Else Kröner-Fresenius Symposia Editor: S. Pahernik Vol. 4

Innate Immunity: Resistance and Disease-Promoting Principles

Editors Gunther Hartmann Hermann Wagner

Innate Immunity: Resistance and Disease-Promoting Principles Else Kröner-Fresenius Symposia

Vol. 4

Series Editor

S. Pahernik Heidelberg Innate Immunity: Resistance and Disease- Promoting Principles

Volume Editors

Gunther Hartmann Bonn Hermann Wagner Munich

13 figures, 10 in color, 2013

Basel • Freiburg • Paris • London • New York • New Delhi • Bangkok • Beijing • Tokyo • Kuala Lumpur • Singapore • Sydney

Prof. Dr. med. Gunther Hartmann Prof. Dr. med. Hermann Wagner, PhD, Dr. h.c. Universitätsklinikum Bonn Technische Universität München Abteilung für Klinische Pharmakologie Institut für Medizinische Mikrobiologie Sigmund-Freud-Strasse 25 Immunologie und Hygiene DE–53105 Bonn (Germany) Trogerstrasse 30 DE–81675 München (Germany)

This book is sponsored by the Else Kröner-Fresenius-Stiftung.

Library of Congress Cataloging-in-Publication Data

Innate immunity : resistance and disease-promoting principles / volume editors, Gunther Hartmann, Hermann Wagner. p. ; cm. -- (Else Kröner-Fresenius symposia, ISSN 1663-0114 ; v.4) Includes bibliographical references and index. ISBN 978-3-318-02347-3 (hard cover : alk. paper) -- ISBN 978-3-318-02348-0 (electronic version) I. Hartmann, Gunther, 1966- II. Wagner, H. (Hermann), 1941- III. Series: Else Kröner-Fresenius symposia ; v. 4. 1663-0114 [DNLM: 1. Immunity, Innate--physiology--Congresses. QW 541] QR181 616.07‘9--dc23 2013008801

Bibliographic Indices. This publication is listed in bibliographic services, including Current Contents€.

Disclaimer. The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publisher and the editor(s). The appearance of advertisements in the book is not a warranty, endorsement, or approval of the products or services advertised or of their eff ectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

Drug Dosage. The authors and the publisher have exerted every eff ort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant fl ow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.

All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.

The cover figures are by courtesy of Cecilia Andersson, Eric Jones and Stephanie Wood. © Copyright 2013 by S. Karger AG, P.O. Box, CH–4009 Basel (Switzerland) www.karger.com Printed in Germany on acid-free and non-aging paper (ISO 9706) by Kraft Druck, Ettlingen ISSN 1663–0114 ISBN 978–3–318–02347–3 eISBN 978–3–318–02348–0 Contents

VII Preface Pahernik, S. (Heidelberg)

1 Introduction Wagner, H. (Munich); Hartmann, G. (Bonn)

Chapter 1: Innate Immunity and Inflammation

4 Innate Immunity, Inflammation and Colorectal Cancer Wang, K.; Grivennikov, S.; Karin, M. (La Jolla, Calif.)

11 Immunobiology of C-Type Lectin Receptors Roth, S.; Thomas, C. (Munich); Ruland, J. (Munich/Neuherberg)

15 Mechanisms of IL-1␤ Maturation in Neutrophils Mankan, A.K.; Hornung, V. (Bonn)

Chapter 2: Innate Immunity and Transplantation

24 The Innate Immune System: Its Rediscovery before Toll Was Described Land, W.G.; Messmer, K. (Munich)

29 Innate-Adaptive Immune Responses in Organ Ischemia/Reperfusion Injury Ji, H.; Zhai, Y.; Kupiec-Weglinski, J.W. (Los Angeles, Calif.)

35 Tolerogenic Dendritic Cells in Transplantation: From Preclinical to Clinical Application Moreau, A.; Cuturi, M.-C. (Nantes)

V Chapter 3: Innate Immunity and Intestinal Microbiota

43 Microbiota’s Influence on Immunity Honda, K.; Tanoue, T.; Nagano, Y.; Atarashi, K. (Tokyo)

48 Inflammasomes and Mucosal Immune Response Elinav, E.; Henao-Mejia, J. (New Haven, Conn.); Flavell, R.A. (New Haven, Conn./Chevy Chase, Md.)

53 Microbial Endocrinology: An Evolution-Based Shared Mechanism Determining Microbiota’s Influence on Health and Disease Lyte, M. (Abilene, Tex.)

Chapter 4: Innate Immunity and Disease Promotion

59 Standing Guard: Innate Lymphoid Cells Promote Immunity and Tissue Homeostasis at Barrier Surfaces Monticelli, L.A.; Artis, D. (Philadelphia, Pa.)

73 miRNAs That Shape the Innate Immune System: Regulation through Toll-Like Receptor Signaling Curtis, A.M.; O’Neill, L.A.J. (Dublin)

80 Type 2 Diabetes and Islet Inflammation Donath, M.Y. (Basel)

86 The Innate Immune System in Alzheimer’s Disease Beckert, H.; Halle, A. (Bonn)

91 Role of Inflammasomes in Obesity Dixit, V.D. (Baton Rouge, La.)

96 Gut-Brain Communication in the Regulation of System Metabolism Müller, T.D.; Pfluger, P.T.; Tschöp, M.H. (Munich/Cincinnati, Ohio)

Chapter 5: Drosophila and Immunity

103 Metabolism and Innate Immunity: FOXO Regulation of Antimicrobial Peptides in Drosophila Loch, G.; Jentgens, E.; Bülow, M.; Zinke, I. (Bonn); Mori, T.; Suzuki, S.; Takeyama, H. (Tokyo); Hoch, M. (Bonn)

Chapter 6: Innate Signaling and Adaptive Immunity

112 Dendritic Cells Orchestrate Innate Immunity against Bacterial Kidney Infection Tittel, A.P.; Heuser, C.; Garbi, N.; Kurts, C. (Bonn)

Chapter 7: Speakers at the Symposium

120 Speakers at the Symposium

127 Author Index 128 Subject Index

VI

Preface

Innate Immunity tissue injury-induced molecules denoted as dam- age-associated molecular patterns (DAMPs). This book series features the proceedings of the The recognition of either PAMPs or DAMPs Else Kröner-Fresenius Symposia, which are in- then initiates an infectious or sterile inflamma- tended to cover clinically relevant topics at the tory response, followed by tissue repair if neces- forefront of biomedical research. The meetings sary. should give experts the opportunity to discuss In this sense, innate immunity has definitely the most recent findings in evolving fields of bio- developed beyond the boundaries of classical im- medicine and outline future research strategies. munology and can now be regarded as an im- Today’s research is characterized by the accel- mune sensory system evolutionarily determined erated generation of biomedical data, the increas- to directly or indirectly sense alterations in cell or ingly interdisciplinary and translational nature tissue integrity. In fact, as recognized today, the of biomedical science, as well as efforts to inte- innate immune system encompasses a much grate the data into complex biological systems. broader field of life-saving biological functions: These developments emphasize the need for new when adequately controlled, it is essential for forums of discussion. maintaining homeostasis and, thus, guarantees The innate immune system is an evolutionari- the health of an individual. However – and this is ly highly conserved, first rapid line of host de- the other side of the coin – when functionally un- fense that precedes and instructs the adaptive im- controlled and exaggerated, the same defense mune system. The defense system detects not system plays a pivotal deleterious role in most hu- only pathogen-mediated injury but also any oth- man pathologies and diseases including sepsis, er type of physical, chemical or radiation tissue atherosclerosis, metabolic disorders, and neuro- insult. Its general strategy of defense relies on dis- degenerative diseases. tinct innate immune-sensing receptors that are In this context, Prof. Hermann Wagner orga- present not only in immune cells but in most so- nized in May 2012 a distinguished meeting to matic cells. These receptors are able to recognize discuss the current knowledge of the innate im- both microbial molecules known as pathogen as- mune system, the 4th Else Kröner-Fresenius sociated molecular patterns (PAMPs) and sterile Symposium. World-renowned experts in the field of the innate immune system discussed per- company was progressively rebuilt and the need spectives of the current advances integrating the to maintain it determined all activities. data from diverse fields of research into the med- It were these important and far-sighted entre- ical perspective. preneurial business decisions in the 1950s and The Else Kröner-Fresenius Stiftung thanks 1960s that ensured the successful future develop- Prof. Hermann Wagner for his inspiring scien- ment of the company. Decades of growth fol- tific work and personal outstanding input in or- lowed, in particular within the field of dialysis, ganizing together with his team, the 4th Else nutrition and intensive care, leading to an inter- Kröner-Fresenius Symposium. nationally competitive enterprise and market leader in special areas of health care. Until 1981, Else Kröner led the company. After The Else Kröner-Fresenius-Stiftung the transformation of Fresenius into a stock com- pany, she remained chairwoman of the Supervi- In 1983, Else Kröner (1925–1988) founded the sory Board until her death on June 5, 1988. From Else Kröner-Fresenius-Stiftung, a private foun- 1981 to 1992, her husband Hans Kröner led the dation and nonprofit organization dedicated to company as CEO. Thereafter, he significantly promoting medical science, supporting medical shaped the policy of the Else Kröner-Fresenius- education, and providing humanitarian aid. Stiftung, of which he was chairman of the board Else Kröner, née Fernau, was born on May 15, from 1995 to 2005. 1925, in Frankfurt am Main, Germany. When Today, the Fresenius group, of which the Else she was 3 years old, her father died. After his Kröner-Fresenius-Stiftung is the leading share death, she lived with her mother in the home of holder, is an international healthcare conglomer- Dr. Eduard Fresenius, a pharmacist and owner of ate with products and services for dialysis, hospi- the Hirsch Pharmacy in Frankfurt, who had tal, and medical care of patients. The Fresenius founded the pharmaceutical company Fresenius group currently employs nearly 150,000 people in in 1912. Dr. Fresenius, whose marriage remained more than 100 countries and generates annual childless, took care of Else Fernau. In 1944, she sales of over EUR 16 billion. started an internship at the Hirsch Pharmacy and Else Kröner entrusted nearly her entire prop- decided to study pharmacy, which was supported erty to the foundation. She laid down that its fi- by her patron. In 1946, Dr. Fresenius unexpect- nancial resources should be employed to promote edly died. At that time, Else Fernau had not com- medical science, advance health care and provide pleted her pharmaceutical education. However, humanitarian aid. It is in accordance with her vi- Dr. Eduard Fresenius bequeathed the Hirsch sion and requirements that the Else Kröner-Fre- Pharmacy and the Fresenius company to her. senius-Stiftung continues to put the founder’s At the age of 21 years, Else Fernau decided, fortune at the service of nonprofit projects and against the advice of many, to take responsibility objectives. The symposia are published as part of for the Hirsch Pharmacy and the Fresenius com- the foundation’s commitment to the advance- pany, which were then experiencing severe finan- ment of medical research and treatment. cial difficulties during the post-war years. Of the Sascha Pahernik, Heidelberg original 400 employees, all but 30 had to be laid Member of the Scientific Committee of off. In her efforts to ensure the survival and force the Else Kröner-Fresenius-Stiftung the re-expansion of the company, she was later Series Editor supported by her husband, Hans Kröner. The

VIII Preface

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 1–3 (DOI: 10.1159/000346498)

Introduction

a b H. Wagner G. Hartmann a b Institute for Medical Microbiology, Immunology and Hygiene, Technical University Munich, Munich, and Department for Clinical Pharmacology, University Clinics Bonn, Bonn, Germany

Innate Immunity’s Resurrection: workers had shown that in fruit fly embryos the How It Perceives Infection Toll gene controls the establishment of the dorso- ventral axis. Using Toll mutants originally gener- In 1908, E. Metchnikoff and P. Ehrlich jointly re- ated for embryological studies, Jules Hoffmann ceived the Nobel Prize ‘in recognition of their and Bruno Lemaitre then reported in 1996 that a work in immunity’. Metchnikoff advocated the functioning Toll gene was essential to control idea that phagocytes constituted a first line of in- fungal infections in adult flies. The fact that the nate defense by nonspecifically ingesting and di- innate immune system of flies relies upon germ gesting invading pathogens. In contrast, Ehrlich line-encoded and ligand-specific receptors to proposed the ‘side chain theory’ to explain how sense infection was a revelation to many immu- antibodies specific for diphtheria and tetanus nologists. Later, the observation that inbred exotoxins functioned. For decades, innate immu- mouse strains C3H/HeJ and C57BL/10ScCr re- nity was thus considered as ‘non-specific’, while sisted otherwise lethal doses of lipopolysaccha- the hallmark ‘specificity’ was confined to adap- ride (LPS; endotoxin) prompted the speculation tive immunity – as mediated by T and B lympho- as to whether these inbred mice harbor a non- cytes. Although microbes had long been recog- functional (mutated) receptor sensing LPS. Con- nized as the cause of infectious diseases, and sequently, Bruce Beutler and colleagues used Charles Janeway in 1989 had speculated that LPS-resistant C3H/HeJ mice and searched via pathogen-associated molecular patterns alerted ‘positional cloning’ for the postulated LPS recep- adaptive immunity by upregulating costimula- tor. In 1998, they discovered that LPS is sensed by tory signals on antigen-presenting dendritic cells, Toll-like receptor 4 (TLR4); enforced cross-link- the fundamental question as to how innate im- ing of TLR4 had previously been shown by mune cells perceive infections remained largely C. Janeway and R. Medzhitow to cause NF-␬ B- unknown. One guiding discovery and two semi- dependent cytokine production. Beutler’s mile- nal discoveries subsequently gave ‘limited’ speci- stone discovery was the first to link the TLR sys- ficity to innate immune cells. In 1985, Nüsslein tem with recognition of structurally defined Vollhard (Nobel Prize Laureate in 1995) and co- molecules of utmost biological relevance. By generating TLR gene knockout mice, containing receptors), the function of NLRP3 has Shizou Akira and his group made important con- been highlighted by the work of J. Tschopp. tributions to the identification of TLR ligands NALP3 was found to trigger ‘inflammasome’ and TLR signaling pathways that induce proin- formation upon ligand-driven oligomerization of flammatory cytokines or type 1 interferons. Al- cytosolic NLRP3 and ASC proteins. NLRP6 has together, the pioneering work of Akira, Beutler, recently been identified as a component of an in- Hoffmann and Medzhitow brought about a shift flammasome that activates IL-18 and negatively in our understanding how the host perceives in- regulates colonic inflammation through altera- fection: Innate immune cells and many other cell tions of the intestinal microbiota. ASC-depen- types express evolutionary conserved germ line- dent formation of inflammasomes is also a func- encoded pattern recognition receptors (PRRs) tion of AIM2: the inflammasome then causes via able to sense pathogen-derived ligands. Upon caspase 11 and 1 the production/secretion of bio- recognition, such ligands specifically activate in- logically active IL-1 family members. nate immune cells and function as powerful ad- juvants to alert adaptive immunity. The Nobel Prize in Physiology/Medicine to J. Hoffmann, Innate Immunity’s Vibrancy: How Does It B. Beutler and R. Steinmann (for his pioneering Promote Diseases? work on dendritic cells) highlighted this para- digm shift in our understanding of innate im- In recent years, a second paradigm shift (the first munity. concerns the germ line-encoded limited reper- TLRs were the first PRRs (or immune sensing toire of innate immune cells) has appeared on the receptors) to be described. Numerous additional horizon. Innate PRRs appear to be ‘promiscuous’ immune sensing receptors have now been de- in that they recognize not only exogenous, patho- scribed. TLRs and C-type lectin are membrane gen-derived ligands but also endogenous, host- bound and either located on the cell surface or in derived molecules. In her ‘danger theory’, P. the endosomal membrane. More immune sens- Massinger has collectively termed such endoge- ing receptors are found in the cytoplasm. For ex- nous ‘danger’ signals ‘danger-associated molecu- ample, retinoic acid-inducible gene I-like recep- lar patterns’ (DAMPs). DAMPS sensed by PRRs tors (RIG-I like receptors including RIG-I and include liberated intracellular components (heat melanoma differentiation-associated protein 5) shock proteins, high-mobility box proteins, and are members of the DExD/H box helicase super- extracellular host DNA), cleaved matrix hyaluro- family. They function as cytosolic RNA sensors nan proteins, misfolded proteins including amy- alerting innate immunity towards virus- and loid- ␤, or CEPs (carboxyethyl pyrrole), the end bacteria-derived RNA. Absent in melanoma 2 products of lipid oxidation that are present in (AIM2)-like receptors represent a group of DNA- low-density lipoproteins. These discoveries add- sensing receptors that comprises two members of ed a second dimension: the degree to which in- the Pyrin and HIN domain-containing protein nate immune responses cause or promote chron- family: AIM2 and interferon-␥ -inducible protein ic autoinflammatory diseases. One striking ex- 16. While STING (stimulator of IFN genes) is ample is the autoinflammatory disease gout: uric mostly known as adaptor molecule, recent work acid crystals activate the NALP3-driven inflam- highlighted its ability to bind bacterial DNA as masome and IL-1␤ drives acute inflammation. well as cyclic di-GMP, a signaling molecule re- Autoinflammatory responses have also been stricted to bacteria. Within the group of NLRs linked to atherosclerosis, certain aspects of the (nucleotide-binding domain leucine-rich repeat- metabolic syndrome, as well as to type 2 diabetes.

