 McInerney, M. F. et al. High density insulin receptor-positive T lymphocytes from nonobese diabetic mice transfer insulitis and diabetes. J Immunol 157, 3716-3726 (1996).

209 Tsai, S. et al. Insulin Receptor-Mediated Stimulation Boosts T Cell Immunity during Inflammation and Infection. Cell Metab 28, 922-934 e924, doi:10.1016/j.cmet.2018.08.003 (2018).

210 Komarowska, I. et al. Hepatocyte Growth Factor Receptor c-Met Instructs T Cell Cardiotropism and Promotes T Cell Migration to the Heart via Autocrine Chemokine Release. Immunity 42, 1087-1099, doi:10.1016/j.immuni.2015.05.014 (2015).

211 Benkhoucha, M., Senoner, I. & Lalive, P. H. c-Met is expressed by highly autoreactive encephalitogenic CD8+ cells. J Neuroinflammation 17, 68, doi:10.1186/s12974-019-1676-0 (2020).

212 Adams, D. H. et al. Hepatocyte growth factor and macrophage inflammatory protein 1 beta: structurally distinct cytokines that induce rapid cytoskeletal changes and subset-preferential migration in T cells. Proc Natl Acad Sci U S A 91, 7144- 7148, doi:10.1073/pnas.91.15.7144 (1994).

213 Organ, S. L. et al. Quantitative phospho-proteomic profiling of hepatocyte growth factor (HGF)-MET signaling in colorectal cancer. J Proteome Res 10, 3200-3211, doi:10.1021/pr200238t (2011).

214 Organ, S. L. & Tsao, M. S. An overview of the c-MET signaling pathway. Ther Adv Med Oncol 3, S7-S19, doi:10.1177/1758834011422556 (2011).

266

215 Boccaccio, C. et al. Induction of epithelial tubules by growth factor HGF depends on the STAT pathway. Nature 391, 285-288, doi:10.1038/34657 (1998).

216 Stentz, F. B. & Kitabchi, A. E. Transcriptome and proteome expression in activated human CD4 and CD8 T-lymphocytes. Biochem Biophys Res Commun 324, 692- 696, doi:10.1016/j.bbrc.2004.09.113 (2004).

217 Schillaci, R., Brocardo, M. G., Galeano, A. & Roldan, A. Downregulation of insulin-like growth factor-1 receptor (IGF-1R) expression in human T lymphocyte activation. Cell Immunol 183, 157-161, doi:10.1006/cimm.1997.1237 (1998).

218 Johannesson, B. et al. Insulin-like growth factor-1 induces regulatory T cell- mediated suppression of allergic contact dermatitis in mice. Dis Model Mech 7, 977-985, doi:10.1242/dmm.015362 (2014).

219 Bilbao, D., Luciani, L., Johannesson, B., Piszczek, A. & Rosenthal, N. Insulin-like growth factor-1 stimulates regulatory T cells and suppresses autoimmune disease. EMBO Mol Med 6, 1423-1435, doi:10.15252/emmm.201303376 (2014).

220 Kooijman, R. K., Scholtens, L. E., Rijkers, G. T. & Zegers, B. J. Differential expression of type I insulin-like growth factor receptors in different stages of human T cells. Eur J Immunol 25, 931-935, doi:10.1002/eji.1830250411 (1995).

221 Segretin, M. E., Galeano, A., Roldan, A. & Schillaci, R. Insulin-like growth factor- 1 receptor regulation in activated human T lymphocytes. Horm Res 59, 276-280, doi:10.1159/000070625 (2003).

222 Xu, X., Mardell, C., Xian, C. J., Zola, H. & Read, L. C. Expression of functional insulin-like growth factor-1 receptor on lymphoid cell subsets of rats. Immunology 85, 394-399 (1995).

223 Kooijman, R., Scholtens, L. E., Rijkers, G. T. & Zegers, B. J. Type I insulin-like growth factor receptor expression in different developmental stages of human thymocytes. J Endocrinol 147, 203-209, doi:10.1677/joe.0.1470203 (1995).

224 Hindmarsh, E. J., Staykova, M. A., Willenborg, D. O. & Parish, C. R. Cell surface expression of the 300 kDa mannose-6-phosphate receptor by activated T lymphocytes. Immunol Cell Biol 79, 436-443, doi:10.1046/j.1440- 1711.2001.01026.x (2001).

267

225 Ahmed, K. A., Wang, L., Griebel, P., Mousseau, D. D. & Xiang, J. Differential expression of mannose-6-phosphate receptor regulates T cell contraction. J Leukoc Biol 98, 313-318, doi:10.1189/jlb.2HI0215-049RR (2015).

226 Yang, G. et al. Insulin-like growth factor 2 enhances regulatory T-cell functions and suppresses food allergy in an experimental model. J Allergy Clin Immunol 133, 1702-1708 e1705, doi:10.1016/j.jaci.2014.02.019 (2014).

227 Miyagawa, I. et al. Induction of Regulatory T Cells and Its Regulation with Insulin- like Growth Factor/Insulin-like Growth Factor Binding Protein-4 by Human Mesenchymal Stem Cells. J Immunol 199, 1616-1625, doi:10.4049/jimmunol.1600230 (2017).

228 Ikushima, H. et al. Internalization of CD26 by mannose 6-phosphate/insulin-like growth factor II receptor contributes to T cell activation. Proc Natl Acad Sci U S A 97, 8439-8444, doi:10.1073/pnas.97.15.8439 (2000).

229 Wood, R. J. & Hulett, M. D. Cell surface-expressed cation-independent mannose 6-phosphate receptor (CD222) binds enzymatically active heparanase independently of mannose 6-phosphate to promote extracellular matrix degradation. J Biol Chem 283, 4165-4176, doi:10.1074/jbc.M708723200 (2008).

230 Byrd, V., Zhao, X. M., McKeehan, W. L., Miller, G. G. & Thomas, J. W. Expression and functional expansion of fibroblast growth factor receptor T cells in rheumatoid synovium and peripheral blood of patients with rheumatoid arthritis. Arthritis Rheum 39, 914-922, doi:10.1002/art.1780390607 (1996).

231 Farahnak, S. et al. Basic Fibroblast Growth Factor 2 Is a Determinant of CD4 T Cell-Airway Smooth Muscle Cell Communication through Membrane Conduits. J Immunol 199, 3086-3093, doi:10.4049/jimmunol.1700164 (2017).

232 Zhao, X. M. et al. Costimulation of human CD4+ T cells by fibroblast growth factor-1 (acidic fibroblast growth factor). J Immunol 155, 3904-3911 (1995).

233 Minutti, C. M. et al. Epidermal Growth Factor Receptor Expression Licenses Type- 2 Helper T Cells to Function in a T Cell Receptor-Independent Fashion. Immunity 47, 710-722 e716, doi:10.1016/j.immuni.2017.09.013 (2017).

268

234 Zaiss, D. M. et al. Amphiregulin enhances regulatory T cell-suppressive function via the epidermal growth factor receptor. Immunity 38, 275-284, doi:10.1016/j.immuni.2012.09.023 (2013).

235 Zeboudj, L. et al. Selective EGF-Receptor Inhibition in CD4(+) T Cells Induces Anergy and Limits Atherosclerosis. J Am Coll Cardiol 71, 160-172, doi:10.1016/j.jacc.2017.10.084 (2018).

236 Wee, P. & Wang, Z. Epidermal Growth Factor Receptor Cell Proliferation Signaling Pathways. Cancers (Basel) 9, doi:10.3390/cancers9050052 (2017).

237 Ehrhard, P. B., Erb, P., Graumann, U., Schmutz, B. & Otten, U. Expression of functional trk tyrosine kinase receptors after T cell activation. J Immunol 152, 2705-2709 (1994).

238 Lambiase, A. et al. Human CD4+ T cell clones produce and release nerve growth factor and express high-affinity nerve growth factor receptors. J Allergy Clin Immunol 100, 408-414, doi:10.1016/s0091-6749(97)70256-2 (1997).

239 Herbrich, S. M. et al. Characterization of TRKA signaling in acute myeloid leukemia. Oncotarget 9, 30092-30105, doi:10.18632/oncotarget.25723 (2018).

240 Marlin, M. C. & Li, G. Biogenesis and function of the NGF/TrkA signaling endosome. Int Rev Cell Mol Biol 314, 239-257, doi:10.1016/bs.ircmb.2014.10.002 (2015).

241 Tian, B. et al. Peripheral blood brain-derived neurotrophic factor level and tyrosine kinase B expression on T lymphocytes in systemic lupus erythematosus: Implications for systemic involvement. Cytokine 123, 154764, doi:10.1016/j.cyto.2019.154764 (2019).

242 Linker, R. A. et al. Thymocyte-derived BDNF influences T-cell maturation at the DN3/DN4 transition stage. Eur J Immunol 45, 1326-1338, doi:10.1002/eji.201444985 (2015).

243 Besser, M. & Wank, R. Cutting edge: clonally restricted production of the neurotrophins brain-derived neurotrophic factor and neurotrophin-3 mRNA by human immune cells and Th1/Th2-polarized expression of their receptors. J Immunol 162, 6303-6306 (1999).

269

244 Maroder, M. et al. Expression of trKB neurotrophin receptor during T cell development. Role of brain derived neurotrophic factor in immature thymocyte survival. J Immunol 157, 2864-2872 (1996).

245 Minichiello, L. TrkB signalling pathways in LTP and learning. Nat Rev Neurosci 10, 850-860, doi:10.1038/nrn2738 (2009).

246 Sekimoto, M. et al. Functional expression of the TrkC gene, encoding a high affinity receptor for NT-3, in antigen-specific T helper type 2 (Th2) cells. Immunol Lett 88, 221-226, doi:10.1016/s0165-2478(03)00080-4 (2003).

247 Jin, W. Roles of TrkC Signaling in the Regulation of Tumorigenicity and Metastasis of Cancer. Cancers (Basel) 12, doi:10.3390/cancers12010147 (2020).

248 Lisowska, K. A., Debska-Slizien, A., Bryl, E., Rutkowski, B. & Witkowski, J. M. Erythropoietin receptor is expressed on human peripheral blood T and B lymphocytes and monocytes and is modulated by recombinant human erythropoietin treatment. Artif Organs 34, 654-662, doi:10.1111/j.1525- 1594.2009.00948.x (2010).

249 Lisowska, K. A., Frackowiak, J. E., Mikosik, A. & Witkowski, J. M. Changes in the Expression of Transcription Factors Involved in Modulating the Expression of EPO-R in Activated Human CD4-Positive Lymphocytes. PLoS One 8, e60326, doi:10.1371/journal.pone.0060326 (2013).

250 Licona, P., Chimal-Monroy, J. & Soldevila, G. Inhibins are the major activin ligands expressed during early thymocyte development. Dev Dyn 235, 1124-1132, doi:10.1002/dvdy.20707 (2006).

251 Tousa, S. et al. Activin-A co-opts IRF4 and AhR signaling to induce human regulatory T cells that restrain asthmatic responses. Proc Natl Acad Sci U S A 114, E2891-E2900, doi:10.1073/pnas.1616942114 (2017).

252 Gongrich, C. et al. ALK4 coordinates extracellular and intrinsic signals to regulate development of cortical somatostatin interneurons. J Cell Biol 219, doi:10.1083/jcb.201905002 (2020).

253 Goh, B. C. et al. Activin receptor type 2A (ACVR2A) functions directly in osteoblasts as a negative regulator of bone mass. J Biol Chem 292, 13809-13822, doi:10.1074/jbc.M117.782128 (2017).

270

254 Ihn, H. J. et al. Identification of Acvr2a as a Th17 cell-specific gene induced during Th17 differentiation. Biosci Biotechnol Biochem 75, 2138-2141, doi:10.1271/bbb.110436 (2011).

255 Olsen, O. E. et al. Activin A inhibits BMP-signaling by binding ACVR2A and ACVR2B. Cell Commun Signal 13, 27, doi:10.1186/s12964-015-0104-z (2015).

256 Martinez, V. G. et al. The BMP Pathway Participates in Human Naive CD4+ T Cell Activation and Homeostasis. PLoS One 10, e0131453, doi:10.1371/journal.pone.0131453 (2015).

257 Goulley, J., Dahl, U., Baeza, N., Mishina, Y. & Edlund, H. BMP4-BMPR1A signaling in beta cells is required for and augments glucose-stimulated insulin secretion. Cell Metab 5, 207-219, doi:10.1016/j.cmet.2007.01.009 (2007).

258 Zhao, G. J. et al. Growth Arrest-Specific 6 Enhances the Suppressive Function of CD4(+)CD25(+) Regulatory T Cells Mainly through Axl Receptor. Mediators Inflamm 2017, 6848430, doi:10.1155/2017/6848430 (2017).

259 Goruppi, S., Ruaro, E., Varnum, B. & Schneider, C. Gas6-mediated survival in NIH3T3 cells activates stress signalling cascade and is independent of Ras. Oncogene 18, 4224-4236, doi:10.1038/sj.onc.1202788 (1999).

260 Bellosta, P., Zhang, Q., Goff, S. P. & Basilico, C. Signaling through the ARK tyrosine kinase receptor protects from apoptosis in the absence of growth stimulation. Oncogene 15, 2387-2397, doi:10.1038/sj.onc.1201419 (1997).

261 Lemke, G. Biology of the TAM receptors. Cold Spring Harb Perspect Biol 5, a009076, doi:10.1101/cshperspect.a009076 (2013).

262 Melaragno, M. G. et al. Gas6 inhibits apoptosis in vascular smooth muscle: role of Axl kinase and Akt. J Mol Cell Cardiol 37, 881-887, doi:10.1016/j.yjmcc.2004.06.018 (2004).

263 Chuckran, C. A., Liu, C., Bruno, T. C., Workman, C. J. & Vignali, D. A. Neuropilin-1: a checkpoint target with unique implications for cancer immunology and immunotherapy. J Immunother Cancer 8, doi:10.1136/jitc-2020-000967 (2020).

271

264 Hwang, J. Y., Sun, Y., Carroll, C. R. & Usherwood, E. J. Neuropilin-1 Regulates the Secondary CD8 T Cell Response to Virus Infection. mSphere 4, doi:10.1128/mSphere.00221-19 (2019).

265 Kalekar, L. A. et al. CD4(+) T cell anergy prevents autoimmunity and generates regulatory T cell precursors. Nat Immunol 17, 304-314, doi:10.1038/ni.3331 (2016).

266 Liu, C. et al. Neuropilin-1 is a T cell memory checkpoint limiting long-term antitumor immunity. Nat Immunol, doi:10.1038/s41590-020-0733-2 (2020).