2 Wagner Hartmann

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 1–3 (DOI: 10.1159/000346498) Innate Immunity and the Gut: How It Impacts type 2 diabetes and atherosclerosis, and certain Gut Microbiota Homeostasis metabolic disorders for the large part is still puz- zling and subject to intense investigation. Being Defensins represent major determinants of gut engaged in unraveling the immunobiology of homeostasis with its microbiota. While TLRs are PRRs, we sensed that there is an urgent need for unlikely to discriminate between commensals an information platform to discuss where this and pathogens, and NLRP6 expressed in gut epi- field of science stands now, and where it is likely thelial cells control defensin production, IL-22 to develop. The best platform envisaged was to produced by gut-homing ‘innate lymphoid cells’ organize a symposium in a remote place, in which keeps commensal bacteria contained in their an- an invited international Faculty of Scientists was atomical niches. Furthermore, the composition to discuss their views on the state of the art in this of gut microbiota appears to impact on the devel- field. The Else Kröner-Fresenius-Stiftung gener- opment of inflammatory Th17 (T) cells and that ously funded this symposium that focused on the of regulatory T cells. If so, T cell functions appear role of innate immunity ‘in protection against in- to be imprinted not only in the thymus but also fection’ and ‘in promoting chronic autoinflam- in the gut. matory diseases’. The symposium, which took place in May 2012 at Schloss Elmau/Upper Ba- varia, brought together leading experts, fostered Concluding Remarks scientific exchange, open and unsparing discus- sion, as well as future concepts. This meeting is Innate immunity’s impact on protection against the first of a series of biannual meetings which we viral or bacterial pathogens is increasingly being plan in the context of the new DFG-funded Ex- understood on a molecular level. To combat in- cellence Cluster ImmunoSensation: the Immune truders and driven by a limited repertoire of Sensory System with a scientific focus on these germ line-encoded PRRs, innate immune cells newly developing fields of immune sensing with respond to infection with the balanced and well- connection to the metabolic, endocrine and ner- controlled production of proinflammatory cyto- vous systems. This book represents a meeting re- kines and type I interferon. The impact of innate port summarizing the current knowledge in this immunity on autoinflammatory diseases, age- vibrant field of research at the starting point of related chronic inflammatory disorders such as this new excellence cluster.

Prof. Dr. Hermann Wagner Technical University Munich Institute for Medical Microbiology, Immunology and Hygiene Trogerstrasse 30 DE–81675 Munich (Germany)

E-Mail hermann.wagner@ mikrobio.med.tum.de

Introduction 3

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 1–3 (DOI: 10.1159/000346498) Chapter 1: Innate Immunity and Inflammation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526)

Innate Immunity, Inflammation and Colorectal Cancer

Kepeng Wang Sergei Grivennikov Michael Karin

Laboratory of Gene Regulation and Signal Transduction, Department of Pharmacology and Pathology, School of Medicine, University of California, San Diego, La Jolla, Calif., USA

A b s t r a c t A link between inflammation and cancer has Up to 20% of cancers arise from chronic inflammation been suspected for a long time since Rudolf Vir- and persistent infections. Moreover, most solid tumors chow observed infiltration of leukocytes in ma- contain immune infiltrates. Tumor-associated inflamma- lignant tumors [1] . Yet, experimental evidence tory cells play broad roles in different stages of cancer demonstrating the important role of inflamma- development, including initiation, promotion, progres- tion in tumorigenesis has become available only sion and metastasis. Cytokines are important mediators in the last decade [2] . About 15–20% of cancers of the inflammatory effect on tumorigenesis. Our work arise in the context of preceding chronic inflam- showed that IL-6 is an important tumor promoter in ear- mation. For instance, Helicobacter pylori infec- ly colitis-associated cancer (CAC). IL-6 is mainly produced tion is associated with gastric cancer, while HBV by tumor-infiltrating myeloid cells under the control of and HCV infections and hepatitis are associated NF-␬ B. IL-6 promotes proliferation of tumor-initiating with hepatocellular carcinoma. Chronic colonic cells derived from the intestinal epithelium and protects inflammation manifested in ulcerative colitis them from apoptosis during colitis. The pro-survival and other inflammatory bowel diseases (IBD) and proliferative effects of IL-6 are mainly mediated by significantly increases the probability of colorec- STAT3, whose ablation in intestinal epithelial cells signif- tal cancer (CRC) development [3–6]. While it is icantly reduces CAC tumorigenesis and augments the already known that chronic inflammation in- host susceptibility to experimental colitis. Taken togeth- duces and promotes cancer and that the usage of er, the NF- ␬ B-IL-6-STAT3 pathway links innate immunity non-steroidal anti-inflammatory drugs, such as to tumorigenesis by promoting the survival and prolif- aspirin, decreases the life-long risk of cancer eration of tumor-initiating cells. IL-6 and other cytokines death [7, 8] , less is known about molecular and may also enhance cancer progression. cellular mechanisms connecting inflammation Copyright © 2013 S. Karger AG, Basel and cancer. Recent studies have begun to deci- pher such links, focusing on signaling within and interactions between myeloid, immune and epi- thelial cells [6, 9–12]. Inflammatory cytokines O v e r a l l , T N F - I K K ␤ -NF- ␬B signaling acti- are the key mediators and regulators of these in- vates myeloid cells to produce proinflammatory teractions. cytokines, which in turn serve as growth factors CRC is the third most common and deadly that increase malignant cell proliferation. NF-␬ B cancer around the globe [13] . Patients with ulcer- signaling in epithelial cells mainly prevents pre- ative colitis have significantly higher risk of CRC malignant cell from undergoing apoptosis; there- as approximately 20% of IBD patients develop fore, in its absence, we observe fewer and smaller colitis-associated cancer (CAC) within 30 years tumors being formed upon AOM treatment com- of disease emergence [14, 15] . CAC is a classical bined with chronic colitis [18] . inflammation-driven cancer, which can be rela- I nactivation of NF-␬ B in myeloid cells led to tively easily modeled in mice by three cycles of reduced production of IL-6 during DSS-evoked dextran sodium sulfate (DSS) in the drinking wa- intestinal inflammation [21] . IL-6 is a multifunc- ter subsequent to a single dose of the pro-carcin- tional cytokine that plays important roles in im- ogen azoxymethane (AOM) [16, 17] . mune responses, cell survival and proliferation It has been previously shown by our lab that [22] . IL-6 binds to soluble or membrane-bound an overall reduction of the DSS-evoked intesti- IL-6R ␣, and cell surface gp130, and activates in- nal inflammatory response by inactivation of tracellular signaling mediated by STAT3, Ras and IKK ␤ in myeloid cells leads to decreased tumor PI3K-Akt [22] . In immunity, IL-6 is important size, with concomitant reduction in the expres- for T cell survival and differentiation, and there- sion of multiple proinflammatory cytokines that fore plays a pivotal role in the pathogenesis of au- may serve as tumor growth factors [18]. Impor- toimmune disorders [23]. Blocking IL-6 signal- tantly, inactivation of NF-␬ B signaling in intes- ing by neutralizing antibody against IL-6R or by tinal epithelial cells (IEC) by cell type-specific gp130-Fc fusion protein caused suppression of disruption of the Ikk ␤ gene reduced tumor count colitis in a mouse model of Crohn’s disease [24] . and increased apoptosis in tumors and injured IL-6 also plays an important role in tissue ho- tissue [18] . That study first suggested that cyto- meostasis and regeneration, suggesting that it kines produced by inflammatory cells in an NF- may directly promote tumorigenesis and malig- ␬ B-dependent manner can act on premalignant nant cell survival [25, 26] . Moreover, IL-6 mRNA cells derived from IEC to activate the NF-␬ B- is upregulated in multiple human cancers, in- dependent pro-survival gene expression pro- cluding breast, lung, prostate, liver and colon gram. One of the cytokines which is induced by cancer [27] , and its expression levels often corre- NF-␬ B and can activate NF-␬ B is tumor necro- late with tumor mass and poor prognosis. IL-6 sis factor (TNF), whose role in CAC has been promotes proliferation of human colon carcino- demonstrated. Mice lacking TNFR1 showed re- ma cells in vitro via activation of STAT3 [28] . In duced mucosal damage, reduced infiltration of the settings of CAC, IL-6 may perpetuate chron- macrophages and neutrophils, and attenuated ic inflammation and maintain production of formation of colon tumors [19] . Likely, TNF ex- proinflammatory cytokines responsible for erts its tumorigenic properties by acting both growth and survival of malignant cells. Blocking on immune and epithelial cells. While adoptive IL-6 signaling by injection of chimeric gp130-Fc transfer experiments showed that TNFR1 sig- protein which blocks IL-6 transsignaling result- naling is particularly important in the radiosen- ed in reduced CAC tumor burden in mice [29] . sitive compartment [19] , other studies demon- These lines of evidence suggest that IL-6 may strated the potential importance of TNFR2 sig- play an oncogenic role in CAC by acting on both naling in IEC [20] . immune cells and malignant epithelial cells.

Innate Immunity, Inflammation and Colorectal Cancer 5

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) By using genetic tools to disrupt IL-6 signal- or Il6 –/– mice subjected to acute colitis. DSS-ex- ing in vivo, we demonstrated that IL-6 regulates posed Il6 –/– mice exhibited elevated apoptosis in survival and proliferation of IEC and preneoplas- IECs. Consistently, we observed downregulation –/– tic cells during induction of CAC and that the ef- of antiapoptotic protein Bcl-X L in IECs of Il6 fects of IL-6 are in part mediated by its ability to mice undergoing acute colitis. We also observed activate transcription factor STAT3 in premalig- a significant decrease in the proliferation of basal nant cells [30] . These data were further corrobo- crypt cells of Il6 –/– mice subjected to intestinal rated by genetic inactivation of STAT3 in IECs, injury caused by DSS exposure. Therefore, IL-6 which inhibits CAC induction and growth [30, promotes both proliferation and survival of IECs 31] . Therefore, cytokines produced at high levels during acute colitis. Presumably, IL-6 exerts the during chronic inflammatory responses can also same effect on malignant IECs; it protects them act on premalignant and cancerous cells to en- from apoptosis, and promotes their proliferation hance tumor promotion and progression. in CAC. To determine the origin of IL-6 in CAC, we performed reciprocal adoptive transfer experi- R e s u l t s ments by introducing WT or Il6–/– bone marrow into WT or Il6 –/– lethally irradiated recipient IL-6 Is Required for CAC Tumorigenesis mice. Reduction in tumor number, size and load It is generally believed that activation of NF- ␬ B in mice deficient in IL-6 in hematopoietic cells in myeloid cells promotes neoplastic growth of suggested the importance of immune cells in CAC in mice in part by induction of proinflam- overall IL-6 production during CAC tumorigen- matory cytokines and growth factors [2, 21] . IL-6 esis. To further delineate the source of IL-6 in fulfills the criteria of being an NF-␬ B target and CAC, we purified different myeloid and immune an important regulator of inflammation on one cells from CAC adenomas by FACS sorting. hand and a potent growth factor for epithelial Analysis of IL-6 mRNA by q-PCR revealed that and malignant cells on the other. Since levels of IL-6 is mainly produced by dendritic cells and IL-6 are typically upregulated in many cancers macrophages, followed by T cells. In patients suf- and chronic inflammatory conditions, we sought fering from ulcerative colitis and CAC, IL-6 is ex- to examine the effect of complete IL-6 deficiency pressed in colonic epithelial cells and more po- on CAC development. CAC was induced in wild- tently by infiltrating immune cells. Expression of type (WT) and Il6 –/– mice by injection of a single IL-6 correlates with downregulation of SOCS3 dose of AOM followed by 3 cycles of DSS in the and activation of STAT3 in both epithelial/tumor drinking water [16, 17]. As expected, colonic IL-6 cells and immune cells [32] . was upregulated upon DSS treatment. Ablation Cancer cells from IL-6-deficient mice showed of IL-6 resulted in reduced tumor number, size reduced number of proliferating PCNA+ cells and total tumor load in mice. These data indicate and decreased levels of cyclin D expression. We that IL-6 is important for both tumor develop- also tested the effect of ectopic IL-6 administra- ment and growth in CAC. Differences in tumor tion on CAC tumorigenesis. WT mice subjected multiplicity and size may be explained by altered to the CAC induction protocol were injected with cancer cell apoptosis and/or proliferation. Re- recombinant IL-6 or ‘hyper-IL-6’, which is a fu- duced tumor number in Il6 –/– mice suggested sion protein of IL-6 and soluble IL-6R that trig- that IL-6 may contribute to cancer cell survival gers IL-6 transsignaling [33, 34] . When adminis- and/or proliferation. To test this hypothesis, we tered at the late stage after the last cycle of DSS examined apoptosis and cell proliferation in WT treatment, IL-6 and hyper-IL-6 contributed to a

6 Wang Grivennikov Karin

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) significant increase in tumor size but not tumor esis in a mouse model of inflammation-associat- multiplicity. When administered during early ed skin cancer [45] . We tested the role of IL-23 stage of CAC induction, they increased tumor and its downstream Th17 effector cytokines in number. Therefore, IL-6 promotes both tumor CAC development. Expression of IL-23 is upreg- formation and growth. ulated in intestinal tissues subjected to DSS- induced colitis. Compared with WT controls, STAT3 in Epithelial Cells Mediates the Il23 –/– (p19 –/– ) mice develop fewer and smaller co- Tumorigenic Effect of IL-6 lonic adenomas when subjected to CAC induc- Among downstream signaling pathways activat- tion. Expression of IL-6, IL-17A and IL-22 in ed by IL-6, we observed a marked reduction in CAC are also reduced in the absence of IL-23. Ab- the level of active STAT3 in IEC of Il6 –/– mice un- lation of the Il17r or Il22 genes also resulted in dergoing DSS-induced colitis. It is plausible that reduced CAC tumorigenesis, suggesting an im- IL-6 acts through STAT3 to promote the survival portant role for Th17-derived cytokines in CAC and proliferation of IEC during colitis, and stim- promotion. IL-23 promotes CAC development ulate growth of premalignant cells in the course mainly through activation of IEC-expressed of CAC induction. To this end, we crossed Stat3 F/F STAT3 to enhance the proliferation and survival mice [35] with villin-cre mice [36] to generate of transformed enterocytes. However, unlike Stat3 ⌬IEC mice where Stat3 is selectively ablated IL-6 receptor, IL-23 receptor is not expressed in in IECs. The phenotype of Stat3 ⌬IEC mice is sim- IECs or adenoma cells. Therefore, IL-23 pro- ilar to that of Il6 –/– mice, although Stat3 ⌬IEC mice motes CAC indirectly by effects on tumor-infil- developed much more severe colitis when ex- trating immune cells, particularly Th17 cells, and posed to DSS due to enhanced IEC apoptosis. upregulation of other cytokines like IL-6, IL-17A Consistent with this, expression of Bcl-XL was re- and IL-22 that are capable of signaling more di- duced in IEC lysates of Stat3 ⌬IEC mice compared rectly to malignant cells to promote their prolif- to WT controls. Most importantly, fewer adeno- eration. mas were found in Stat3 ⌬IEC mice subjected to CAC induction, an effect that was greater than that of the Il6 knockout. These data indicated Discussion that STAT3 is required for transduction of the IL-6 tumor-promoting signal, and by doing this Our studies indicate that IL-6 produced by my- STAT3 promotes the survival, growth and regen- eloid cells serves as tumor promoter by activating eration of enterocytes in colitis and CAC. STAT3 in IECs [30] . An article published at the same time by Bollrath et al. [31] confirmed the Th17 Cells Promote Tumorigenesis in CAC important role of IL-6 and IL-11/STAT3 signal- Emerging evidence suggests that Th17 cells po- ing in the development of CAC. These authors tentiate tumor-associated inflammation [37, 38] . were also able to show that increased activation of IL-23 promotes Th17 cell proliferation and IL-17 STAT3 achieved by expression of hyperactive production, and is important for inflammatory gp130 resulted in increased tumor multiplicity responses including experimental colitis [39–42] . and size [31] . Deletion of IEC-expressed STAT3 IL-23 is a heterodimeric cytokine composed of a also resulted in more severe mucosal damage unique p19 subunit and a p40 subunit, which it during DSS-induced colitis, while excessive acti- shares with IL-12 [43, 44]. Expression of IL-23 is vation of STAT3 confers resistance to DSS colitis upregulated in multiple human cancers [45] . De- [31]. Taken together, IL-6 and possibly other letion of Il23p19 resulted in reduced tumorigen- STAT3-activating cytokines promote CAC devel-