267 Renand, A. et al. Neuropilin-1 expression characterizes T follicular helper (Tfh) cells activated during B cell differentiation in human secondary lymphoid organs. PLoS One 8, e85589, doi:10.1371/journal.pone.0085589 (2013).

268 Yadav, M. et al. Neuropilin-1 distinguishes natural and inducible regulatory T cells among regulatory T cell subsets in vivo. J Exp Med 209, 1713-1722, S1711-1719, doi:10.1084/jem.20120822 (2012).

269 Sharfe, N., Freywald, A., Toro, A. & Roifman, C. M. Ephrin-A1 induces c-Cbl phosphorylation and EphA receptor down-regulation in T cells. J Immunol 170, 6024-6032, doi:10.4049/jimmunol.170.12.6024 (2003).

270 Freywald, A., Sharfe, N., Miller, C. D., Rashotte, C. & Roifman, C. M. EphA receptors inhibit anti-CD3-induced apoptosis in thymocytes. J Immunol 176, 4066- 4074, doi:10.4049/jimmunol.176.7.4066 (2006).

271 Smith, L. M. et al. EphA3 is induced by CD28 and IGF-1 and regulates cell adhesion. Exp Cell Res 292, 295-303, doi:10.1016/j.yexcr.2003.08.021 (2004).

272 Luo, H. et al. EphrinB1 and EphrinB2 regulate T cell chemotaxis and migration in experimental autoimmune encephalomyelitis and multiple sclerosis. Neurobiol Dis 91, 292-306, doi:10.1016/j.nbd.2016.03.013 (2016).

273 Yu, G., Luo, H., Wu, Y. & Wu, J. EphrinB1 is essential in T-cell-T-cell cooperation during T-cell activation. J Biol Chem 279, 55531-55539, doi:10.1074/jbc.M410814200 (2004).

272

274 Jin, W., Qi, S. & Luo, H. The effect of conditional EFNB1 deletion in the T cell compartment on T cell development and function. BMC Immunol 12, 68, doi:10.1186/1471-2172-12-68 (2011).

275 Yu, G., Mao, J., Wu, Y., Luo, H. & Wu, J. Ephrin-B1 is critical in T-cell development. J Biol Chem 281, 10222-10229, doi:10.1074/jbc.M510320200 (2006).

276 Yu, G., Luo, H., Wu, Y. & Wu, J. Mouse ephrinB3 augments T-cell signaling and responses to T-cell receptor ligation. J Biol Chem 278, 47209-47216, doi:10.1074/jbc.M306659200 (2003).

277 Freywald, A., Sharfe, N., Rashotte, C., Grunberger, T. & Roifman, C. M. The EphB6 receptor inhibits JNK activation in T lymphocytes and modulates T cell receptor-mediated responses. J Biol Chem 278, 10150-10156, doi:10.1074/jbc.M208179200 (2003).

278 Shimoyama, M. et al. T-cell-specific expression of kinase-defective Eph-family receptor protein, EphB6 in normal as well as transformed hematopoietic cells. Growth Factors 18, 63-78, doi:10.3109/08977190009003234 (2000).

279 Sriram, U. et al. Methamphetamine induces trace amine-associated receptor 1 (TAAR1) expression in human T lymphocytes: role in immunomodulation. J Leukoc Biol 99, 213-223, doi:10.1189/jlb.4A0814-395RR (2016).

280 Underhill, S. M. et al. Amphetamines signal through intracellular TAAR1 receptors coupled to Galpha13 and GalphaS in discrete subcellular domains. Mol Psychiatry, doi:10.1038/s41380-019-0469-2 (2019).

281 Pandiella, A., Beguinot, L., Velu, T. J. & Meldolesi, J. Transmembrane signalling at epidermal growth factor receptors overexpressed in NIH 3T3 cells. Phosphoinositide hydrolysis, cytosolic Ca2+ increase and alkalinization correlate with epidermal-growth-factor-induced cell proliferation. Biochem J 254, 223-228, doi:10.1042/bj2540223 (1988).

282 Graus-Porta, D., Beerli, R. R., Daly, J. M. & Hynes, N. E. ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling. EMBO J 16, 1647-1655, doi:10.1093/emboj/16.7.1647 (1997).

273

283 Sweeney, C. et al. Growth factor-specific signaling pathway stimulation and gene expression mediated by ErbB receptors. J Biol Chem 276, 22685-22698, doi:10.1074/jbc.M100602200 (2001).

284 Burrack, A. L., Martinov, T. & Fife, B. T. T Cell-Mediated Beta Cell Destruction: Autoimmunity and Alloimmunity in the Context of Type 1 Diabetes. Front Endocrinol (Lausanne) 8, 343, doi:10.3389/fendo.2017.00343 (2017).

285 Pugliese, A. Autoreactive T cells in type 1 diabetes. J Clin Invest 127, 2881-2891, doi:10.1172/JCI94549 (2017).

286 Mollinedo, F. & Gajate, C. Lipid rafts as signaling hubs in cancer cell survival/death and invasion: implications in tumor progression and therapy. J Lipid Res 61, 611-635, doi:10.1194/jlr.TR119000439 (2020).

287 Zhang, Z. et al. Lipid raft localization of epidermal growth factor receptor alters matrix metalloproteinase-1 expression in SiHa cells via the MAPK/ERK signaling pathway. Oncol Lett 12, 4991-4998, doi:10.3892/ol.2016.5307 (2016).

288 Tracey, K. J. Understanding immunity requires more than immunology. Nat Immunol 11, 561-564, doi:10.1038/ni0710-561 (2010).

289 Veiga-Fernandes, H. & Freitas, A. A. The S(c)ensory Immune System Theory. Trends Immunol, doi:10.1016/j.it.2017.02.007 (2017).

290 Andersson, U. & Tracey, K. J. Reflex principles of immunological homeostasis. Annu Rev Immunol 30, 313-335, doi:10.1146/annurev-immunol-020711-075015 (2012).

291 Ordovas-Montanes, J. et al. The Regulation of Immunological Processes by Peripheral Neurons in Homeostasis and Disease. Trends Immunol 36, 578-604, doi:10.1016/j.it.2015.08.007 (2015).

292 Kipnis, J. Multifaceted interactions between adaptive immunity and the central nervous system. Science 353, 766-771, doi:10.1126/science.aag2638 (2016).

293 Sun, J., Singh, V., Kajino-Sakamoto, R. & Aballay, A. Neuronal GPCR controls innate immunity by regulating noncanonical unfolded protein response genes. Science 332, 729-732, doi:10.1126/science.1203411 (2011).

274

294 Scanzano, A. & Cosentino, M. Adrenergic regulation of innate immunity: a review. Frontiers in pharmacology 6, 171, doi:10.3389/fphar.2015.00171 (2015).

295 Hanoun, M., Maryanovich, M., Arnal-Estape, A. & Frenette, P. S. Neural Regulation of Hematopoiesis, Inflammation, and Cancer. Neuron 86, 360-373, doi:10.1016/j.neuron.2015.01.026 (2015).

296 Reddy, K. C., Andersen, E. C., Kruglyak, L. & Kim, D. H. A Polymorphism in npr- 1 Is a Behavioral Determinant of Pathogen Susceptibility in C-elegans. Science 323, 382-384, doi:10.1126/science.1166527 (2009).

297 Katayama, Y. et al. Signals from the sympathetic nervous system regulate hematopoietic stem cell egress from bone marrow. Cell 124, 407-421, doi:10.1016/j.cell.2005.10.041 (2006).

298 Cardoso, V. et al. Neuronal regulation of type 2 innate lymphoid cells via neuromedin U. Nature 549, 277-281, doi:10.1038/nature23469 (2017).

299 Borovikova, L. V. et al. Vagus nerve stimulation attenuates the systemic inflammatory response to endotoxin. Nature 405, 458-462 (2000).

300 Rosas-Ballina, M. et al. Splenic nerve is required for cholinergic anti inflammatory pathway control of TNF in endotoxemia. Proceedings of the National Academy of Sciences of the United States of America 105, 11008-11013, doi:10.1073/pnas.0803237105 (2008).

301 Klose, C. S. N. et al. The neuropeptide neuromedin U stimulates innate lymphoid cells and type 2 inflammation. Nature 549, 282-286, doi:10.1038/nature23676 (2017).

302 Moriyama, S. et al. beta2-adrenergic receptor-mediated negative regulation of group 2 innate lymphoid cell responses. Science 359, 1056-1061, doi:10.1126/science.aan4829 (2018).

303 Nagashima, H. et al. Neuropeptide CGRP Limits Group 2 Innate Lymphoid Cell Responses and Constrains Type 2 Inflammation. Immunity 51, 682-695 e686, doi:10.1016/j.immuni.2019.06.009 (2019).

275

304 Seillet, C. et al. The neuropeptide VIP confers anticipatory mucosal immunity by regulating ILC3 activity. Nat Immunol 21, 168-177, doi:10.1038/s41590-019- 0567-y (2020).

305 Talbot, J. et al. Feeding-dependent VIP neuron-ILC3 circuit regulates the intestinal barrier. Nature 579, 575-580, doi:10.1038/s41586-020-2039-9 (2020).

306 Wallrapp, A. et al. Calcitonin Gene-Related Peptide Negatively Regulates Alarmin-Driven Type 2 Innate Lymphoid Cell Responses. Immunity 51, 709-723 e706, doi:10.1016/j.immuni.2019.09.005 (2019).

307 Wallrapp, A. et al. The neuropeptide NMU amplifies ILC2-driven allergic lung inflammation. Nature 549, 351-356, doi:10.1038/nature24029 (2017).

308 Talbot, S. et al. Silencing Nociceptor Neurons Reduces Allergic Airway Inflammation. Neuron 87, 341-354, doi:10.1016/j.neuron.2015.06.007 (2015).

309 Baral, P. et al. Nociceptor sensory neurons suppress neutrophil and gammadelta T cell responses in bacterial lung infections and lethal pneumonia. Nat Med 24, 417- 426, doi:10.1038/nm.4501 (2018).

310 Pinho-Ribeiro, F. A. et al. Blocking Neuronal Signaling to Immune Cells Treats Streptococcal Invasive Infection. Cell 173, 1083-1097 e1022, doi:10.1016/j.cell.2018.04.006 (2018).

311 Lai, N. Y. et al. Gut-Innervating Nociceptor Neurons Regulate Peyer's Patch Microfold Cells and SFB Levels to Mediate Salmonella Host Defense. Cell 180, 33-49 e22, doi:10.1016/j.cell.2019.11.014 (2020).

312 Levite, M. Neuropeptides, by direct interaction with T cells, induce cytokine secretion and break the commitment to a distinct T helper phenotype. Proc Natl Acad Sci U S A 95, 12544-12549 (1998).

313 Levite, M. Nervous immunity: neurotransmitters, extracellular K+ and T-cell function. Trends Immunol 22, 2-5 (2001).

314 Rosas-Ballina, M. et al. Acetylcholine-Synthesizing T Cells Relay Neural Signals in a Vagus Nerve Circuit. Science 334, 98-101, doi:10.1126/science.1209985 (2011).

276

315 Reardon, C. et al. Lymphocyte-derived ACh regulates local innate but not adaptive immunity. Proceedings of the National Academy of Sciences of the United States of America 110, 1410-1415, doi:10.1073/pnas.1221655110 (2013).

316 Albuquerque, E. X., Pereira, E. F., Alkondon, M. & Rogers, S. W. Mammalian nicotinic acetylcholine receptors: from structure to function. Physiol Rev 89, 73- 120, doi:10.1152/physrev.00015.2008 (2009).

317 Scarr, E. Muscarinic receptors: their roles in disorders of the central nervous system and potential as therapeutic targets. CNS Neurosci Ther 18, 369-379, doi:10.1111/j.1755-5949.2011.00249.x (2012).

318 Khan, S. M. et al. The expanding roles of Gbetagamma subunits in G protein- coupled receptor signaling and drug action. Pharmacol Rev 65, 545-577, doi:10.1124/pr.111.005603 (2013).

319 Tan, L., Yan, W., McCorvy, J. D. & Cheng, J. Biased Ligands of G Protein-Coupled Receptors (GPCRs): Structure-Functional Selectivity Relationships (SFSRs) and Therapeutic Potential. J Med Chem 61, 9841-9878, doi:10.1021/acs.jmedchem.8b00435 (2018).

320 Betke, K. M., Wells, C. A. & Hamm, H. E. GPCR mediated regulation of synaptic transmission. Prog Neurobiol 96, 304-321, doi:10.1016/j.pneurobio.2012.01.009 (2012).

321 Goetz, T., Arslan, A., Wisden, W. & Wulff, P. GABA(A) receptors: structure and function in the basal ganglia. Prog Brain Res 160, 21-41, doi:10.1016/S0079- 6123(06)60003-4 (2007).

322 North, R. A. P2X receptors. Philos Trans R Soc Lond B Biol Sci 371, doi:10.1098/rstb.2015.0427 (2016).

323 Schmidt, R. W. & Thompson, M. L. Glycinergic signaling in the human nervous system: An overview of therapeutic drug targets and clinical effects. Ment Health Clin 6, 266-276, doi:10.9740/mhc.2016.11.266 (2016).

324 Traynelis, S. F. et al. Glutamate receptor ion channels: structure, regulation, and function. Pharmacol Rev 62, 405-496, doi:10.1124/pr.109.002451 (2010).

277

325 Twomey, E. C. & Sobolevsky, A. I. Structural Mechanisms of Gating in Ionotropic Glutamate Receptors. Biochemistry 57, 267-276, doi:10.1021/acs.biochem.7b00891 (2018).

326 Wu, Z. S., Cheng, H., Jiang, Y., Melcher, K. & Xu, H. E. Ion channels gated by acetylcholine and serotonin: structures, biology, and drug discovery. Acta Pharmacol Sin 36, 895-907, doi:10.1038/aps.2015.66 (2015).

327 Huang, Y., Zheng, Y., Su, Z. & Gu, X. Differences in duplication age distributions between human GPCRs and their downstream genes from a network prospective. BMC Genomics 10 Suppl 1, S14, doi:10.1186/1471-2164-10-S1-S14 (2009).

328 Sanders, V. M. et al. Differential expression of the beta(2)-adrenergic receptor by Th1 and Th2 clones - Implications for cytokine production and B cell help. Journal of Immunology 158, 4200-4210 (1997).

329 Levite, M., Cahalon, L., Hershkoviz, R., Steinman, L. & Lider, O. Neuropeptides, via specific receptors, regulate T cell adhesion to fibronectin. J Immunol 160, 993- 1000 (1998).

330 Swanson, M. A., Lee, W. T. & Sanders, V. M. IFN-gamma production by Th1 cells generated from naive CD4+ T cells exposed to norepinephrine. J Immunol 166, 232-240 (2001).