Innate Immunity, Inflammation and Colorectal Cancer 7

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) opment by impacting IEC survival and growth. ly unraveled. Importantly, in addition to its role Indeed, we found that IL-23 is another important in CAC, IL-6 also promotes spontaneous intesti- promoter of CAC development. Similar to IL-6, nal tumorigenesis, as deletion of the Il6 gene in IL-23 is produced by lamina propria myeloid cells Apc Min mice also resulted in reduced tumor load and it can lead to STAT3 activation in IEC. How- [50] . The molecular and cellular mechanisms by ever, unlike IL-6, IL-23 does not act directly on which IL-6 promotes the development of sporad- IEC, as these cells do not express IL-23 receptor. ic CRC are not yet understood, but may be similar Most likely, IL-23 controls STAT3 activation in to those discovered for CAC. IEC and their premalignant derivatives by affect- While nearly 20% of all human cancers are ing the expression of other cytokines, including associated with premalignant inflammation, re- IL-6. cruitment of inflammatory cells can be detected IL-6-deficient mice develop fewer and smaller in the majority of solid tumors [2, 4], thus rep- adenomas than WT controls. Through IEC-spe- resenting ‘tumor-elicited inflammation’. While cific deletion of Stat3 , we demonstrated that IL-6 this type of inflammation does not affect tumor acts directly on enterocytes to promote tumori- initiation, most experimental evidence suggests genesis [30] . In agreement with this note, recipro- that this ‘tumor-elicited inflammation’ has a piv- cal adoptive transfer experiments showed that otal role in tumor growth, progression and ther- hyperactive gp130 mainly functions in non-he- apy resistance even in tumors of ‘non-inflamma- matopoietic cells to promote CAC development tory’ origin [2, 4, 51, 52] . Moreover, non-steroidal [31]. Nonetheless, the role of IL-6 in CAC likely anti-inflammatory drugs are quite effective in re- extends beyond its direct action on enterocytes ducing cancer-related death and tumor progres- and may include its action on various subsets of sion in otherwise ‘non-inflammatory’ cancers [7, immune cells and fibroblasts. For example, IL-6 8] . Origins and mechanisms of induction that are is known to regulate the recruitment of myeloid responsible for ‘tumor-elicited inflammation’ are cells and neutrophils to the sites of inflammation not yet defined, and in the case of CRC may in- [46] , as well as to inactivate T regulatory cells by clude oncogene-driven chemokine production, altering FoxP3 expression [47] , consistent with hypoxia or therapy-induced necrosis [4] and acti- the role of IL-6 in Th17 cell induction [48, 49] . vation of tumor-associated immune cells by mi- IL-6 can also exert its tumorigenic effects through crobial products or stress signals present within Th17 cells, which are critical regulators of intes- the tumor microenvironment. For example, for- tinal inflammation and emerging players in can- mation of colonic tumors may disrupt intestinal cer [37, 38]. Indeed, in the CAC model, we ob- barrier and allow inflammation induction by served a reduction in IL-17-producing Th17 cells otherwise non-pathogenic bacteria, similar to the in CAC tumors of Il6 –/– mice [30]. Th17 cells may disruption of barrier function that precedes IBD in turn promote tumor development by secreting development [53] . It is therefore important to inflammatory cytokines like IL-17A and IL-22. study the cause and the mechanisms of tumor- The relative contribution of various cellular tar- elicited inflammation as it holds the key for new gets for IL-6 and the molecular pathways under- therapeutic, prognostic and preventive approach- lying its pro-tumorigenic action remain to be ful- es in a variety of cancers.

8 Wang Grivennikov Karin

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) References

1 Balkwill F, Mantovani A: Inflamma- 15 Lakatos PL, Lakatos L: Risk for colorec- 24 Atreya R, Mudter J, Finotto S, Mullberg tion and cancer: back to Virchow? tal cancer in ulcerative colitis: changes, J, Jostock T, Wirtz S, Schutz M, Bartsch

Lancet 2001; 357: 539–545. causes and management strategies. B, Holtmann M, Becker C, Strand D,

2 Grivennikov SI, Greten FR, Karin M: World J Gastroenterol 2008; 14: 3937– Czaja J, Schlaak JF, Lehr HA, Autsch- Immunity, inflammation, and cancer. 3947. bach F, Schurmann G, Nishimoto N,

Cell 2010; 140: 883–899. 16 Neufert C, Becker C, Neurath MF: An Yoshizaki K, Ito H, Kishimoto T, Galle 3 Karin M, Lawrence T, Nizet V: Innate inducible mouse model of colon carci- PR, Rose-John S, Neurath MF: Block- immunity gone awry: linking micro- nogenesis for the analysis of sporadic ade of interleukin 6 trans signaling bial infections to chronic inflammation and inflammation-driven tumor pro- suppresses T-cell resistance against

and cancer. Cell 2006; 124: 823–835. gression. Nat Protoc 2007; 2: 1998–2004. apoptosis in chronic intestinal inflam- 4 Mantovani A, Allavena P, Sica A, Balk- 17 Okayasu I, Ohkusa T, Kajiura K, Kanno mation: evidence in Crohn disease and will F: Cancer-related inflammation. J, Sakamoto S: Promotion of colorectal experimental colitis in vivo. Nat Med

Nature 2008; 454: 436–444. neoplasia in experimental murine ul- 2000; 6: 583–588.

5 Waldner MJ, Neurath MF: Cytokines in cerative colitis. Gut 1996; 39: 87–92. 25 Dann SM, Spehlmann ME, Hammond colitis associated cancer: potential drug 18 Greten FR, Eckmann L, Greten TF, DC, Iimura M, Hase K, Choi LJ, Hanson targets? Inflamm Allergy Drug Targets Park JM, Li ZW, Egan LJ, Kagnoff MF, E, Eckmann L: IL-6-dependent muco-

2008; 7: 187–194. Karin M: IKKbeta links inflammation sal protection prevents establishment 6 Terzic J, Grivennikov S, Karin E, Karin and tumorigenesis in a mouse model of of a microbial niche for attaching/effac-

M: Inflammation and colon cancer. colitis-associated cancer. Cell 2004; 118: ing lesion-forming enteric bacterial

Gastroenterology 2010; 138: 2101–2114. 285–296. pathogens. J Immunol 2008; 180: 6816– e5. 19 Popivanova BK, Kitamura K, Wu Y, 6826. 7 Gierach GL, Lacey JV Jr, Schatzkin A, Kondo T, Kagaya T, Kaneko S, Oshima 26 Tebbutt NC, Giraud AS, Inglese M, Jen- Leitzmann MF, Richesson D, Hollen- M, Fujii C, Mukaida N: Blocking TNF- kins B, Waring P, Clay FJ, Malki S, Al- beck AR, Brinton LA: Nonsteroidal alpha in mice reduces colorectal carci- derman BM, Grail D, Hollande F, Heath anti-inflammatory drugs and breast nogenesis associated with chronic coli- JK, Ernst M: Reciprocal regulation of

cancer risk in the National Institutes of tis. J Clin Invest 2008; 118: 560–570. gastrointestinal homeostasis by SHP2 Health-AARP Diet and Health Study. 20 Onizawa M, Nagaishi T, Kanai T, Na- and STAT-mediated trefoil gene activa-

Breast Cancer Res 2008; 10:R38. gano K, Oshima S, Nemoto Y, Yoshioka tion in gp130 mutant mice. Nat Med

8 Liu X, Plummer SJ, Nock NL, Casey G, A, Totsuka T, Okamoto R, Nakamura T, 2002; 8: 1089–1097. Witte JS: Nonsteroidal antiinflamma- Sakamoto N, Tsuchiya K, Aoki K, Ohya 27 Heikkila K, Ebrahim S, Lawlor DA: Sys- tory drugs and decreased risk of ad- K, Yagita H, Watanabe M: Signaling tematic review of the association be- vanced prostate cancer: modification pathway via TNF-alpha/NF-kappaB in tween circulating interleukin-6 (IL-6)

by lymphotoxin alpha. Am J Epidemiol intestinal epithelial cells may be direct- and cancer. Eur J Cancer 2008; 44: 937–

2006; 164: 984–989. ly involved in colitis-associated carci- 945. 9 Grivennikov SI, Karin M: Dangerous nogenesis. Am J Physiol Gastrointest 28 Becker C, Fantini MC, Wirtz S, Niko-

liaisons: STAT3 and NF-kappaB col- Liver Physiol 2009; 296:G850–G859. laev A, Lehr HA, Galle PR, Rose-John S, laboration and crosstalk in cancer. Cy- 21 Greten FR, Eckmann L, Greten TF, Neurath MF: IL-6 signaling promotes

tokine Growth Factor Rev 2010; 21: 11– Park JM, Li ZW, Egan LJ, Kagnoff MF, tumor growth in colorectal cancer. Cell ␤ 19. Karin M: IKK links inflammation Cycle 2005; 4: 217–220. 10 Hanahan D, Weinberg RA: Hallmarks and tumorigenesis in a mouse model of 29 Becker C, Fantini MC, Schramm C,

of cancer: the next generation. Cell colitis-associated cancer. Cell 2004; 118: Lehr HA, Wirtz S, Nikolaev A, Burg J,

2011; 144: 646–674. 285–296. Strand S, Kiesslich R, Huber S, Ito H, 11 Saleh M, Trinchieri G: Innate immune 22 Kishimoto T: Interleukin-6: from basic Nishimoto N, Yoshizaki K, Kishimoto mechanisms of colitis and colitis-asso- science to medicine – 40 years in im- T, Galle PR, Blessing M, Rose-John S,

ciated colorectal cancer. Nat Rev Im- munology. Annu Rev Immunol 2005; Neurath MF: TGF-beta suppresses tu-

munol 2011; 11: 9–20. 23: 1–21. mor progression in colon cancer by 12 Shiao SL, Ganesan AP, Rugo HS, Cous- 23 Neurath MF, Finotto S: IL-6 signaling inhibition of IL-6 trans-signaling. Im-

sens LM: Immune microenvironments in autoimmunity, chronic inflamma- munity 2004; 21: 491–501. in solid tumors: new targets for thera- tion and inflammation-associated can- 30 Grivennikov S, Karin E, Terzic J,

py. Genes Dev 2011; 25: 2559–2572. cer. Cytokine Growth Factor Rev 2011; Mucida D, Yu GY, Vallabhapurapu S,

13 Jemal A, Siegel R, Xu J, Ward E: Cancer 22: 83–89. Scheller J, Rose-John S, Cheroutre H,

statistics, 2010. CA Cancer J Clin 2010; Eckmann L, Karin M: IL-6 and Stat3

60: 277–300. are required for survival of intestinal 14 Feagins LA, Souza RF, Spechler SJ: Car- epithelial cells and development of coli-

cinogenesis in IBD: potential targets for tis-associated cancer. Cancer Cell 2009;

the prevention of colorectal cancer. Nat 15: 103–113.

Rev Gastroenterol Hepatol 2009; 6: 297– 305.

Innate Immunity, Inflammation and Colorectal Cancer 9

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) 31 Bollrath J, Phesse TJ, von Burstin VA, 39 Langrish CL, Chen Y, Blumenschein 45 Langowski JL, Zhang X, Wu L, Mattson Putoczki T, Bennecke M, Bateman T, WM, Mattson J, Basham B, Sedgwick JD, Chen T, Smith K, Basham B, McCla- Nebelsiek T, Lundgren-May T, Canli O, JD, McClanahan T, Kastelein RA, Cua nahan T, Kastelein RA, Oft M: IL-23 Schwitalla S, Matthews V, Schmid RM, DJ: IL-23 drives a pathogenic T cell promotes tumour incidence and

Kirchner T, Arkan MC, Ernst M, Gre- population that induces autoimmune growth. Nature 2006; 442: 461–465.

ten FR: gp130-mediated Stat3 activa- inflammation. J Exp Med 2005; 201: 46 Romano M, Sironi M, Toniatti C, Po- tion in enterocytes regulates cell sur- 233–240. lentarutti N, Fruscella P, Ghezzi P, Fag- vival and cell-cycle progression during 40 Hue S, Ahern P, Buonocore S, Kullberg gioni R, Luini W, van Hinsbergh V, colitis-associated tumorigenesis. Can- MC, Cua DJ, McKenzie BS, Powrie F, Sozzani S, Bussolino F, Poli V, Ciliberto

cer Cell 2009; 15: 91–102. Maloy KJ: Interleukin-23 drives innate G, Mantovani A: Role of IL-6 and its 32 Li Y, de Haar C, Chen M, Deuring J, and T cell-mediated intestinal inflam- soluble receptor in induction of chemo-

Gerrits MM, Smits R, Xia B, Kuipers EJ, mation. J Exp Med 2006; 203: 2473– kines and leukocyte recruitment. Im-

van der Woude CJ: Disease-related ex- 2483. munity 1997; 6: 315–325. pression of the IL6/STAT3/SOCS3 sig- 41 Cua DJ, Sherlock J, Chen Y, Murphy 47 Dominitzki S, Fantini MC, Neufert C, nalling pathway in ulcerative colitis CA, Joyce B, Seymour B, Lucian L, To Nikolaev A, Galle PR, Scheller J, Mon- and ulcerative colitis-related carcino- W, Kwan S, Churakova T, Zurawski S, teleone G, Rose-John S, Neurath MF,

genesis. Gut 2010; 59: 227–235. Wiekowski M, Lira SA, Gorman D, Becker C: Cutting edge: trans-signaling 33 Fischer M, Goldschmitt J, Peschel C, Kastelein RA, Sedgwick JD: Interleu- via the soluble IL-6R abrogates the in- Brakenhoff JP, Kallen KJ, Wollmer A, kin-23 rather than interleukin-12 is the duction of FoxP3 in naive CD4+CD25 T

Grotzinger J, Rose-John S: I. A bioactive critical cytokine for autoimmune in- cells. J Immunol 2007; 179: 2041–2045.

designer cytokine for human hemato- flammation of the brain. Nature 2003; 48 Bettelli E, Carrier Y, Gao W, Korn T,

poietic progenitor cell expansion. Nat 421: 744–748. Strom TB, Oukka M, Weiner HL, Kuch-

Biotechnol 1997; 15: 142–145. 42 Murphy CA, Langrish CL, Chen Y, roo VK: Reciprocal developmental 34 Mitsuyama K, Sata M, Rose-John S: Blumenschein W, McClanahan T, pathways for the generation of patho- Interleukin-6 trans-signaling in in- Kastelein RA, Sedgwick JD, Cua DJ: genic effector TH17 and regulatory T

flammatory bowel disease. Cytokine Divergent pro- and antiinflammatory cells. Nature 2006; 441: 235–238.