331 Besser, M. J., Ganor, Y. & Levite, M. Dopamine by itself activates either D2, D3 or D1/D5 dopaminergic receptors in normal human T-cells and triggers the selective secretion of either IL-10, TNFalpha or both. Journal of neuroimmunology 169, 161-171, doi:10.1016/j.jneuroim.2005.07.013 (2005).

332 Bjurstom, H. et al. GABA, a natural immunomodulator of T lymphocytes. J Neuroimmunol 205, 44-50, doi:10.1016/j.jneuroim.2008.08.017 (2008).

333 Levite, M. Neurotransmitters activate T-cells and elicit crucial functions via neurotransmitter receptors. Curr Opin Pharmacol 8, 460-471, doi:10.1016/j.coph.2008.05.001 (2008).

334 Contreras, F. et al. Dopamine Receptor D3 Signaling on CD4+ T Cells Favors Th1- and Th17-Mediated Immunity. J Immunol 196, 4143-4149, doi:10.4049/jimmunol.1502420 (2016).

278

335 Galant, S. P. & Remo, R. A. Beta-adrenergic inhibition of human T lymphocyte rosettes. J Immunol 114, 512-513 (1975).

336 Pochet, R., Delespesse, G., Gausset, P. W. & Collet, H. Distribution of beta- adrenergic receptors on human lymphocyte subpopulations. Clinical and experimental immunology 38, 578-584 (1979).

337 Bishopric, N. H., Cohen, H. J. & Lefkowitz, R. J. Beta adrenergic receptors in lymphocyte subpopulations. The Journal of allergy and clinical immunology 65, 29-33 (1980).

338 Feldman, R. D., Hunninghake, G. W. & McArdle, W. L. Beta-adrenergic-receptor- mediated suppression of interleukin 2 receptors in human lymphocytes. J Immunol 139, 3355-3359 (1987).

339 Ramer-Quinn, D. S., Baker, R. A. & Sanders, V. M. Activated T helper 1 and T helper 2 cells differentially express the beta-2-adrenergic receptor: a mechanism for selective modulation of T helper 1 cell cytokine production. J Immunol 159, 4857-4867 (1997).

340 Sanders, V. M. et al. Differential expression of the beta2-adrenergic receptor by Th1 and Th2 clones: implications for cytokine production and B cell help. J Immunol 158, 4200-4210 (1997).

341 Guereschi, M. G. et al. Beta2-adrenergic receptor signaling in CD4+ Foxp3+ regulatory T cells enhances their suppressive function in a PKA-dependent manner. Eur J Immunol 43, 1001-1012, doi:10.1002/eji.201243005 (2013).

342 Kolmus, K., Tavernier, J. & Gerlo, S. beta2-Adrenergic receptors in immunity and inflammation: stressing NF-kappaB. Brain, behavior, and immunity 45, 297-310, doi:10.1016/j.bbi.2014.10.007 (2015).

343 Estrada, L. D., Agac, D. & Farrar, J. D. Sympathetic neural signaling via the beta2- adrenergic receptor suppresses T-cell receptor-mediated human and mouse CD8(+) T-cell effector function. Eur J Immunol 46, 1948-1958, doi:10.1002/eji.201646395 (2016).

344 Nakai, A., Hayano, Y., Furuta, F., Noda, M. & Suzuki, K. Control of lymphocyte egress from lymph nodes through beta2-adrenergic receptors. J Exp Med 211, 2583- 2598, doi:10.1084/jem.20141132 (2014).

279

345 Maestroni, G. J. M. & Conti, A. Modulation of Hematopoiesis via Alpha-1 Adrenergic receptors on Bone marrow cells. Experimental hematology 22, 313-320 (1994).

346 Nakai, A., Hayano, Y., Furuta, F., Noda, M. & Suzuki, K. Control of lymphocyte egress from lymph nodes through beta(2)-adrenergic receptors. Journal of Experimental Medicine 211, 2583-2598, doi:10.1084/jem.20141132 (2014).

347 Suzuki, K., Hayano, Y., Nakai, A., Furuta, F. & Noda, M. Adrenergic control of the adaptive immune response by diurnal lymphocyte recirculation through lymph nodes. J Exp Med 213, 2567-2574, doi:10.1084/jem.20160723 (2016).

348 Maestroni, G. J. M. Dendritic cell migration controlled by alpha(1b)-adrenergic receptors. Journal of Immunology 165, 6743-6747 (2000).

349 Takenaka, M. C., Guereschi, M. G. & Basso, A. S. Neuroimmune interactions: dendritic cell modulation by the sympathetic nervous system. Seminars in immunopathology 39, 165-176, doi:10.1007/s00281-016-0590-0 (2017).

350 Germain, R. N. The art of the probable: system control in the adaptive immune system. Science 293, 240-245 (2001).

351 Lanzavecchia, A. & Sallusto, F. The instructive role of dendritic cells on T cell responses: lineages, plasticity and kinetics. Curr.Opin.Immunol. 13, 291-298 (2001).

352 Fuchs, B. A., Albright, J. W. & Albright, J. F. Beta-adrenergic receptors on murine lymphocytes: density varies with cell maturity and lymphocyte subtype and is decreased after antigen administration. Cell Immunol 114, 231-245 (1988).

353 Estrada, L. D., Agac, D. & Farrar, J. D. Sympathetic neural signaling via the beta2- adrenergic receptor suppresses T-cell receptor-mediated human and mouse CD8 T- cell effector function. European journal of immunology, doi:10.1002/eji.201646395 (2016).

354 Rosenberg, K. M. & Singh, N. J. Mouse T cells express a neurotransmitter-receptor signature that is quantitatively modulated in a subset- and activation-dependent manner. Brain Behav Immun 80, 275-285, doi:10.1016/j.bbi.2019.04.002 (2019).

280

355 Ganor, Y., Besser, M., Ben-Zakay, N., Unger, T. & Levite, M. Human T cells express a functional ionotropic glutamate receptor GluR3, and glutamate by itself triggers integrin-mediated adhesion to laminin and fibronectin and chemotactic migration. J Immunol 170, 4362-4372 (2003).

356 Ganor, Y., Teichberg, V. I. & Levite, M. TCR activation eliminates glutamate receptor GluR3 from the cell surface of normal human T cells, via an autocrine/paracrine granzyme B-mediated proteolytic cleavage. J Immunol 178, 683-692 (2007).

357 Doering, T. A. et al. Network Analysis Reveals Centrally Connected Genes and Pathways Involved in CD8(+) T Cell Exhaustion versus Memory. Immunity 37, 1130-1144, doi:10.1016/j.immuni.2012.08.021 (2012).

358 Chang, J. T., Wherry, E. J. & Goldrath, A. W. Molecular regulation of effector and memory T cell differentiation. Nat Immunol 15, 1104-1115, doi:10.1038/ni.3031 (2014).

359 Kaech, S. M., Wherry, E. J. & Ahmed, R. Effector and memory T-cell differentiation: Implications for vaccine development. Nature Reviews Immunology 2, 251-262, doi:10.1038/nri778 (2002).

360 Joshi, N. S. et al. Inflammation directs memory precursor and short-lived effector CD8+ T cell fates via the graded expression of T-bet transcription factor. Immunity 27, 281-295, doi:10.1016/j.immuni.2007.07.010 (2007).

361 Yu, B. et al. Epigenetic landscapes reveal transcription factors that regulate CD8+ T cell differentiation. Nat Immunol, doi:10.1038/ni.3706 (2017).