Growth Factor Rev 2006; 17: 451–461. roles for IL-23 and IL-12 in joint auto- 49 Veldhoen M, Hocking RJ, Atkins CJ, 35 Takeda K, Clausen BE, Kaisho T, Tsu- immune inflammation. J Exp Med Locksley RM, Stockinger B: TGFbeta in

jimura T, Terada N, Forster I, Akira S: 2003; 198: 1951–1957. the context of an inflammatory cyto- Enhanced Th1 activity and develop- 43 Oppmann B, Lesley R, Blom B, Timans kine milieu supports de novo differen- ment of chronic enterocolitis in mice JC, Xu Y, Hunte B, Vega F, Yu N, Wang tiation of IL-17-producing T cells. Im-

devoid of Stat3 in macrophages and J, Singh K, Zonin F, Vaisberg E, Chura- munity 2006; 24: 179–189.

neutrophils. Immunity 1999; 10: 39–49. kova T, Liu M, Gorman D, Wagner J, 50 Baltgalvis KA, Berger FG, Pena MM, 36 Madison BB, Dunbar L, Qiao XT, Zurawski S, Liu Y, Abrams JS, Moore Davis JM, Muga SJ, Carson JA: Interleu- Braunstein K, Braunstein E, Gumucio KW, Rennick D, de Waal-Malefyt R, kin-6 and cachexia in ApcMin/+ mice. DL: Cis elements of the villin gene con- Hannum C, Bazan JF, Kastelein RA: Am J Physiol Regul Integr Comp Physi-

trol expression in restricted domains of Novel p19 protein engages IL-12p40 to ol 2008; 294:R393–R401. the vertical (crypt) and horizontal (du- form a cytokine, IL-23, with biological 51 Tan W, Zhang W, Strasner A, Griven- odenum, cecum) axes of the intestine. J activities similar as well as distinct nikov S, Cheng JQ, Hoffman RM, Karin

Biol Chem 2002; 277: 33275–33283. from IL-12. Immunity 2000; 13: 715– M: Tumour-infiltrating regulatory T 37 Zou W, Restifo NP: T(H)17 cells in tu- 725. cells stimulate mammary cancer me- mour immunity and immunotherapy. 44 Trinchieri G, Pflanz S, Kastelein RA: tastasis through RANKL-RANK sig-

Nat Rev Immunol 2010; 10: 248–256. The IL-12 family of heterodimeric cy- nalling. Nature 2011; 470: 548–553. 38 Martin-Orozco N, Dong C: The IL-17/ tokines: new players in the regulation 52 Ammirante M, Luo JL, Grivennikov S,

IL-23 axis of inflammation in cancer: of T cell responses. Immunity 2003; 19: Nedospasov S, Karin M: B-cell-derived friend or foe? Curr Opin Investig Drugs 641–644. lymphotoxin promotes castration-re-

2009; 10: 543–549. sistant prostate cancer. Nature 2010;

464: 302–305. 53 Turner JR: Intestinal mucosal barrier function in health and disease. Nat Rev

Immunol 2009; 9: 799–809.

Michael Karin Laboratory of Gene Regulation and Signal Transduction Department of Pharmacology and Pathology, School of Medicine University of California, San Diego 9500 Gilman Drive La Jolla, CA 92093-0723 (USA)

E-Mail karinoffice @ ucsd.edu

10 Wang Grivennikov Karin

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 4–10 (DOI: 10.1159/000346526) Chapter 1: Innate Immunity and Inflammation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 11–14 (DOI: 10.1159/000346538)

Immunobiology of C-Type Lectin Receptors

a a a, b Susanne Roth Christina Thomas Jürgen Ruland a Institut für Klinische Chemie und Pathobiochemie, Klinikum rechts der Isar, Technische Universität München, Munich , and b Laboratory of Signaling in the Immune System, Helmholtz Zentrum München – German Research Center for Environmental Health, Neuherberg , Germany

A b s t r a c t naling pathways that ultimately allow myeloid C-type lectin receptors (CLRs) that signal via the kinase cells to elicit inflammation and shape adaptive Syk are an important class of pattern recognition recep- immune responses. One key signaling cascade tors in the innate immune system. They recognize path- that is crucial for inflammatory responses, in- ogen- and host-derived danger signals, and are best cluding those initiated by CLRs, is the NF-␬ B sig- known for their role in antifungal immunity. Here, we re- naling pathway [4] . view recent insights into the molecular mechanisms of CLR signaling, and their significance in host defense. Copyright © 2013 S. Karger AG, Basel Danger Recognition by C-Type Lectin Receptors

Cells of the innate immune system such as den- This discussion focuses on a specific sub-family dritic cells, macrophages and neutrophils are of CLRs that are characterized by their ability to equipped with a wide array of germline-encoded recruit and activate the tyrosine kinase Syk [5–7] . pattern recognition receptors (PRRs) that are es- Dectin-1 is the prototypic Syk-activating CLR, sential for host protection and tissue homeostasis and the first CLR identified to be essential for [1] . These include the transmembrane Toll-like host defense [5–7] . It is known for its role in anti- receptors and C-type lectin receptors (CLRs), as fungal immunity, and possesses an extracellular well as cytosolic sensors of the nucleotide-bind- C-type lectin domain that recognizes ␤ -glucans ing oligomerization domain-like receptor and in fungal cell walls [7]. The intracellular signaling retinoic acid-inducible gene-I-like helicase fami- domain of dectin-1 contains an immunoreceptor lies [2, 3]. PRRs detect not only pathogen-associ- tyrosine-based activation motif (ITAM)-like mo- ated molecular patterns derived from viruses, tif [2, 6]. Upon ligand binding, the ITAM is phos- bacteria, or fungi, but also endogenous damage- phorylated by Src family tyrosine kinases, there- associated molecular patterns, which are released by creating a docking site for the recruitment and upon tissue injury. Upon sensing their cognate activation of Syk [2, 3, 6]. Other CLRs that are ligands, PRRs engage distinct intracellular sig- involved in anti-fungal immunity are dectin-2 and mincle [8–11] . However, in contrast to dec- [22] . Initial work demonstrated that Card9-defi- tin-1, dectin-2 and mincle do not possess ITAMs cient mice are highly susceptible to infection themselves, but instead associate with the ITAM- with the opportunistic fungal pathogen Candida containing signaling adapter Fc receptor-␥ chain albicans, and that Card9 is critically required for for Syk activation [6, 10, 11] . cytokine production upon dectin-1 stimulation In addition to fungi, CLRs can also detect mi- [20, 21] . The requirement of Card9 for anti-fun- crobial pathogens and endogenous ligands. Min- gal immunity is also reflected in humans, where cle, for example, has been recently identified as a homozygous loss-of-function mutation in the activating receptor that recognizes the my- CARD9 results in high susceptibility to fungal cobacterial cord factor, trehalose-6,6-dimycolate infections [23] . However, in the context of infec- [12, 13], and also the self-ligand SAP130 that is tion, Card9 probably integrates signals not only released by necrotic cells [14] . The ITAM-con- from dectin-1, but also from dectin-2 [8] and taining CLR CLEC9a is a danger receptor for the mincle [13, 24] . Thus, although there is some re- recognition of endogenous signals released upon dundancy at the receptor level, Card9 represents cellular damage [15] . Recent work has identified a non-redundant factor critical for anti-fungal the ligands for CLEC9a to be cytoskeletal compo- defense. nents [16, 17] . Interestingly, recognition of non- Molecularly, Card9 cooperates with the adapt- fungal ligands by CLRs is not always protective. er protein Bcl10 and the paracaspase Malt1 to se- Dectin-2 can, in addition to fungi, also recognize lectively transduce signals from Syk to the canon- the helminth Schistosoma mansoni [18] . Howev- ical IKK-dependent NF-␬ B pathway. Signaling er, it is thought that signaling via dectin-2 during via the Card9, Bcl10, Malt1 complex operates in- schistosomal infection contributes to immuno- dependently from CLR-induced ROS production pathology [18]. Similarly, recognition of dengue and phagocytosis [6] . Recent findings have given virus by CLEC5a, which signals via Syk by using insight into the mechanisms of how Syk-coupled the ITAM-containing adapter protein DAP-12, is CLRs activate the Card9 complex. Stimulation of required for lethal disease caused by this virus innate immune cells with CLR ligands induces [19] . Syk-dependent phosphorylation and activation of the serine/threonine kinase PKC ␦ [25] . PKC ␦ then phosphorylates Card9 at Thr231, which is C-Type Lectin Receptor Effector Pathways required for the signal-induced association of Card9 with Bcl10 and Malt1, and the subsequent Syk activation by CLRs drives several cellular re- recruitment of TAK1 for activation of the canon- sponses, including phagocytosis of fungal parti- ical NF- ␬B pathway. Consistently, PKC ␦-defi- cles, production of microbicidal reactive oxygen cient dendritic cells are defective in innate im- species (ROS), as well as altered gene expression mune responses to dectin-1, dectin-2 or mincle [6]. Activation of NF- ␬B signaling is a critical stimulation, and PKC␦ -deficient mice are highly event downstream of numerous Syk-coupled susceptible to fungal infection [25] . CLRs, and the ability of these receptors to acti- The proinflammatory cytokine interleukin- vate NF- ␬ B signaling requires the adapter pro- 1 ␤ (IL-1␤ ) is critical for host defense against tein Card9 [6, 20, 21] . Card9 possesses a caspase fungal infection [26] . The production of IL-1 ␤ recruitment domain (CARD) and a coiled-coil requires NF-␬ B-mediated upregulation of pro- region, and is a myeloid cell-specific member of IL-1␤ , and subsequent proteolytic conversion of a small family of CARD-coiled-coil proteins pro-IL-1 ␤ to the bioactive and secreted form. The which also includes Card10, Card11 and Card14 latter event is typically mediated by caspase 1

12 Roth Thomas Ruland

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 11–14 (DOI: 10.1159/000346538) within the context of cytosolic complexes termed Conclusions inflammasomes [27] . Upon cellular infection with C. albicans, CLR-induced Syk signaling is Syk-coupled CLRs play a broad role in innate im- required for both pro-IL-1␤ synthesis and acti- munity and can also couple innate to adaptive vation of the Nlrp3 inflammasome [28] . While immune responses. Recent studies have uncov- pro-IL-1␤ synthesis selectively requires the Card9 ered critical signaling molecules downstream of pathway, inflammasome activation by fungi in- CLRs, such as Syk, Card9 and PKC␦ . Most work volves ROS production and potassium efflux. has focused on the role of CLR and Card9 signal- Activation of the Nlrp3 inflammasome down- ing in anti-fungal defense. However, Card9-defi- stream of CLRs is required for anti-fungal immu- cient mice are also highly susceptible to Mycobac- nity, since Nlrp3-deficient mice are highly sus- terium tuberculosis infection [29] , presumably ceptible to C. albicans infection [28]. Thus, there due to the critical role of Card9 downstream of is a cross talk between Syk-coupled CLRs and the cord factor receptor mincle [13, 24] . Thus, ad- Nlrp3, at least in the context of fungal infection. ditional work is required to further characterize However, the exact mechanism of how CLR sig- the signaling mechanisms of CLRs, and their role naling couples to the inflammasome remains un- in other infections and sterile tissue damage. characterized. These studies will provide important insights into the molecular regulation of host defense and the control of tissue homeostasis.

References

1 Medzhitov R: Approaching the asymp- 8 Robinson MJ, Osorio F, Rosas M, Frei- 11 Wells CA, Salvage-Jones JA, Li X,

tote: 20 years later. Immunity 2009; 30: tas RP, Schweighoffer E, Gross O, Ver- Hitchens K, Butcher S, Murray RZ, 766–775. beek JS, Ruland J, Tybulewicz V, Brown Beckhouse AG, Lo YL, Manzanero S, 2 Kerrigan AM, Brown GD: Syk-coupled GD, Moita LF, Taylor PR, Reis e Sousa Cobbold C, Schroder K, Ma B, Orr S, C-type lectins in immunity. Trends C: Dectin-2 is a Syk-coupled pattern Stewart L, Lebus D, Sobieszczuk P,

Immunol 2011; 32: 151–156. recognition receptor crucial for Th17 Hume DA, Stow J, Blanchard H, Ash- 3 Osorio F, Reis e Sousa C: Myeloid C- responses to fungal infection. J Exp man RB: The macrophage-inducible

type lectin receptors in pathogen rec- Med 2009; 206: 2037–2051. C-type lectin, mincle, is an essential ognition and host defense. Immunity 9 Saijo S, Ikeda S, Yamabe K, Kakuta S, component of the innate immune re-

2011; 34: 651–664. Ishigame H, Akitsu A, Fujikado N, Ku- sponse to Candida albicans . J Immu-

4 Vallabhapurapu S, Karin M: Regulation saka T, Kubo S, Chung SH, Komatsu R, nol 2008; 180: 7404–7413. and function of NF-kappaB transcrip- Miura N, Adachi Y, Ohno N, Shibuya K, 12 Ishikawa E, Ishikawa T, Morita YS, tion factors in the immune system. Yamamoto N, Kawakami K, Yamasaki Toyonaga K, Yamada H, Takeuchi O,

Annu Rev Immunol 2009; 27: 693–733. S, Saito T, Akira S, Iwakura Y: Dectin-2 Kinoshita T, Akira S, Yoshikai Y, Ya- 5 Brown GD, Gordon S: Immune recogni- recognition of alpha-mannans and in- masaki S: Direct recognition of the my- tion. A new receptor for beta-glucans. duction of Th17 cell differentiation is cobacterial glycolipid, trehalose dimy-

Nature 2001; 413: 36–37. essential for host defense against Can- colate, by C-type lectin Mincle. J Exp

6 Mocsai A, Ruland J, Tybulewicz VL: dida albicans . Immunity 2010; 32: 681– Med 2009; 206: 2879–2888. The SYK tyrosine kinase: a crucial 691. 13 Schoenen H, Bodendorfer B, Hitchens player in diverse biological functions. 10 Sato K, Yang XL, Yudate T, Chung JS, K, Manzanero S, Werninghaus K, Nim-

Nat Rev Immunol 2010; 10: 387–402. Wu J, Luby-Phelps K, Kimberly RP, Un- merjahn F, Agger EM, Stenger S, An- 7 Taylor PR, Tsoni SV, Willment JA, Den- derhill D, Cruz PD Jr, Ariizumi K: Dec- dersen P, Ruland J, Brown GD, Wells C, nehy KM, Rosas M, Findon H, Haynes tin-2 is a pattern recognition receptor Lang R: Cutting edge: Mincle is essen- K, Steele C, Botto M, Gordon S, Brown for fungi that couples with the Fc re- tial for recognition and adjuvanticity of GD: Dectin-1 is required for beta-glu- ceptor gamma chain to induce innate the mycobacterial cord factor and its

can recognition and control of fungal immune responses. J Biol Chem 2006; synthetic analog trehalose-dibehenate.

infection. Nat Immunol 2007; 8: 31–38. 281: 38854–38866. J Immunol 2010; 184: 2756–2760.

Immunobiology of C-Type Lectin Receptors 13

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 11–14 (DOI: 10.1159/000346538) 14 Yamasaki S, Ishikawa E, Sakuma M, 20 Gross O, Gewies A, Finger K, Schafer 24 Werninghaus K, Babiak A, Gross O, Hara H, Ogata K, Saito T: Mincle is an M, Sparwasser T, Peschel C, Forster I, Holscher C, Dietrich H, Agger EM, ITAM-coupled activating receptor that Ruland J: Card9 controls a non-TLR Mages J, Mocsai A, Schoenen H, Finger senses damaged cells. Nat Immunol signalling pathway for innate anti-fun- K, Nimmerjahn F, Brown GD,

2008; 9: 1179–1188. gal immunity. Nature 2006; 442: 651– Kirschning C, Heit A, Andersen P, 15 Sancho D, Joffre OP, Keller AM, Rogers 656. Wagner H, Ruland J, Lang R: Adjuvan- NC, Martinez D, Hernanz-Falcon P, 21 Hara H, Ishihara C, Takeuchi A, Imani- ticity of a synthetic cord factor ana- Rosewell I, Reis e Sousa C: Identifica- shi T, Xue L, Morris SW, Inui M, Takai logue for subunit Mycobacterium tion of a dendritic cell receptor that T, Shibuya A, Saijo S, Iwakura Y, Ohno tuberculosis vaccination requires couples sensing of necrosis to immu- N, Koseki H, Yoshida H, Penninger JM, FcRgamma-Syk-Card9-dependent in-

nity. Nature 2009; 458: 899–903. Saito T: The adaptor protein CARD9 is nate immune activation. J Exp Med

16 Ahrens S, Zelenay S, Sancho D, Hanc P, essential for the activation of myeloid 2009; 206: 89–97. Kjaer S, Feest C, Fletcher G, Durkin C, cells through ITAM-associated and 25 Strasser D, Neumann K, Bergmann H,

Postigo A, Skehel M, Batista F, Thomp- Toll-like receptors. Nat Immunol 2007; Marakalala MJ, Guler R, Rojowska A,

son B, Way M, Reis e Sousa C, Schulz O: 8: 619–629. Hopfner KP, Brombacher F, Urlaub H, F-actin is an evolutionarily conserved 22 Bertin J, Guo Y, Wang L, Srinivasula Baier G, Brown GD, Leitges M, Ruland damage-associated molecular pattern SM, Jacobson MD, Poyet JL, Merriam S, J: Syk kinase-coupled C-type lectin re- recognized by DNGR-1, a receptor for Du MQ, Dyer MJ, Robison KE, DiSte- ceptors engage protein kinase C- ␦ to

dead cells. Immunity 2012; 36: 635–645. fano PS, Alnemri ES: CARD9 is a novel elicit Card9 adaptor-mediated innate

17 Zhang JG, Czabotar PE, Policheni AN, caspase recruitment domain-contain- immunity. Immunity 2012; 36: 32–42. Caminschi I, Wan SS, Kitsoulis S, Tul- ing protein that interacts with BCL10/ 26 Vonk AG, Netea MG, van Krieken JH, lett KM, Robin AY, Brammananth R, CLAP and activates NF-kappa B. J Biol Iwakura Y, van der Meer JW, Kullberg

van Delft MF, Lu J, O’Reilly LA, Josef- Chem 2000; 275: 41082–41086. BJ: Endogenous interleukin (IL)-1 alpha sson EC, Kile BT, Chin WJ, Mintern JD, 23 Glocker EO, Hennigs A, Nabavi M, and IL-1 beta are crucial for host de- Olshina MA, Wong W, Baum J, Wright Schaffer AA, Woellner C, Salzer U, fense against disseminated candidiasis.

MD, Huang DC, Mohandas N, Coppel Pfeifer D, Veelken H, Warnatz K, Taha- J Infect Dis 2006; 193: 1419–1426. RL, Colman PM, Nicola NA, Shortman mi F, Jamal S, Manguiat A, Rezaei N, 27 Strowig T, Henao-Mejia J, Elinav E, K, Lahoud MH: The dendritic cell re- Amirzargar AA, Plebani A, Hannes- Flavell R: Inflammasomes in health

ceptor Clec9A binds damaged cells via schlager N, Gross O, Ruland J, Grim- and disease. Nature 2012; 481: 278–286. exposed actin filaments. Immunity bacher B: A homozygous CARD9 muta- 28 Gross O, Poeck H, Bscheider M, Dostert

2012; 36: 646–657. tion in a family with susceptibility to C, Hannesschlager N, Endres S, Hart-

18 Ritter M, Gross O, Kays S, Ruland J, fungal infections. N Engl J Med 2009; mann G, Tardivel A, Schweighoffer E,

Nimmerjahn F, Saijo S, Tschopp J, Lay- 361: 1727–1735. Tybulewicz V, Mocsai A, Tschopp J, land LE, Prazeres da Costa C: Schisto- Ruland J: Syk kinase signalling couples soma mansoni triggers Dectin-2, to the Nlrp3 inflammasome for anti-

which activates the Nlrp3 inflamma- fungal host defence. Nature 2009; 459: some and alters adaptive immune re- 433–436.

sponses. Proc Natl Acad Sci USA 2010; 29 Dorhoi A, Desel C, Yeremeev V, Pradl

107: 20459–20464. L, Brinkmann V, Mollenkopf HJ, Han- 19 Chen ST, Lin YL, Huang MT, Wu MF, ke K, Gross O, Ruland J, Kaufmann SH: Cheng SC, Lei HY, Lee CK, Chiou TW, The adaptor molecule CARD9 is essen- Wong CH, Hsieh SL: CLEC5A is critical tial for tuberculosis control. J Exp Med

for dengue-virus-induced lethal dis- 2010; 207: 777–792.

ease. Nature 2008; 453: 672–676.

Prof. Jürgen Ruland Klinikum r.d. Isar Institute for Immunology Schneckenburgerstrasse 8 DE–81675 Munich (Germany)

E-Mail jruland@ lrz.tu-muenchen.de

14 Roth Thomas Ruland

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 11–14 (DOI: 10.1159/000346538) Chapter 1: Innate Immunity and Inflammation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524)

Mechanisms of IL-1 Maturation in Neutrophils

Arun K. Mankan Veit Hornung

Unit for Clinical Biochemistry, Institute for Clinical Chemistry and Clinical Pharmacology, University Hospital, University of Bonn, Bonn , Germany

A b s t r a c t The presence of exogenous or endogenous harm- The presence of foreign biological or physical matter and ful stimuli can trigger inflammatory responses. also endogenous harmful material can elicit potent im- A key component of the initial inflammatory re- mune responses. Upon contact with these agents, che- sponse is the secretion of proinflammatory che- mokines are released that promote the recruitment of mokines and cytokines that promote the recruit- specialized immune cells, including neutrophils. Neutro- ment and activation of different immune cells. phils can not only phagocytose these noxious substanc- Neutrophils are, depending on the context, es but also secrete cytokines that initiate inflammation amongst the first set of immune cells that are re- and also help to amplify and direct an adaptive immune cruited to the inflamed site. Neutrophils along response. Among these, IL-1␤ is a key cytokine that plays with resident macrophages phagocytose, digest an important role in the immediate inflammatory re- and try to eliminate the injurious agent. If the in- sponse and also in the orchestration of subsequent flammation-initiating factor is not resolved by adaptive immune responses. In macrophages and mono - this first response, additional cells including cells cytes, IL-1 ␤ is produced as a pro-cytokine and undergoes of the adaptive immune system are recruited to enzymatic cleavage by the cysteine protease caspase 1. help neutralize the injurious agent [1, 2] . Caspase 1 itself is processed into its active form upon activation of a multimeric protein complex known as the inflammasome. The mechanisms of processing and se- Neutrophils at the Interface of an cretion of IL-1 ␤ have been extensively studied, and many Inflammatory Response of the components of the inflammasome complex re- sponsible for the processing of pro-IL-1 ␤ have been iden- Microbe-associated molecular patterns (MAMPs, tified. However, the activation of IL-1 ␤ by neutrophils and also termed pathogen-associated molecular pat- the role of the inflammasome complex in these cells are terns or PAMPs) consist of a limited number of only now being addressed. In this review, we discuss the well-conserved microbial structures encoded by recent advances in our understanding of IL-1 ␤ process- the non-self genome. In analogy to MAMPs, ing by neutrophils. Copyright © 2012 S. Karger AG, Basel danger-associated molecular patterns (DAMPs) refer to the various endogenous signals that arise golysosomal vacuole, where the proteases enzy- during tissue/cell damage or cellular stress. matically destroy the engulfed particle. In the While lipopolysaccharide and peptidoglycan are majority of cases, this process of phagocytosis is some of the classical examples of MAMPs, tightly regulated; however, sometimes phagocy- HMGB1 and ATP represent typical endogenous tosis can lead to lysosomal damage and rupture, DAMPs that are released by damaged cells. The thereby releasing the contents (e.g. serine prote- cellular response to the MAMPs or DAMPs is ases) of these granules into the extracellular initiated by a set of germline-encoded receptors space. termed pattern recognition receptors (PRRs). Another key microbicidal weapon in the Four major families of PRRs are currently known: neutrophilic arsenal is the ROS. Exposure of the the Toll-like receptor system (TLRs), the RIG- neutrophils to the injurious stimuli initiates a like receptors, the C-type lectin receptors and the process termed as respiratory burst. During this nucleotide-binding domain leucine-rich repeats process, neutrophils rapidly take up oxygen, me- (NLRs). While predominately expressed on my- tabolize glucose and produce large amounts of eloid cells that are ascribed to the innate immune ROS like superoxide, hydrogen peroxide, hy- system, PRRs are also present on organ-resident droxy radicals, etc. [6]. The NADPH (nicotin- non-immune cells. The activation of PRRs by amide adenine dinucleotide phosphate oxidase) MAMPs or DAMPs leads to the recruitment of system, present in the plasma membrane and the immune cells, with neutrophils being among the phagosomes of neutrophils, generates superox- first responders. Following their chemoattrac- ide, which spontaneously forms hydrogen perox- tion, rolling, adhesion, tight adhesion and trans- ide [7]. Superoxide also reacts with nitric oxide to migration across the endothelial membrane, form peroxynitrite, which is a strong oxidant. neutrophil migration is directed towards the fo- Myeloperoxidase, an important enzyme within cus of damage. Subsequently, neutrophils sur- the azurophilic granule, promotes the enzymatic round dead cells or microbes and engulf them reaction between hydrogen peroxide and a ha- by forming pseudopodia, which finally results in lide, like chloride, to generate the potent anti- the formation of a phagosome. This phagosome microbial compound hypochlorous acid. Apart fuses with the intracellular granules forming a from their microbicidal activity, ROS species are phagolysosome, where the different cocktail of also known to activate several signal pathways proteases, peptides and reactive oxygen species [8] . (ROS) ensures the neutralization of the engulfed Neutrophil extracellular traps (NETs) consti- particle [3, 4] . Neutrophils are distinguished by tute an additional unique process by which neu- the presence of several different types of granules trophils control spread of potentially harmful that orchestrate distinct effector functions. These agents. NETs are formed as a result of release of intracellular granules, namely azurophilic gran- the genomic DNA and mitochondrial DNA by ules, specific granules and gelatinase granules are neutrophils following cell death. NETs are coated produced during different stages of maturation with proteases from intracellular granules, and of the neutrophils [5]. While azurophilic gran- by capturing floating microbial species they pro- ules contain key microbicidal proteases includ- mote their interaction with these digestive prote- ing myeloperoxidase, cathepsin G, elastase and ases [7, 9] . proteinase 3 (PR3), specific granules mainly con- While phagocytosis, ROS production and the tain lactoferrin and lysozyme, whereas the gela- formation of NETs are independent of de novo tinase granules contain gelatinase [4] . These gene expression and part of a broad non-specific granules release their contents within the pha- response directed against many exogenous or en-

16 Mankan Hornung

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) dogenous injurious agents, binding of different bridging molecule ASC (or PYCARD) that har- MAMPs or DAMPs to PRRs initiates responses bors a C-terminal caspase recruitment domain that involve activation of transcription factors (CARD) to recruit caspase 1. Upon dimerization and the subsequent transcription and translation of caspase 1, autoproteolysis is triggered, which of a large set of genes. Neutrophils are known to leads to processing at three aspartic acids leading express most of the TLR family members with to two subunits of 20 kDa (p20) and 10 kDa (p10). the exception of TLR3 [9] . Moreover, neutrophils In the end, the active caspase 1 complex contains have also been shown to express some members two heterodimers of p20 and p10. Activated cas- of the NLR and RIG-like receptor families [9–11] . pase 1 can cleave pro-IL-1␤ or pro-IL-18 and also As it has been well established, one of the down- induce a special type of cell death that is known stream consequences of the activation of these as pyroptosis [13] . At the same time, it has also PRRs includes activation of the different tran- been shown that caspase 1 is required for the se- scription factors and synthesis and secretion of cretion of IL-1␤ and various other target proteins several cytokines. The repertoire of cytokines se- [14] . Interestingly, this unconventional type of creted by neutrophils consists of proinflamma- protein secretion appears to be independent of its tory, anti-inflammatory and immunoregulatory proteolytic activity [15] . The proximity-induced cytokines of all major cytokine families, includ- activation of caspase 1 via ASC is mediated ing IL-1␣ / ␤, IL-6, TNF and type I IFNs [9] . At the through the formation of a multiprotein plat- per-cell level, the amount of cytokines secreted form, commonly referred to as the inflamma- by neutrophils appears to be lower than in mono- some [16] . Currently, at least seven different in- nuclear cells of the myeloid lineage. However, flammasome sensors, namely NLRP1, NLRP3, since at the site of inflammation neutrophils out- NLRC4, NLRP6, NLRP12, pyrin and AIM2 have number other immune cells by several orders of been identified that all utilize ASC to trigger the magnitude, even this limited secretion of cyto- formation of processive caspase 1 [17] . kines becomes significant [1] .

The Inflammasomes Synthesis and Maturation of IL-1 The NLRs form the biggest group of inflamma- Upon the activation of many PRR pathways, IL- some-forming proteins and include NLRP1, 1 ␤ is synthesized as a pro-form that is kept with- NLRP3, NLRC4, NLRP6 and NLRP12. In gen- in the cytoplasm in an inactive state. In order to eral, NLRs share three structural domains. First- activate this cytokine, IL-1␤ has to be cleaved so ly, an N-terminal effector domain, which can be that the C-terminal part can be secreted and in- a pyrin domain, a CARD, a baculovirus inhibitor teract with the IL-1 receptor complex. Caspase 1, of apoptosis protein repeat domain or as in case a cysteine protease, is the most important prote- of NLRX1 an N-terminal domain of undefined ase that enzymatically processes pro-IL-1␤ . Its function; secondly, the intermediate NACHT exact cleavage sites within the human pro-IL-1␤ (NBS; nucleotide oligomerization domain) do- are D27 and D116, whereas the latter cleavage site main, which is essential for the activation of the is the critical cleavage site to generate biologically NLRs by oligomerization and formation of the active IL-1␤ [12] . Caspase 1 is also expressed as an core structure of the inflammasome, and thirdly, inactive pro-form, and activation of caspase 1 is the C-terminal leucine-rich repeat domain [13] . thought to be initiated by dimerization, which it- Some of these NLRs have clearly defined activa- self is triggered by the multimerization of the tors; for example, anthrax lethal toxin is a well-

Mechanisms of IL-1␤ Maturation in Neutrophils 17

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) defined trigger for NLRP1 activation [18] , where- IL-1␤ -dependent inflammatory responses that as flagellin is a ligand for the NAIP5/NLRC4 occur independent of caspase 1 [31, 32] . From inflammasome [19]. NLRP3, one of the most these reports, it was concluded that proteases studied NLR proteins, is activated upon stimula- other than caspase 1 could mature pro-IL-1␤ . On tion with numerous different stimuli such as the other hand, different phenotypes of mice de- ATP, nigericin and crystalline material. The lack ficient for caspase 1 or IL-1R that were preferen- of identification of a defined NLRP3 ligand sug- tially used in in vivo studies could also be due to gests that NLRP3 acts downstream of another a redundant role of IL-1 ␣ , a cytokine that is close- upstream signaling event that integrates the sig- ly related to IL-1␤ and shares the same receptor naling pathways from several stimuli to the and as such many biological functions. NLRP3 inflammasome [13]. As such, activation Like IL-1 ␤ , IL-1 ␣ is also expressed as a pro- of NLRP3 has been the subject of several in- form, yet both the cleaved and the pro-form of IL- t ensive studies, and various models have been 1␣ can be secreted and display biological activity. proposed to explain NLRP3 activation. These Cleavage of IL-1 ␣ was initially described to be in- include: (a) potassium efflux by membrane dependent of caspase 1, yet sensitive to inhibitors channels or ionophoric compounds [20, 21] ; (b) of calpain-like proteases. Moreover, calpains have lysosomal disintegration and release of its con- been shown to process IL-1␣ in vitro [33, 34] . tent by phagocytosed material [22, 23]; (c) induc- However, since both the pro-form as well as the tion of ROS production at mitochondrial mem- cleaved cytokine are biologically active, the role of branes [24, 25], and (d) exposure to DNA from its processing by calpains remains to be identi- mitochondria [24, 26]. While it is possible that fied. Furthermore, it was initially postulated that these processes can independently activate IL-1␣ in contrast to IL-1␤ is more widely and also NLRP3, it is also conceivable that they coopera- constitutively expressed [12, 35]. However, a re- tively function to activate NLRP3. Once activat- cent report indicates that the synthesis and secre- ed, NLR proteins recruit a small bridging or tion of IL-1␣ is similar to IL-1 ␤, primarily re- adaptor molecule, known as ASC through pyrin- stricted to PRR-activated cells of the myeloid lin- pyrin domain interaction. As described above, eage [15] . In this context, it was further shown ASC in turn recruits caspase 1 through its C-ter- that IL-1 ␣ was largely cosecreted with IL-1 ␤ in minal CARD, thereby initiating the activation of response to classical inflammasome triggers and caspase 1. AIM2, although not a member of the that this secretion was also dependent on respec- NLR family of proteins, binds to dsDNA and also tive components of the inflammasome. However, promotes the formation of an ASC-dependent particulate NLRP3 activators (e.g. uric acid crys- inflammasome complex [27–29] . Additionally, tals or alum) could also trigger the release of IL-1 ␣ pyrin has been identified to form an inflamma- in an NLRP3-independent fashion. Indeed, stim- some with ASC, yet the respective ligand is still uli that induced robust Ca2+ influx with or with- unknown [30] . out concomitant NLRP3 activation triggered IL- 1␣ release. This is well in line with the fact that calpains are calcium-dependent proteases. Inter- Caspase 1-Independent Release of IL-1 estingly, the processing of pro-IL-1 ␣ by calpains in response to non-particulate NLRP3 stimuli While it is clearly documented that proteolytic (e.g. nigericin or ATP) was still NLRP3 depen- processing of IL-1␤ by caspase 1 is the limiting dent. Moreover, in line with the identified role of step for the secretion of bioactive IL-1 ␤ in most caspase 1 in unconventional protein secretion, the in vitro systems, several studies have shown release of IL-1 ␣ turned out to require caspase 1,

18 Mankan Hornung

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) Particulate stimuli ATP (e.g. MSU, alum, Nigericin cholesterol crystals)