362 Schenkel, J. M. & Masopust, D. Tissue-resident memory T cells. Immunity 41, 886- 897, doi:10.1016/j.immuni.2014.12.007 (2014).

363 Wakim, L. M. et al. The Molecular Signature of Tissue Resident Memory CD8 T Cells Isolated from the Brain. Journal of Immunology 189, 3462-3471, doi:10.4049/jimmunol.1201305 (2012).

364 Henney, C. S. & Lichtenstein, L. M. The role of cyclic AMP in the cytolytic activity of lymphocytes. J Immunol 107, 610-612 (1971).

281

365 Lichtenstein, L. M., Henney, C. S., Bourne, H. R. & Greenough, W. B., 3rd. Effects of cholera toxin on in vitro models of immediate and delayed hypersensitivity. Further evidence for the role of cyclic adenosine 3',5'-monophosphate. J Clin Invest 52, 691-697, doi:10.1172/JCI107230 (1973).

366 Wolberg, G., Zimmerman, T. P., Hiemstra, K., Winston, M. & Chu, L. C. Adenosine inhibition of lymphocyte-mediated cytolysis: possible role of cyclic adenosine monophosphate. Science 187, 957-959, doi:10.1126/science.167434 (1975).

367 Kammer, G. M. The adenylate cyclase-cAMP-protein kinase A pathway and regulation of the immune response. Immunol Today 9, 222-229, doi:10.1016/0167- 5699(88)91220-0 (1988).

368 Hynes, T. R. et al. Inhibition of Galphas/cAMP Signaling Decreases TCR- Stimulated IL-2 transcription in CD4(+) T Helper Cells. J Mol Signal 10, 2, doi:10.5334/1750-2187-10-2 (2015).

369 Staus, D. P. et al. Regulation of beta2-adrenergic receptor function by conformationally selective single-domain intrabodies. Mol Pharmacol 85, 472-481, doi:10.1124/mol.113.089516 (2014).

370 Vayttaden, S. J. et al. Quantitative modeling of GRK-mediated beta2AR regulation. PLoS Comput Biol 6, e1000647, doi:10.1371/journal.pcbi.1000647 (2010).

371 Weinberg, Z. Y. & Puthenveedu, M. A. Regulation of G protein-coupled receptor signaling by plasma membrane organization and endocytosis. Traffic 20, 121-129, doi:10.1111/tra.12628 (2019).

372 Kawasaki, H. et al. A family of cAMP-binding proteins that directly activate Rap1. Science 282, 2275-2279 (1998).

373 Vossler, M. R. et al. cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent pathway. Cell 89, 73-82 (1997).

374 Ramstad, C., Sundvold, V., Johansen, H. K. & Lea, T. cAMP-dependent protein kinase (PKA) inhibits T cell activation by phosphorylating ser-43 of raf-1 in the MAPK/ERK pathway. Cell Signal 12, 557-563, doi:10.1016/s0898- 6568(00)00097-8 (2000).

282

375 Tamir, A., Granot, Y. & Isakov, N. Inhibition of T lymphocyte activation by cAMP is associated with down-regulation of two parallel mitogen-activated protein kinase pathways, the extracellular signal-related kinase and c-Jun N-terminal kinase. J Immunol 157, 1514-1522 (1996).

376 Levite, M. Nerve-driven immunity. The direct effects of neurotransmitters on T- cell function. Annals of the New York Academy of Sciences 917, 307-321 (2000).

377 Mignini, F. et al. T-cell subpopulations express a different pattern of dopaminergic markers in intra- and extra-thymic compartments. Journal of biological regulators and homeostatic agents 27, 463-475 (2013).

378 Takenaka, M. C. et al. Norepinephrine Controls Effector T Cell Differentiation through beta2-Adrenergic Receptor-Mediated Inhibition of NF-kappaB and AP-1 in Dendritic Cells. J Immunol 196, 637-644, doi:10.4049/jimmunol.1501206 (2016).

379 Scheiermann, C. et al. Adrenergic Nerves Govern Circadian Leukocyte Recruitment to Tissues. Immunity 37, 290-301, doi:10.1016/j.immuni.2012.05.021 (2012).

380 Klein, T. W., Newton, C. & Friedman, H. Cannabinoid receptors and the cytokine network. Adv Exp Med Biol 437, 215-222 (1998).

381 Felsner, P. et al. Continuous in vivo treatment with catecholamines suppresses in vitro reactivity of rat peripheral blood T-lymphocytes via alpha-mediated mechanisms. J Neuroimmunol 37, 47-57 (1992).

382 Jutel, M. et al. Histamine regulates T-cell and antibody responses by differential expression of H1 and H2 receptors. Nature 413, 420-425, doi:10.1038/35096564 (2001).

383 Bao, J. Y., Huang, Y., Wang, F., Peng, Y. P. & Qiu, Y. H. Expression of alpha-AR subtypes in T lymphocytes and role of the alpha-ARs in mediating modulation of T cell function. Neuroimmunomodulation 14, 344-353, doi:10.1159/000129670 (2007).

384 Sato, K. Z. et al. Diversity of mRNA expression for muscarinic acetylcholine receptor subtypes and neuronal nicotinic acetylcholine receptor subunits in human mononuclear leukocytes and leukemic cell lines. Neurosci Lett 266, 17-20 (1999).

283

385 Ramseier, H., Lichtensteiger, W. & Schlumpf, M. In vitro inhibition of cellular immune responses by benzodiazepines and PK 11195. Effects on mitogen- and alloantigen-driven lymphocyte proliferation and on IL-1, IL-2 synthesis and IL-2 receptor expression. Immunopharmacol Immunotoxicol 15, 557-582, doi:10.3109/08923979309019731 (1993).

386 Zagon, I. S., Donahue, R. N., Bonneau, R. H. & McLaughlin, P. J. T lymphocyte proliferation is suppressed by the opioid growth factor ([Met(5)]-enkephalin)- opioid growth factor receptor axis: implication for the treatment of autoimmune diseases. Immunobiology 216, 579-590, doi:10.1016/j.imbio.2010.09.014 (2011).

387 Hucklebridge, F. H., Hudspith, B. N., Lydyard, P. M. & Brostoff, J. Stimulation of human peripheral lymphocytes by methionine enkephalin and delta-selective opioid analogues. Immunopharmacology 19, 87-91 (1990).

388 Sorensen, A. N. & Claesson, M. H. Effect of the opioid methionine enkephalinamide on signal transduction in human T-lymphocytes. Life Sci 62, 1251-1259 (1998).

389 Wybran, J., Appelboom, T., Famaey, J. P. & Govaerts, A. Suggestive evidence for receptors for morphine and methionine-enkephalin on normal human blood T lymphocytes. J Immunol 123, 1068-1070 (1979).

390 Woehrle, T. et al. Pannexin-1 hemichannel-mediated ATP release together with P2X1 and P2X4 receptors regulate T-cell activation at the immune synapse. Blood 116, 3475-3484, doi:10.1182/blood-2010-04-277707 (2010).

391 Anholt, R. R., De Souza, E. B., Oster-Granite, M. L. & Snyder, S. H. Peripheral- type benzodiazepine receptors: autoradiographic localization in whole-body sections of neonatal rats. J Pharmacol Exp Ther 233, 517-526 (1985).