NLRNLRP3P3

Pattern recognition receptor (e.g. TLR4) AASCS Ca2+ influx

Pro-caspase 1

Activation of NLRP3 proinflammatory Calpains gene expression

Pro-IL-1␣ Calpains Caspase 1 IL-1␣ Pro-IL-1␤ Pro-IL-1␣ Unconventional protein secretion

Caspase 1 Pro-IL-1␤ Active IL-1␤ caspase 1 Pyroptosis

Fig. 1. Mechanism of maturation and secretion of IL-1 in myeloid cells. Upon PRR activation, the expression of IL-1 ␣ and IL-1 ␤ is induced. At the same time, the expression of NLRP3 is also tightly regulated by this priming signal (blue arrows). Various stimuli subsequently trigger the formation of the NLRP3 inflammasome (red arrows). This multimer- ic signaling platform is formed by interaction between NLRP3, ASC and pro-caspase 1, which results in the activation of caspase 1 and subsequent secretion of bioactive IL-1 ␤. Particulate NLRP3 stimuli (e.g. monosodium urate, MSU, and alum) additionally induce Ca2+ influx, which results in the activation of calpains that subsequently process IL-1 ␣ . Both IL-1 ␣ and IL-1 ␤ are subsequently secreted in a caspase 1-dependent manner.

yet with the exception of particulate inflamma- A second explanation for caspase 1-indepen- some stimuli ( fig. 1 ). These types of stimuli (e.g. dent but IL-1␤ -dependent inflammatory re- endogenous crystals), however, constitute an im- sponse is the presence of alternative proteases portant component of IL-1-driven inflammation that might promote the maturation of pro-IL-1␤ . in vivo, so that under these conditions IL-1␣ re- These alternative mechanisms could either be lease would be independent of caspase 1. cell intrinsic, i.e. proteases other than caspase 1

Mechanisms of IL-1␤ Maturation in Neutrophils 19

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) that are active within the cell to cleave pro-IL-1 ␤ , active IL-1 ␤ when membranes from myeloid cells or cell extrinsic, which involve proteases that are were coincubated. Subsequent studies using puri- secreted and process pro-IL-1␤ extracellularly. fied proteases (e.g. elastase) corroborated the con- Recently, caspase 8 was shown to process pro-IL- cept of extracellular pro-IL-1␤ maturation. Most 1 ␤ in a caspase 1-indepenent manner. Maelfait et proteases that have been shown to process IL-1␤ al. [36] reported that caspase 8-mediated pro-IL- independent of caspase 1 cleave several amino ac- 1 ␤ processing occurs in response to TLR3 and ids upstream of the authentic caspase 1 cleavage TLR4 stimulation. Furthermore, they observed site with only minor loss in biological activity. that both caspase 1 and caspase 8 cleave IL-1␤ at Elastase, for example, cleaves pro-IL-1 ␤ at Y113 the same cleavage site. The authors suggested that generating bioactive IL-1 ␤, whereas PR3, another in an in vivo setting where both caspase 1 and serine protease, processes pro-IL-1 ␤ at V114. In- caspase 8 could be independently activated, there deed, most of these pro-IL-1␤ -cleaving proteases could be an enhanced secretion of IL-1␤ by mac- are serine proteases that are found in azurophilic rophages undergoing apoptosis. In another study, granules of neutrophils (see below). At the same Vince et al. [37] set out to explore the role of IAPs time, proteases derived from non-myeloid im- (inhibitor of apoptosis proteins) in the secretion mune cells were also shown to mature pro-IL-1 ␤ of IL-1␤ . In this respect, the authors observed an into bioactive IL-1␤ . Granzyme A, a serine prote- enhanced secretion of mature IL-1 ␤ from primed ase that is released from granules of cytotoxic T macrophages treated with synthetic antagonists cells and NK cells cleaves pro-IL-1 ␤ downstream of IAPs. This release of IL-1␤ was only partially of its canonical caspase 1 cleavage site at R120 reduced in NLRP3-deficient macrophages, sug- [41] . Another set of proteins belonging to the ma- gesting the involvement of another pathway. Fur- trix metalloproteinase family including MMP2, ther experiments revealed a key role for RIP3-de- MMP3 and MMP9 were also reported to process pendent caspase 8 activation in this process. On the pro-IL-1 ␤ protein to the active form [42] . the same note, Gringhuis et al. [38] recently un- raveled the relevance of the caspase 8-dependent pro-IL-1␤ processing in the recognition of fungal Mechanism of Secretion of IL-1 by pathogens by the extracellular sensor dectin-1. Neutrophils In contrast to the cell-intrinsic mechanism, several studies have suggested a whole gamut of The secretion of bioactive IL-1␤ by neutrophils extracellular proteases that can cleave pro-IL-1 ␤ . has been the subject of a number of studies. It has Keratinocytes have been reported to respond to been known for a long time that neutrophils have inflammatory stimuli by synthesizing pro-IL-1 ␤ , the capacity to secrete mature IL-1␤ [43, 44] . The yet these cells do not express caspase 1 and cannot secretion of processed IL-1␤ by in vitro cultured process pro-IL-1 ␤ [39]. Indeed, various groups macrophages was clearly dependent on caspase 1. have implicated several extracellular proteases However, considerable production of processed that can process pro-IL-1␤ to generate bioactive IL-1␤ could be detected in caspase 1-deficient IL-1␤ . One of the earliest reports in this direction mice. This was mainly attributed to caspase 1-in- was produced by Diane Mochizuki’s group [40] . dependent processing of pro-IL-1 ␤ by neutro- In this study, recombinant pro-IL-1 ␤ was incu- phils [45] . As it was already established that neu- bated with crude plasma membrane extracts from trophils, next to macrophages, were a major different myeloid and non-myeloid cells. Using source of IL-1 ␤ in vivo, suitable protease candi- IL-1␤ bioassays and Western blot, the authors dates for IL-1 processing included the above- were able to confirm the presence of cleaved bio- mentioned serine proteases present in azurophil-

20 Mankan Hornung

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) ic granules of neutrophils (neutrophil elastase, blocked both PR3 and caspase 1 and observed PR3 and cathepsin G). These proteases require better protection against cartilage and bone de- proteolytic activation themselves, which is gov- struction. Altogether, these studies raised the no- erned by the cysteine protease dipeptidyl pepti- tion that neutrophil-derived IL-1␤ was matured dase I (also known as cathepsin C). To under- independent of the inflammasome pathway, pre- stand the role of these different proteases, Coe- sumably in the extracellular space via neutro- shott et al. [46] coincubated THP-1 cells with phil-derived proteases. human neutrophils and stimulated them with Others and we have recently shown that the in- LPS and a neutrophil-specific stimulant (N-for- flammasome components NLRP3, ASC and cas- myl-Nle-Leu-Ph). Interestingly, they observed an pase 1 are expressed in neutrophils [10, 50]. Guar- increased amount of biologically active IL-1 ␤ in da et al. [50] generated a knock-in mouse that ex- the supernatant of coincubated cells when com- presses GFP downstream of the endogenous pared to individual cells alone. They subsequent- NLRP3 promoter. As expected, in myeloid cells ly incubated stimulated THP-1 cells with puri- from these mice, GFP expression was highly in- fied neutrophil elastase or PR3 and specific in- ducible upon LPS priming, and surprisingly the hibitors of these proteases and found that purified highest expression of NLRP3 was seen in neutro- PR3 was more efficient at processing pro-IL-1␤ phils. We subsequently demonstrated a non-re- than neutrophil elastase. On the same note, to dundant role for the NLRP3/ASC/caspase 1 axis in understand the relevance of some of these prote- the secretion of biologically active IL-1 ␤ by neu- ases in vivo, Hazuda et al. [47] incubated recom- trophils in response to various well-characterized binant IL-1␤ with exudate collected from differ- NLRP3 stimuli, including crystalline material ent inflammatory sites including synovial fluid [10]. To study the contribution of PR3 and elastase and bronchoalveolar lavage fluid. The authors in the secretion of IL-1␤ in our experimental set- observed secretion of biologically active IL-1␤ ting, we used PR3/elastase double-knockout mice. and identified cathepsin G as one of the several Surprisingly, we found that the neutrophils from proteases responsible for processing of pro-IL-1␤ . these mice secreted similar amounts of IL-1 ␤ as More recently, two different groups induced ar- compared to the WT mice in response to inflam- thritis in WT and caspase 1-deficient mice [48, masome activators. Of note, these experiments do 49] . The response to arthritis involves infiltration not rule out that cell-extrinsic pro-IL-1␤ process- and activation of several different immune cells ing by these enzymes can indeed occur in vivo, yet along with a large increase in the number of neu- they unequivocally demonstrate that in neutro- trophils. Despite the induction of inflammation, phils the NLRP3 inflammasome is functional and not only was the degree of inflammatory re- also non-redundant (for typical NLRP3 stimuli) to sponse between WT and Casp1 –/– mice similar, process IL-1 ␤ in a caspase 1-dependent manner. but also the production of measurable amounts of IL-1 ␤ that was reduced upon treatment with neutrophil elastase inhibitors [48]. The second Conclusion group tested response to acute and chronic ar- thritis in a similar set of mice, and detected that While ‘neutrophil-rich’ pus injected into rabbits while in a chronic arthritis model the inflamma- was already used in early attempts to identify the tory response was dependent on caspase 1, in causative agent for fever – which turned out to be acute neutrophil-infiltrated arthritis the activity IL-1␤ [44] – the exact mechanism of IL-1 ␤ pro- of caspase 1 was redundant [49]. To validate their cessing and secretion by neutrophils is still the observations in a therapeutic setting, the authors subject of controversial debates. Attempts to de-

Mechanisms of IL-1␤ Maturation in Neutrophils 21

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) termine the protease responsible for the process- that, the novel concept of caspase 8 maturing ing of IL-1␤ led to the discovery of caspase 1 and pro-IL-1 ␤ upon the initiation of the extrinsic the inflammasome complex. While it was initial- apoptosis pathway should also be considered. ly thought that neutrophils process IL-1 ␤ inde- pendent of the inflammasome system, recent studies point out a key role of the NLRP3 inflam- Acknowledgements masome in neutrophilic IL-1 ␤ secretion. At the same time, cell-extrinsic maturation of pro-IL- This work was supported by grants from the German Re- ␤ search Foundation (SFB704 and SFB670), and the Euro- 1 by neutrophil-derived proteases might be op- pean Research Council (ERC-2009-StG 243046) to V.H. erational in vivo, yet more mechanistic evidence We thank Franz Bauernfeind for critical reading of the is required to support this concept. Apart from manuscript.

References

1 Nathan C: Neutrophils and immunity: 11 Berger M, Hsieh CY, Bakele M, Marcos 19 Franchi L, Munoz-Planillo R, Nunez G: challenges and opportunities. Nat Rev V, Rieber N, Kormann M, Mays L, Sensing and reacting to microbes

Immunol 2006; 6: 173–182. Hofer L, Neth O, Vitkov L, Krautgart- through the inflammasomes. Nat

2 Williams MR, Azcutia V, Newton G, ner WD, von Schweinitz D, Kappler R, Immunol 2012; 13: 325–332. Alcaide P, Luscinskas FW: Emerging Hector A, Weber A, Hartl D: Neutro- 20 Perregaux D, Gabel CA: Interleukin-1 mechanisms of neutrophil recruitment phils express distinct RNA receptors in beta maturation and release in re-

across endothelium. Trends Immunol a non-canonical way. J Biol Chem 2012; sponse to ATP and nigericin. Evidence

2011; 32: 461–469. 287: 19409–19417. that potassium depletion mediated by 3 Mayer-Scholl A, Averhoff P, Zychlinsky 12 Dinarello CA: Immunological and in- these agents is a necessary and com- A: How do neutrophils and pathogens flammatory functions of the interleu- mon feature of their activity. J Biol

interact? Curr Opin Microbiol 2004; 7: kin-1 family. Annu Rev Immunol 2009; Chem 1994; 269: 15195–15203.

62–66. 27: 519–550. 21 Petrilli V, Papin S, Dostert C, Mayor A, 4 Segal AW: How neutrophils kill mi- 13 Mankan AK, Kubarenko A, Hornung V: Martinon F, Tschopp J: Activation of

crobes. Annu Rev Immunol 2005; 23: Immunology in clinic review series; focus the NALP3 inflammasome is triggered 197–223. on autoinflammatory diseases: inflam- by low intracellular potassium concen-

5 Borregaard N, Sorensen OE, Theil- masomes: mechanisms of activation. Clin tration. Cell Death Differ 2007; 14:

gaard-Monch K: Neutrophil granules: a Exp Immunol 2012; 167: 369–381. 1583–1589. library of innate immunity proteins. 14 Keller M, Ruegg A, Werner S, Beer HD: 22 Hornung V, Bauernfeind F, Halle A, Sam-

Trends Immunol 2007; 28: 340–345. Active caspase-1 is a regulator of un- stad EO, Kono H, Rock KL, Fitzgerald 6 Babior BM: The respiratory burst of conventional protein secretion. Cell KA, Latz E: Silica crystals and aluminum

phagocytes. J Clin Invest 1984; 73: 599– 2008; 132: 818–831. salts activate the NALP3 inflammasome 601. 15 Gross O, Yazdi AS, Thomas CJ, Masin M, through phagosomal destabilization. Nat

7 Amulic B, Cazalet C, Hayes GL, Metzler Heinz LX, Guarda G, Quadroni M, Drex- Immunol 2008; 9: 847–856. KD, Zychlinsky A: Neutrophil function: ler SK, Tschopp J: Inflammasome activa- 23 Halle A, Hornung V, Petzold GC, from mechanisms to disease. Annu Rev tors induce interleukin-1alpha secretion Stewart CR, Monks BG, Reinheckel T,

Immunol 2012; 30: 459–489. via distinct pathways with differential Fitzgerald KA, Latz E, Moore KJ, 8 Finkel T: Signal transduction by reactive requirement for the protease function of Golenbock DT: The NALP3 inflamma-

oxygen species. J Cell Biol 2011; 194: 7–15. caspase-1. Immunity 2012; 36: 388–400. some is involved in the innate immune 9 Mantovani A, Cassatella MA, Costan- 16 Schroder K, Tschopp J: The inflamma- response to amyloid-beta. Nat

tini C, Jaillon S: Neutrophils in the acti- somes. Cell 2010; 140: 821–832. Immunol 2008; 9: 857–865. vation and regulation of innate and 17 Rathinam VA, Vanaja SK, Fitzgerald 24 Nakahira K, Haspel JA, Rathinam VA, adaptive immunity. Nat Rev Immunol KA: Regulation of inflammasome sig- Lee SJ, Dolinay T, Lam HC, Englert JA,

2011; 11: 519–531. naling. Nat Immunol 2012; 13: 333–332. Rabinovitch M, Cernadas M, Kim HP, 10 Mankan AK, Dau T, Jenne D, Hornung 18 Boyden ED, Dietrich WF: Nalp1b con- Fitzgerald KA, Ryter SW, Choi AM: V: The NLRP3/ASC/Caspase-1 axis reg- trols mouse macrophage susceptibility Autophagy proteins regulate innate

ulates IL-1beta processing in neutro- to anthrax lethal toxin. Nat Genet 2006; immune responses by inhibiting the

phils. Eur J Immunol 2012; 42: 710–715. 38: 240–244. release of mitochondrial DNA medi- ated by the NALP3 inflammasome. Nat

Immunol 2011; 12: 222–230.

22 Mankan Hornung

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) 25 Zhou R, Yazdi AS, Menu P, Tschopp J: A 33 Kobayashi Y, Yamamoto K, Saido T, 41 Irmler M, Hertig S, MacDonald HR, role for mitochondria in NLRP3 in- Kawasaki H, Oppenheim JJ, Matsushi- Sadoul R, Becherer JD, Proudfoot A,

flammasome activation. Nature 2011; ma K: Identification of calcium-activat- Solari R, Tschopp J: Granzyme A is an

469: 221–225. ed neutral protease as a processing en- interleukin 1 beta-converting enzyme.

26 Shimada K, Crother TR, Karlin J, zyme of human interleukin 1 alpha. J Exp Med 1995; 181: 1917–1922.

Dagvadorj J, Chiba N, Chen S, Ramanu- Proc Natl Acad Sci USA 1990; 87: 5548– 42 Schonbeck U, Mach F, Libby P: Genera- jan VK, Wolf AJ, Vergnes L, Ojcius DM, 5552. tion of biologically active IL-1 beta by Rentsendorj A, Vargas M, Guerrero C, 34 Carruth LM, Demczuk S, Mizel SB: In- matrix metalloproteinases: a novel cas- Wang Y, Fitzgerald KA, Underhill DM, volvement of a calpain-like protease in pase-1-independent pathway of IL-1

Town T, Arditi M: Oxidized mitochon- the processing of the murine interleu- beta processing. J Immunol 1998; 161: drial DNA activates the NLRP3 inflam- kin 1 alpha precursor. J Biol Chem 3340–3346.

masome during apoptosis. Immunity 1991; 266: 12162–12167. 43 Cassatella MA: The production of cyto-

2012; 36: 401–414. 35 Rock KL, Latz E, Ontiveros F, Kono H: kines by polymorphonuclear neutro-

27 Burckstummer T, Baumann C, Bluml The sterile inflammatory response. phils. Immunol Today 1995; 16: 21–26.

S, Dixit E, Durnberger G, Jahn H, Plan- Annu Rev Immunol 2010; 28: 321–342. 44 Dinarello CA: IL-1: discoveries, contro- yavsky M, Bilban M, Colinge J, Bennett 36 Maelfait J, Vercammen E, Janssens S, versies and future directions. Eur J

KL, Superti-Furga G: An orthogonal Schotte P, Haegman M, Magez S, Bey- Immunol 2010; 40: 599–606. proteomic-genomic screen identifies aert R: Stimulation of Toll-like receptor 45 Fantuzzi G, Ku G, Harding MW, Liv- AIM2 as a cytoplasmic DNA sensor for 3 and 4 induces interleukin-1beta mat- ingston DJ, Sipe JD, Kuida K, Flavell

the inflammasome. Nat Immunol uration by caspase-8. J Exp Med 2008; RA, Dinarello CA: Response to local

2009; 10: 266–272. 205: 1967–1973. inflammation of IL-1 beta-converting 28 Fernandes-Alnemri T, Yu JW, Datta P, 37 Vince JE, Wong WW, Gentle I, Lawlor enzyme- deficient mice. J Immunol

Wu J, Alnemri ES: AIM2 activates the KE, Allam R, O’Reilly L, Mason K, 1997; 158: 1818–1824. inflammasome and cell death in re- Gross O, Ma S, Guarda G, Anderton H, 46 Coeshott C, Ohnemus C, Pilyavskaya A, sponse to cytoplasmic DNA. Nature Castillo R, Hacker G, Silke J, Tschopp J: Ross S, Wieczorek M, Kroona H,

2009; 458: 509–513. Inhibitor of apoptosis proteins limit Leimer AH, Cheronis J: Converting 29 Hornung V, Ablasser A, Charrel-Den- RIP3 kinase-dependent interleukin-1 enzyme-independent release of tumor

nis M, Bauernfeind F, Horvath G, Caf- activation. Immunity 2012; 36: 215–227. necrosis factor alpha and IL-1beta from frey DR, Latz E, Fitzgerald KA: AIM2 38 Gringhuis SI, Kaptein TM, Wevers BA, a stimulated human monocytic cell line recognizes cytosolic dsDNA and forms Theelen B, van der Vlist M, Boekhout in the presence of activated neutrophils a caspase-1-activating inflammasome T, Geijtenbeek TB: Dectin-1 is an extra- or purified proteinase 3. Proc Natl Acad

with ASC. Nature 2009; 458: 514–518. cellular pathogen sensor for the induc- Sci USA 1999; 96: 6261–6266. 30 Chae JJ, Cho YH, Lee GS, Cheng J, Liu tion and processing of IL-1beta via a 47 Hazuda DJ, Strickler J, Kueppers F, Si- PP, Feigenbaum L, Katz SI, Kastner DL: noncanonical caspase-8 inflamma- mon PL, Young PR: Processing of pre-

Gain-of-function Pyrin mutations in- some. Nat Immunol 2012; 13: 246–254. cursor interleukin 1 beta and inflam-

duce NLRP3 protein-independent in- 39 Mizutani H, Black R, Kupper TS: Hu- matory disease. J Biol Chem 1990; 265: terleukin-1beta activation and severe man keratinocytes produce but do not 6318–6322. autoinflammation in mice. Immunity process pro-interleukin-1 (IL-1) beta. 48 Guma M, Ronacher L, Liu-Bryan R,

2011; 34: 755–768. Different strategies of IL-1 production Takai S, Karin M, Corr M: Caspase 31 Dinarello CA: Interleukin-1 in the and processing in monocytes and kera- 1-independent activation of interleu-

pathogenesis and treatment of inflam- tinocytes. J Clin Invest 1991; 87: 1066– kin-1beta in neutrophil-predominant

matory diseases. Blood 2011; 117: 3720– 1071. inflammation. Arthritis Rheum 2009;

3732. 40 Black RA, Kronheim SR, Cantrell M, 60: 3642–3650. 32 Netea MG, Simon A, van de Veerdonk Deeley MC, March CJ, Prickett KS, 49 Joosten LA, Netea MG, Fantuzzi G, F, Kullberg BJ, Van der Meer JW, Wignall J, Conlon PJ, Cosman D, Hopp Koenders MI, Helsen MM, Sparrer H, Joosten LA: IL-1beta processing in host TP, et al: Generation of biologically ac- Pham CT, van der Meer JW, Dinarello defense: beyond the inflammasomes. tive interleukin-1 beta by proteolytic CA, van den Berg WB: Inflammatory

PLoS Pathog 2010; 6:e1000661. cleavage of the inactive precursor. J Biol arthritis in caspase 1 gene-deficient

Chem 1988; 263: 9437–9442. mice: contribution of proteinase 3 to caspase 1-independent production of bioactive interleukin-1beta. Arthritis

Rheum 2009; 60: 3651–3662. 50 Guarda G, Zenger M, Yazdi AS, Schro- der K, Ferrero I, Menu P, Tardivel A, Veit Hornung Mattmann C, Tschopp J: Differential Institute for Clinical Chemistry and Clinical Pharmacology expression of NLRP3 among hemato-

Unit for Clinical Biochemistry, University Hospital, University of Bonn poietic cells. J Immunol 2011; 186: 2529– Sigmund-Freud-Strasse 25 2534. DE–53127 Bonn (Germany)

E-Mail veit.hornung @ uni-bonn.de

Mechanisms of IL-1␤ Maturation in Neutrophils 23

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 15–23 (DOI: 10.1159/000346524) Chapter 2: Innate Immunity and Transplantation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 24–28 (DOI: 10.1159/000346531)

The Innate Immune System: Its Rediscovery before Toll Was Described

Walter G. Land Konrad Messmer

German Academy of Transplantation Medicine, Munich , Former Institute for Surgical Research, University of Munich, Munich , Germany

A b s t r a c t With respect to these modern notions on a In 1994, in a prospective control trial in cyclosporine- fundamental role of the innate immune system in treated, kidney transplant patients, we observed that inflammation and adaptive immunity, a peer-re- treatment of a non-specific allograft injury (postisch- viewed article that dealt with this issue already at emic reperfusion injury) leads to a significant reduction an early stage published in Transplantation in in the incidence of both specific alloimmune-mediated 1994 appears to be of interest [4] . In this article, allograft rejection and chronic allograft failure. From we described the existence of a defense system in these convincing clinical data, we concluded in terms of humans (without calling it ‘innate immunity’) an ‘argumentum e contrario’: it is the tissue injury that before first studies on the discovery of the innate induces immunity. As from where we stand today in in- immune receptor Toll were published [5, 6] . nate immunity research, these early clinical observations Interestingly enough, our article appeared just can be regarded as the discovery of the existence of a a few months before Matzinger [7] published her human innate immune system activated by tissue injury famous danger hypothesis which was remarkably and preceding adaptive immunity. in line with our clinical observations. Copyright © 2013 S. Karger AG, Basel These clinical observations derived from a prospective, randomized double-blind placebo controlled clinical trial that showed a beneficial Current notions in immunology hold that not effect of human recombinant superoxide-dis- only pathogen-induced injury but any tissue in- mutase (SOD) on acute and chronic rejection jury activates the innate immune system leading events in kidney-transplanted patients [4] . to infectious/sterile tissue inflammation and pre- The principal design of this clinical study in ceding adaptive immunity [1–3] . recipients under cyclosporine-based immunosup- Fig. 1. Clinical data from the Munich SOD trial in kidney-transplanted pa- % Placebo % tients under cyclosporine-based im- rhSOD munosuppression. Left: incidence p < 0.008 of acute rejection episodes and irre- versible graft rejection during the 50 75 first year after transplantation was p < 0.027 p < 0.033 statistically significantly reduced in rhSOD-treated recipients (n = 81) 33.3% compared to placebo-treated pa- p < 0.038 50 tients (n = 96). Right: long-term re- 18.5% sults were also significantly im- 12.5% proved in rhSOD-treated patients. This difference of survival was most 3.7% 24 68 obvious 4 years after transplanta- 0 tion, still statistically significant at 6 Acute rejection Irreversible 8-year years, and still demonstrable at 8 episodes graft rejection graft survival years although having lost its statis- tical significance.

pression consisted of intravenous administration e contrario’: tissue injury (here: allograft injury) of 200 mg of the free radical scavenger SOD given activates a biological immune system that pre- just once during surgery, that is, a few minutes be- cedes and activates adaptive immunity (here al- fore renal allograft reperfusion. During the subse- loimmunity). In the same article [4] , this conclu- quent 8-year monitoring phase, SOD-treated pa- sion was extended into a working hypothesis, tients revealed a statistically significant reduction today known as the Injury Hypothesis. As illus- in the incidence of acute rejection episodes to only trated in figure 2, a human immune system in its 18% and irreversible graft loss to 3.7%, respective- own right was proposed that is activated by non- ly. The long-term results were also significantly pathogen-induced tissue injury (here the post- improved, and most remarkably the beneficial ef- ischemic reperfusion injury to a renal allograft, fect was even demonstrable 8 years after a single that is, a situation where pathogens are obviously injection of SOD ( fig. 1 ). The therapeutic effect ob- absent) and that, after activation, leads to the in- served was dramatic: With regard to the incidence duction of an adaptive immune response (here an of acute rejection episodes under cyclosporine- adaptive alloimmune response resulting in al- based immunosuppression, the administration of lograft rejection). In the center of this immune a single dose of SOD prior to reperfusion is com- system, besides others, we proposed a role of an- parable to the application of 2 g mycophenolate tigen-presenting cells (later appreciated to be mofetil (MMF) daily, that is, an accumulating dendritic cells) activated by injury and subse- dose of 730 g MMF within 1 year! quently leading to development of adaptive im- Thus, these clinical observations indicated munity, that is cells operating as a bridge between that treatment of a non-specific allograft injury injury and adaptive immunity. (=postischemic reperfusion injury) results in a In addition, in this 1994 article, the possibility significant reduction in specific adaptive im- was discussed that adaptive immune response mune events. From these convincing clinical products (cytotoxic T lymphocytes, alloantibod- data, we concluded in terms of an ‘argumentum ies) induced by this system contribute – via endo-

The Innate Immune System: Its Rediscovery before Toll Was Described 25

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 24–28 (DOI: 10.1159/000346531) Our Injury Hypothesis – based on statistically Ischemia significant clinical data – together with Matzing- er’s Danger Hypothesis – proposed entirely on Reperfusion injury theoretical grounds – could now extend the con- ceptual framework of the late Charles Janeway Upregulation (?) of proposing that the immune system did not re- (1) DR expression, spond to all foreign antigens but only to those (2) adhesion Ag expression, (3) phagocytosis that are potentially associated with infection. (APC reactivity) Janeway’s underlying idea was that the immune system evolved to discriminate infectious non- Increased immunogenicity self from non-infectious self [10, 11]. In fact, Jane- way’s hypothesis turned out to be too simplistic. Increased host immune response Most importantly, however, his model could not explain all immune responses, in particular, not Increased incidence of acute rejection episodes the robust T cell-mediated alloimmune response Early graft loss leading to allograft rejection, a process in the ap- parent absence of microbial infection. In fact, the danger/injury model can now ex- Fig. 2. This figure is traced, redrawn, and modified from figure 2 of the original 1994 article [4]. We proposed a hu- plain why the innate immune system is able to man immune system in its own right that is activated by mount an efficient immune response against non-pathogen-induced tissue injury (here the postisch- harmful injurious pathogenic microorganisms, emic reperfusion injury to a renal allograft) and that, after but not against harmless non-pathogenic micro- activation, leads to the induction of an adaptive immune response (here an adaptive alloimmune response result- organisms: it is the presentation of microbial ing in allograft rejection). In the center of this immune antigens in the context of pathogen-induced tis- system (apart from others), we proposed a role of anti- sue injury that triggers an efficient immune re- gen-presenting cells activated by injury and subsequent- sponse – not simply the foreignness of microbial ly leading to the development of adaptive immunity. antigens. Likewise, the danger/injury model can also explain why the innate immune system sometimes mounts an efficient immune response against non-self foreign tissue such as transplant- thelial injuries (at that time called ‘allograft en- ed alloantigens, but sometimes not, for example dothelitis’) – to chronic allograft dysfunction (at in case of fetal semi-alloantigens [12] . The answer that time called chronic obliterative rejection is: the system distinguishes between an injured vasculopathy). transplant (rejection) and a non-injured fetus In other words, as from where we stand today, (tolerance). Again, it is the presentation of alloan- in 1994, that is before Matzinger published her tigens in the context of tissue injury that triggers danger model [7, 8] and before the groups of Hoff- an efficient alloimmune response, and not simply mann and Beutler published the discovery of Toll the foreignness of allogeneic tissue as reflected, and TLR4 [5, 6] , we had discovered the existence of for example, by an HLA-mismatch. a human innate immune system activated by tissue After the rediscovery of the innate immunity injury and preceding adaptive immunity. We only system, as published in first reports during the missed to call it innate immunity. However, 2 years late 1990s/early 2000s, a concept of the potential later, in 1996, in a review article, at least we briefly impact of the innate immune events on allograft addressed this system as ‘natural immunity’ [9] . rejection was introduced by review articles to the

26 Land Messmer

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 24–28 (DOI: 10.1159/000346531) transplant community in 2002/2003 for the first alloantigens under subimmunogenic conditions time [13–16] . In these reviews, in a subsequent ar- within an intragraft non-inflammatory milieu, ticle [17] as well as in a recently published mono- thereby potentially generating tolerogenic den- graph [18] , the original Injury Hypothesis was ex- dritic cells able to induce Foxp3+ regulatory T tended and modified several times. Along with cell-mediated innate allotolerance [20] – in fact these modifications, we coined the terms ‘innate an allotolerance-inducing principle that has been alloimmunity’ in 2002 [13] and ‘damage-associ- proven to be successful in elegant experiments in ated molecular patterns’ (DAMPs) in 2003 [16] . mice by Verginis et al. [21]. Indeed, such a con- Moreover, in 2003, we predicted that TLR4 medi- cept may be discussed in view of our early clinical ates reperfusion injury-induced inflammatory observation that the effect of a single intravenous response, a prediction that was confirmed only 1 injection of SOD to transplant patients is demon- year later by data of the group of Kupiec-Weglin- strable even 8 years after its application – indicat- ski at UCLA showing that TLR4 activation medi- ing that this free radical scavenger must have ates liver ischemia/reperfusion inflammatory re- induced a fundamental long-lasting active sup- sponse [19] . pressive process. In particular, in these review articles, we pro- Although in earlier times heavily opposed and posed that oxidative stress to the brain-dead do- later on notoriously neglected by the transplant nor organism as well as the generation of reactive community, our Injury Hypothesis has just re- oxygen species during reperfusion of the al- cently gained center stage and obviously appears lograft represent acute injurious events to the do- well accepted by leading transplant immunolo- nor organ that, in turn, lead to acute rejection. By gists [22, 23] . Time seems now to be ripe to think activation of donor/recipient PRR-bearing den- of new immunosuppressive strategies in organ dritic cells of the innate immune system via inter- transplantation such as interfering with the do- action of DAMPs with Toll-like receptors, these nor’s innate immune system during organ re- events lead to initiation of adaptive alloimmuni- moval and the recipient’s innate immune system ty [17] . during allograft reperfusion, for example with In our last published review article, evidence the use of antioxidants, anti-IL-1␤ inhibitors, is provided in support of the notion that preven- anticomplement agents, and polyclonal antilym- tion of oxidative allograft injury may operate as phocyte preparations. an efficient tool in the clinical situation to present

References

1 Chen GY, Nunez G: Sterile inflamma- acute and chronic rejection events in 7 Matzinger P: Tolerance, danger, and tion: sensing and reacting to damage. recipients of cadaveric renal transplants. the extended family. Annu Rev Immu-

Nat Rev Immunol 2010; 10: 826–837. Transplantation 1994; 57: 211–217. nol 1994; 12: 991–1045. 2 Cassel SL, Sutterwala FS: Sterile in- 5 Lemaitre B, Nikolas E, Michaut L, et al: 8 Matzinger P: The evolution of the flammatory responses mediated by the The dorsoventral regulatory gene cas- danger theory. Interview by Lauren NLRP3 inflammasome. Eur J Immunol sette spätzle/Toll/cactus controls the Constable, Commissioning Editor.