392 Benavides, J., Dubois, A., Dennis, T., Hamel, E. & Scatton, B. Omega 3 (peripheral type benzodiazepine binding) site distribution in the rat immune system: an autoradiographic study with the photoaffinity ligand [3H]PK 14105. J Pharmacol Exp Ther 249, 333-339 (1989).

393 Zhu, J., Yamane, H. & Paul, W. E. in Annual Review of Immunology, Vol 28 Vol. 28 Annual Review of Immunology (eds W. E. Paul, D. R. Littman, & W. M. Yokoyama) 445-489 (2010).

284

394 Tubo, N. J. & Jenkins, M. K. TCR signal quantity and quality in CD4 T cell differentiation. Trends Immunol 35, 591-596, doi:10.1016/j.it.2014.09.008 (2014).

395 Tubo, N. J. et al. Single Naive CD4(+) T Cells from a Diverse Repertoire Produce Different Effector Cell Types during Infection. Cell 153, 785-796, doi:10.1016/j.cell.2013.04.007 (2013).

396 Klein, T. W., Kawakami, Y., Newton, C. & Friedman, H. Marijuana components suppress induction and cytolytic function of murine cytotoxic T cells in vitro and in vivo. J Toxicol Environ Health 32, 465-477, doi:10.1080/15287399109531496 (1991).

397 Borges da Silva, H. et al. The purinergic receptor P2RX7 directs metabolic fitness of long-lived memory CD8(+) T cells. Nature 559, 264-268, doi:10.1038/s41586- 018-0282-0 (2018).

398 Scheiermann, C., Kunisaki, Y. & Frenette, P. S. Circadian control of the immune system. Nat Rev Immunol 13, 190-198, doi:10.1038/nri3386 (2013).

399 Druzd, D. et al. Lymphocyte Circadian Clocks Control Lymph Node Trafficking and Adaptive Immune Responses. Immunity 46, 120-132, doi:10.1016/j.immuni.2016.12.011 (2017).

400 Godinho-Silva, C. et al. Light-entrained and brain-tuned circadian circuits regulate ILC3s and gut homeostasis. Nature 574, 254-258, doi:10.1038/s41586-019-1579- 3 (2019).

401 Basu, B. et al. D1 and D2 dopamine receptor-mediated inhibition of activated normal T cell proliferation is lost in jurkat T leukemic cells. The Journal of biological chemistry 285, 27026-27032, doi:10.1074/jbc.M110.144022 (2010).

402 Melnikov, M., Belousova, O., Murugin, V., Pashenkov capital Em, C. & Boysmall ka, C. o. C. A. The role of dopamine in modulation of Th-17 immune response in multiple sclerosis. J Neuroimmunol 292, 97-101, doi:10.1016/j.jneuroim.2016.01.020 (2016).

403 Sanders, V. M. The beta2-adrenergic receptor on T and B lymphocytes: do we understand it yet? Brain Behav Immun 26, 195-200, doi:10.1016/j.bbi.2011.08.001 (2012).

285

404 Yshii, L., Gebauer, C., Bernard-Valnet, R. & Liblau, R. Neurons and T cells: Understanding this interaction for inflammatory neurological diseases. European journal of immunology 45, 2712-2720, doi:10.1002/eji.201545759 (2015).

405 Felten, D. L., Felten, S. Y., Carlson, S. L., Olschowka, J. A. & Livnat, S. Noradrenergic and peptidergic innervation of lymphoid tissue. J Immunol 135, 755s-765s (1985).

406 Gallino, L. et al. VIP Promotes Recruitment of Tregs to the Uterine-Placental Interface During the Peri-Implantation Period to Sustain a Tolerogenic Microenvironment. Frontiers in immunology 10, 2907, doi:10.3389/fimmu.2019.02907 (2019).

407 Jimeno, R. et al. New insights into the role of VIP on the ratio of T-cell subsets during the development of autoimmune diabetes. Immunol Cell Biol 88, 734-745, doi:10.1038/icb.2010.29 (2010).

408 Deng, S. et al. Regulatory effect of vasoactive intestinal peptide on the balance of Treg and Th17 in collagen-induced arthritis. Cell Immunol 265, 105-110, doi:10.1016/j.cellimm.2010.07.010 (2010).

409 Fraccaroli, L. et al. VIP boosts regulatory T cell induction by trophoblast cells in an in vitro model of trophoblast-maternal leukocyte interaction. J Leukoc Biol 98, 49-58, doi:10.1189/jlb.1A1014-492RR (2015).

410 Grasso, E. et al. VIP contribution to the decidualization program: regulatory T cell recruitment. The Journal of endocrinology 221, 121-131, doi:10.1530/joe-13-0565 (2014).

411 Szema, A. M., Hamidi, S. A., Golightly, M. G., Rueb, T. P. & Chen, J. J. VIP Regulates the Development & Proliferation of Treg in vivo in spleen. Allergy, asthma, and clinical immunology : official journal of the Canadian Society of Allergy and Clinical Immunology 7, 19, doi:10.1186/1710-1492-7-19 (2011).

412 Abad, C. et al. Vasoactive intestinal peptide loss leads to impaired CNS parenchymal T-cell infiltration and resistance to experimental autoimmune encephalomyelitis. Proc Natl Acad Sci U S A 107, 19555-19560, doi:10.1073/pnas.1007622107 (2010).

286

413 Yadav, M., Huang, M. C. & Goetzl, E. J. VPAC1 (vasoactive intestinal peptide (VIP) receptor type 1) G protein-coupled receptor mediation of VIP enhancement of murine experimental colitis. Cell Immunol 267, 124-132, doi:10.1016/j.cellimm.2011.01.001 (2011).

414 Tan, Y. V., Abad, C., Wang, Y., Lopez, R. & Waschek, J. VPAC2 (vasoactive intestinal peptide receptor type 2) receptor deficient mice develop exacerbated experimental autoimmune encephalomyelitis with increased Th1/Th17 and reduced Th2/Treg responses. Brain Behav Immun 44, 167-175, doi:10.1016/j.bbi.2014.09.020 (2015).

415 Prasse, A. et al. Inhaled vasoactive intestinal peptide exerts immunoregulatory effects in sarcoidosis. American journal of respiratory and critical care medicine 182, 540-548, doi:10.1164/rccm.200909-1451OC (2010).

416 Korkmaz, O. T. et al. Vasoactive intestinal peptide (VIP) treatment of Parkinsonian rats increases thalamic gamma-aminobutyric acid (GABA) levels and alters the release of nerve growth factor (NGF) by mast cells. Journal of molecular neuroscience : MN 41, 278-287, doi:10.1007/s12031-009-9307-3 (2010).

417 Pozo, D., Anderson, P. & Gonzalez-Rey, E. Induction of alloantigen-specific human T regulatory cells by vasoactive intestinal peptide. J Immunol 183, 4346- 4359, doi:10.4049/jimmunol.0900400 (2009).

418 Fraccaroli, L. et al. VIP modulates the pro-inflammatory maternal response, inducing tolerance to trophoblast cells. British journal of pharmacology 156, 116- 126, doi:10.1111/j.1476-5381.2008.00055.x (2009).

419 Szema, A. M. et al. Mice lacking the VIP gene show airway hyperresponsiveness and airway inflammation, partially reversible by VIP. American journal of physiology. Lung cellular and molecular physiology 291, L880-886, doi:10.1152/ajplung.00499.2005 (2006).

420 Hamidi, S. A. et al. Clues to VIP function from knockout mice. Annals of the New York Academy of Sciences 1070, 5-9, doi:10.1196/annals.1317.035 (2006).

421 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).

287

422 Delgado, M. & Ganea, D. Vasoactive intestinal peptide: a neuropeptide with pleiotropic immune functions. Amino Acids 45, 25-39, doi:10.1007/s00726-011- 1184-8 (2013).

423 Iwasaki, M., Akiba, Y. & Kaunitz, J. D. Recent advances in vasoactive intestinal peptide physiology and pathophysiology: focus on the gastrointestinal system. F1000Res 8, doi:10.12688/f1000research.18039.1 (2019).

424 Yadav, M., Rosenbaum, J. & Goetzl, E. J. Cutting edge: vasoactive intestinal peptide (VIP) induces differentiation of Th17 cells with a distinctive cytokine profile. J Immunol 180, 2772-2776, doi:10.4049/jimmunol.180.5.2772 (2008).

425 Jimeno, R. et al. Vasoactive intestinal peptide maintains the nonpathogenic profile of human th17-polarized cells. J Mol Neurosci 54, 512-525, doi:10.1007/s12031- 014-0318-3 (2014).

426 Laurence, A. et al. Interleukin-2 signaling via STAT5 constrains T helper 17 cell generation. Immunity 26, 371-381, doi:10.1016/j.immuni.2007.02.009 (2007).

427 Delgado, M., Chorny, A., Gonzalez-Rey, E. & Ganea, D. Vasoactive intestinal peptide generates CD4+CD25+ regulatory T cells in vivo. J Leukoc Biol 78, 1327- 1338, doi:10.1189/jlb.0605299 (2005).

428 Bollyky, P. L. et al. CD44 costimulation promotes FoxP3+ regulatory T cell persistence and function via production of IL-2, IL-10, and TGF-beta. J Immunol 183, 2232-2241, doi:10.4049/jimmunol.0900191 (2009).

429 Liu, T., Soong, L., Liu, G., Konig, R. & Chopra, A. K. CD44 expression positively correlates with Foxp3 expression and suppressive function of CD4+ Treg cells. Biol Direct 4, 40, doi:10.1186/1745-6150-4-40 (2009).

430 Bellinger, D. L. et al. Vasoactive intestinal polypeptide (VIP) innervation of rat spleen, thymus, and lymph nodes. Peptides 18, 1139-1149, doi:10.1016/s0196- 9781(97)00075-2 (1997).

431 Guerrero, J. M., Prieto, J. C., Elorza, F. L., Ramirez, R. & Goberna, R. Interaction of vasoactive intestinal peptide with human blood mononuclear cells. Mol Cell Endocrinol 21, 151-160, doi:10.1016/0303-7207(81)90052-6 (1981).

288

432 Said, S. I. & Rosenberg, R. N. Vasoactive intestinal polypeptide: abundant immunoreactivity in neural cell lines and normal nervous tissue. Science 192, 907- 908, doi:10.1126/science.1273576 (1976).

433 Delgado, M., Pozo, D. & Ganea, D. The significance of vasoactive intestinal peptide in immunomodulation. Pharmacol Rev 56, 249-290, doi:10.1124/pr.56.2.7 (2004).

434 Azzam, H. S. et al. Fine tuning of TCR signaling by CD5. J Immunol 166, 5464- 5472, doi:10.4049/jimmunol.166.9.5464 (2001).

435 Azzam, H. S. et al. CD5 expression is developmentally regulated by T cell receptor (TCR) signals and TCR avidity. J Exp Med 188, 2301-2311, doi:10.1084/jem.188.12.2301 (1998).

436 Matson, C. A. et al. CD5 dynamically calibrates basal NF-kappaB signaling in T cells during thymic development and peripheral activation. Proc Natl Acad Sci U S A 117, 14342-14353, doi:10.1073/pnas.1922525117 (2020).

437 Voisinne, G., Gonzalez de Peredo, A. & Roncagalli, R. CD5, an Undercover Regulator of TCR Signaling. Front Immunol 9, 2900, doi:10.3389/fimmu.2018.02900 (2018).

438 Klein, M. & Bopp, T. Cyclic AMP Represents a Crucial Component of Treg Cell- Mediated Immune Regulation. Front Immunol 7, 315, doi:10.3389/fimmu.2016.00315 (2016).

439 Riccomi, A., Gesa, V., Sacchi, A., De Magistris, M. T. & Vendetti, S. Modulation of Phenotype and Function of Human CD4(+)CD25(+) T Regulatory Lymphocytes Mediated by cAMP-Elevating Agents. Front Immunol 7, 358, doi:10.3389/fimmu.2016.00358 (2016).

440 Rueda, C. M., Jackson, C. M. & Chougnet, C. A. Regulatory T-Cell-Mediated Suppression of Conventional T-Cells and Dendritic Cells by Different cAMP Intracellular Pathways. Front Immunol 7, 216, doi:10.3389/fimmu.2016.00216 (2016).

441 Tubo, N. J. et al. Single naive CD4+ T cells from a diverse repertoire produce different effector cell types during infection. Cell 153, 785-796, doi:10.1016/j.cell.2013.04.007 (2013).

289

442 Patin, E. et al. Natural variation in the parameters of innate immune cells is preferentially driven by genetic factors. Nat Immunol, doi:10.1038/s41590-018- 0049-7 (2018).

443 Piasecka, B. et al. Distinctive roles of age, sex, and genetics in shaping transcriptional variation of human immune responses to microbial challenges. Proceedings of the National Academy of Sciences of the United States of America, doi:10.1073/pnas.1714765115 (2017).

444 Li, X. et al. The impact of rare variation on gene expression across tissues. Nature 550, 239-243, doi:10.1038/nature24267 (2017).

445 Klutstein, M., Moss, J., Kaplan, T. & Cedar, H. Contribution of epigenetic mechanisms to variation in cancer risk among tissues. Proceedings of the National Academy of Sciences of the United States of America, doi:10.1073/pnas.1616556114 (2017).

446 Kaczorowski, K. J. et al. Continuous immunotypes describe human immune variation and predict diverse responses. Proceedings of the National Academy of Sciences of the United States of America 114, E6097-e6106, doi:10.1073/pnas.1705065114 (2017).

447 Hall, A. B., Tolonen, A. C. & Xavier, R. J. Human genetic variation and the gut microbiome in disease. Nature reviews. Genetics 18, 690-699, doi:10.1038/nrg.2017.63 (2017).

448 Brodin, P. & Davis, M. M. Human immune system variation. Nature reviews. Immunology 17, 21-29, doi:10.1038/nri.2016.125 (2017).

449 Lu, Y. et al. Systematic Analysis of Cell-to-Cell Expression Variation of T Lymphocytes in a Human Cohort Identifies Aging and Genetic Associations. Immunity 45, 1162-1175, doi:10.1016/j.immuni.2016.10.025 (2016).

450 Liston, A., Carr, E. J. & Linterman, M. A. Shaping Variation in the Human Immune System. Trends Immunol 37, 637-646, doi:10.1016/j.it.2016.08.002 (2016).

451 Li, Y. et al. A Functional Genomics Approach to Understand Variation in Cytokine Production in Humans. Cell 167, 1099-1110.e1014, doi:10.1016/j.cell.2016.10.017 (2016).

290

452 Chen, L. et al. Genetic Drivers of Epigenetic and Transcriptional Variation in Human Immune Cells. Cell 167, 1398-1414.e1324, doi:10.1016/j.cell.2016.10.026 (2016).

291