2010; 40: 607–611. potent antifungal response in Drosoph- Expert Rev Clin Immunol 2012; 8: 311–

3 Rock KL, Lai JJ, Kono H: Innate and ila adults. Cell 1996; 86: 973–983. 317. adaptive immune responses to cell 6 Poltorak A, He X, Smirnova I, et al: 9 Land W, Messmer K: The impact of

death. Immunol Rev 2011; 243: 191–205. Defective LPS signaling in C3H/HeJ ischemia/reperfusion injury on specific 4 Land W, Schneeberger H, Schleibner S, and C57BL/10ScCr mice: mutations in and non-specific, early and late chronic

et al: The beneficial effect of human re- Tlr4 gene. Science 1998; 282: 2085– events after organ transplantation.

combinant superoxide dismutase on 2088. Transplant Rev 1996; 10: 108–127.

The Innate Immune System: Its Rediscovery before Toll Was Described 27

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 24–28 (DOI: 10.1159/000346531) 10 Janeway CA Jr: Approaching the 15 Land W: Allograft injury mediated by 19 Zhai Y, Shen XD, O’Connell R, et al: asymptote? Evolution and revolution in reactive oxygen species: from con- Cutting edge: TLR4 activation mediates immunology. Cold Spring Harb Symp served proteins of Drosophila to acute liver ischemia/reperfusion inflamma-

Quant Biol 1989; 54: 1–13. and chronic rejection of human trans- tory response via IFN regulatory factor 11 Janeway CA Jr: The immune system plants. II. Role of reactive oxygen spe- 3-dependent MyD88-independent

evolved to discriminate infectious non- cies in the induction of the heat shock pathway. J Immunol 2004; 173: 7115– self from noninfectious self. Immunol response as a regulator of innate im- 7119.

Today 1992; 13: 11–16. munity. Transplant Rev 2003; 17: 31–44. 20 Land WG: Emerging role of innate im- 12 Kanellopoulos-Langevin C, Caucheteux 16 Land W: Allograft injury mediated by munity in organ transplantation part SM, Verbeke P, Ojcius DM: Tolerance of reactive oxygen species: from con- III: the quest for transplant tolerance the fetus by the maternal immune sys- served proteins of Drosophila to acute via prevention of oxidative allograft tem: role of inflammatory mediators at and chronic rejection of human trans- injury and its consequences. Trans-

the feto-maternal interface. Reprod plants. III. Interaction of (oxidative) plant Rev (Orlando) 2012; 26: 88–102.

Biol Endocrinol 2003; 1: 121. stress-induced heat shock proteins with 21 Verginis P, McLaughlin KA, Wucher- 13 Land W: Postischemic reperfusion in- Toll-like receptor-bearing cells of in- pfennig KW, et al: Induction of anti- jury to allografts – a case for innate nate immunity and its consequences gen-specific regulatory T cells in wild-

immunity? Eur Surg Res 2002; 34: 160– for the development of acute and type mice: visualization and targets of 169. chronic allograft rejection. Transplant suppression. Proc Natl Acad Sci USA

14 Land W: Allograft injury mediated by Rev 2003; 17: 67–86. 2008; 105: 3479–3484. reactive oxygen species: from con- 17 Land WG: The role of postischemic 22 Wood KJ, Goto R: Mechanisms of rejec- served proteins of Drosophila to acute reperfusion injury and other nonanti- tion: current perspectives. Transplan-

and chronic rejection of human trans- gen-dependent inflammatory pathways tation 2012; 93: 1–10. plants. I. Demonstration of reactive in transplantation. Transplantation 23 Benichou G, Tonsho M, Tocco G, et al:

oxygen species in reperfused allografts 2005; 79: 505–514. Innate immunity and resistance to and their role in the initiation of innate 18 Land WG: Innate Alloimmunity, Part tolerogenesis in allotransplantation.

immunity. Transplant Rev 2002; 16: 2: Innate Immunity and Allograft Re- Front Immunol 2012; 3: 73. 192–204. jection. Lengerich, Pabst Science Pub- lishers, 2011.

Prof. Dr. med. Walter G. Land Köglweg 32 DE–82024 Taufkirchen (Germany)

E-Mail wgland@ gmail.com

28 Land Messmer

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 24–28 (DOI: 10.1159/000346531) Chapter 2: Innate Immunity and Transplantation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527)

Innate-Adaptive Immune Responses in Organ Ischemia/Reperfusion Injury

Haofeng Ji Yuan Zhai Jerzy W. Kupiec-Weglinski

Dumont-UCLA Transplantation Center, Division of Liver and Pancreas Transplantation, Department of Surgery, David Geffen School of Medicine at UCLA, Los Angeles, Calif., USA

A b s t r a c t Introduction Ischemia and reperfusion injury (IRI) in the liver, a major complication of hemorrhagic shock, resection and trans- In 2009, there were 16,104 patients in the US plantation, is a dynamic process that involves two in- awaiting a liver transplant [1] . However, there terrelated phases of ischemic damage and inflamma- were only 6,320 liver transplants performed. tion-mediated reperfusion injury. Multiple cellular and Thus, we are short about 10,000 donor livers per molecular pathways regulate tissue damage, and inte- year to meet the needs, and this shortage has been gration of different players into a unified mechanism is consistent for the past decade. What are the im- warranted. Here, we focus on recent progress in the plications of such a status quo? In 2009, 1,513 pa- mechanism of liver innate immune activation by IR. tients died while waiting on the liver transplant, Kupffer cells, dendritic cells, NK, as well as T cells initiate and another 1,030 patients were removed from local inflammation response, the hallmark of IRI, by utiliz- the waiting list because they became too sick. ing distinctive immune receptors to recognize various This organ shortage has prompted to use ‘ex- molecules, both endogenous and exogenous. The inter- panded criteria donor’ organs, i.e. those from locked molecular signaling pathways in multiple liver cell older, steatotic, or non-heart beating donors, as types, the IRI kinetics, and positive versus negative regu- well as those that have undergone prolonged pe- latory loops in the innate immune activation are dis- riods of cold storage. However, the expanded cri- cussed. Better appreciation of molecular interactions teria organs are more susceptible to ischemia and that mediate these cascades should allow for the devel- reperfusion injury (IRI), the major component of opment of much needed novel therapeutic approaches the ‘harvesting’ insult [2]. Indeed, IRI contrib- against IRI in transplant recipients. utes to the acute donor organ shortage, and it of- Copyright © 2013 S. Karger AG, Basel ten leads to poor early graft function or primary non-function. Moreover, the cellular damage surrounding donor organ removal, storage and engraftment can significantly impact transplan- volved both MyD88-dependent and -indepen- tation outcomes because it is a major risk factor dent pathways [8] . A more recent study points to for both early acute rejection as well as chronic TLR4- and MyD88-mediated signaling in renal allograft dysfunction. Thus, our better apprecia- IRI [9]. Although both liver and renal IRI depend tion of ischemic injury should lead to clinical on TLR4 signaling, liver injury is dominated by a strategies aimed at reducing organ damage and MyD88-independent whereas renal IRI is medi- the ability of transplant centers to use expanded ated rather by a MyD88-dependent pathway. TLR criteria donor livers. However, despite obvious signaling has also been shown to be important in clinical significance, the mechanisms of the ini- myocardial IRI, as TLR4 deficiency decreased the tiating events that account for IR damage are only infarct size and local inflammation [10] . partially understood and remain one of the most The pathogenic role of endogenous DAMPs understudied areas in transplantation immuno- that activate TLRs during IR is becoming eluci- biology [3, 4] . dated. The two broadly categorized endogenous TLR ligands include those released from necrotic cells, such as heat shock proteins (60, 70, Gp96), Toll-Like Receptor System in Ischemia/ high mobility group box-1 (HMGB1) and DNA/ Reperfusion-Triggered Innate Immune RNA complexes, and those derived from degrad- Activation ed extracellular matrix, i.e. heparan sulfate, hyal- uronan, fibrinogen, fibronectin A domain and Figure 1 identifies the major players in the im- tenascin C. Liver non-parenchymal TLR4 is the mune cascade of liver IRI, i.e. Kupffer cells (KCs), main target for HMGB1 [11] . Of note, the HMGB1 dendritic cells (DCs), neutrophils (PMNs), T cells biology is becoming complex, with unsolved is- and NK/NKT cells [4] . In the first, immune-trig- sues concerning molecular nature of TLR4 bind- gering stage, liver-resident KCs and DCs become ing (direct stimulation or by enhancing LPS ac- activated by endogenous damage-associated mo- tivity), and the putative role of other binding lecular pattern (DAMP) and/or pathogen-associ- moieties, such as RAGE (receptor for advanced ated molecular pattern molecules. These danger glycation end products) [12] . molecules or ‘alarmins’, induced during the cel- The intracellular TLR9 at the interface of mi- lular stress, are critical for local inflammation, crobial and sterile inflammation detects bacterial the hallmark of IRI, which propagates the in- and endogenous DNA, and serves as a sensor of flammation response to the whole organ. In the necrotic cell death that exacerbates innate im- second stage, activated monocytes and PMNs are mune activation [13] . TLR9 expressed in PMNs is recruited into the organ to sustain immune acti- essential for IR-induced ROS, IL-6 and TNF- ␣ , vation and to further amplify tissue destruction. data supported by in vitro finding where DNA The sentinel Toll-like receptor (TLR) system released from necrotic hepatocytes activated liv- plays important roles in the pathophysiology of er NPCs to produce a similar proinflammatory IRI in many organ systems. Our group was one of gene pattern in the culture. Since TLR9 signals the first to document that TLR4 activation medi- via MyD88, a question arises as to why TLR4-me- ates liver IR-inflammatory response via IFN reg- diated damage in the liver appears MyD88 inde- ulatory factor-3-dependent but MyD88-indepen- pendent? It is likely that MyD88-independent ac- dent mechanism [5] . Although in a neonatal tivation of KC and DC by DAMPs in the early model of small bowel IRI, TLR2 KO mice sus- phase of organ injury (1–6 h) depends on the di- tained greater intestinal damage [6], TLR2 defi- rect cytotoxicity of soluble TNF-␣ -enriched in- ciency was protective in renal IRI [7] and in- flammatory milieu. In the later stage ( 112 h),

30 Ji Zhai Kupiec-Weglinski

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527) Th1 TCRįDŽ CD154 IL-12R Th1 KC/DC CD40 Stat4 CXCR3

IFNGR ,)1Dž PMN TLR9 RAGE TLR4 NK IL-10R CD39

PMN NKT NKT

A2AR

Inflammatory milieu iTCR TNF-α, IL-1β, IL-6, ROS PMN IL-12, IL-10 CXCL10, CCL2, CXCL8 HMGB1 CD1d

DNA Reperfusion injury

Hepatocyte

Ischemic injury

Fig. 1. A simplified scheme of liver immune activation against IRI. In the first phase, ischemia induces necrotic cell death, providing ‘danger’ molecules (e.g. HMGB1, DNA fragments) to activate TLR4, RAGE and TLR9 signaling on KCs/ DCs and neutrophils. T cells (particularly CD4 effectors) may facilitate innate activation via the CD154-CD40 pathway. In the second phase, IFN-␥ produced by T cells, NKT and NK cells enhances innate activation. In addition, CD1d and CD39 activate NKT and NK cells, respectively. The activation cascade progresses via positive and negative regulatory loops. Proinflammatory microenvironment further activates and recruits immune cells to promote cytotoxicity against liver parenchymal cells. IL-10 counter-regulates proinflammatory activation, whereas adenosine receptor 2A inhibits NKT cell activation. Type II NKT cells may downregulate IFN- ␥ production by type I NKT cells.

however, newly recruited and activated PMNs Non-Toll-Like Receptors in Ischemia/ may require MyD88 signaling. Reperfusion-Mediated Innate Immune The role of other TLR family members in IR Activation innate immune activation remains to be elucidat- ed. Recently, TLR3, which recognizes necrotic The role of non-TLR innate receptors, such as cell-derived RNA products, has been shown to nucleotide-binding domain-like receptor (NLR) sustain inflammation in a murine gastrointesti- and RIG-I-like receptor, in modulating cyto- nal ischemia model [14]. Thus, different TLRs kine/chemokine programs and regulating local may operate in different cell types during the immune responses has only recently become course of IRI in organ transplant recipients. unraveled. Unlike TLRs, imbedded in cell sur-

Innate-Adaptive Immune Responses in Organ IRI 31

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527) face, non-TLR molecules recognize pathogen- seems to be the direct hepatocyte protection. Of associated molecular patterns within the cyto- note, IL-10 neutralization is both necessary and sol. However, analogous to TLRs, they may trig- sufficient to recreate the proinflammatory phe- ger local inflammation and immune activation. notype in otherwise IRI-resistant organs [18] . The necrotic cells are sensed by inflammasomes Multiple innate immune cell types (DCs, mac- to release proinflammatory mediators. One rophages and PMNs) may all produce IL-10 and member of the NLR family, NLRP3 (NLR fam- exert regulatory functions. Recently, KCs were ily, pyrin domain containing 3), has been in- found to prevent organ damage in a bowel-con- volved in PMN recruitment to sites of hepatic gested total liver IRI model by an IL-10-mediat- necrosis in a model of sterile inflammation [15] . ed mechanism [19] . As NALP3 silencing attenuated liver damage, As IR activates the pro- and anti-inflammato- reduced production of IL-1 ␤ , IL-18, TNF-␣ , and ry genes, the question arises which mechanisms IL-6, diminished HMGB1 levels and decreased determine the nature of immune responses, and local cell infiltration [16], the inflammasome dictate IRI outcomes? Is it the kinetics difference signaling network may be essential in organ of innate immune gene induction or cell respon- IRI. Indeed, ATP released from necrotic cells siveness to gene products, in such a way that pro- may activate NLRP3 inflammasome to generate inflammatory phenotype precedes the anti-in- inflammatory microenvironment, which in flammatory development in a self-limited tissue turn alerts circulating PMNs to adhere within damage? Alternatively, endogenous ligands gen- liver sinusoids. The crosstalk between TLR and erated at different IR stages may trigger pro- and inflammasome pathways in IR-triggered organ anti-inflammatory responses sequentially, pos- inflammation/damage warrants further analy- sibly via distinct TLRs and/or in different cell sis. types. Moreover, macrophage (or KC)-T cell in- teractions in IR organs may dictate the nature of local immune response by engaging additional IL-10 in Ischemia/Reperfusion Innate signaling pathways. Addressing these key ques- Immune Signaling tions should help to identify novel targets to sup- press proinflammatory arm without interfering IR-triggered innate activation is a self-limiting with the desired immune regulatory functions in local reaction with active regulatory mechanism transplant recipients. modulated by IL-4, IL-10 and IL-13 [4] . These cytokines are often spared or even heightened in an IR-resistant organ. Although inhibitory to T Cells in Ischemia/Reperfusion Innate IR-induced TNF-␣ and/or IL-1 ␤ when adminis- Immune Activation tered exogenously, the endogenous IL-4, IL-10 and IL-13 may not necessarily exert immune- In addition to KCs and DCs, T cells, NK and regulatory functions. Indeed, although IL-13 NKT cells are also involved in IR innate immune KO mice suffer from exacerbated liver injury, activation ( fig. 1 ). Although liver IRI occurs in compared with IL-13 proficient counterparts, the absence of exogenous T cell antigen (Ag) in IR-induced TNF- ␣ and CXCL8 (MIP-2) pro- syngeneic recipients, and proceeds in the sterile duction in IL-13-deficient and WT mice was environment, CD4 T cells are instrumental in comparable [17]. Although the cellular sources IR-triggered proinflammatory response [20] . In- of IL-4 and IL-13 in the liver have not been deed, livers in CD4-deficient hosts or in WT mice precisely defined, their most significant effect depleted of CD4 T cells are protected from IR

32 Ji Zhai Kupiec-Weglinski

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527) damage. In contrast, CD8 T cell depletion does different T cell types at distinct disease stages not affect the severity of IRI. The immediate ef- regulate IR innate immune activation. fect of Ag-specific CD4 T cells poses a number of In addition to ‘traditional’ T cells, NK and questions regarding their activation and function NKT cells may also play a phase-specific role in in an innate immune response. Since naïve T cell the mechanism of IRI ( fig. 1 ). Although depletion activation requires specific Ag and more than a of NK1.1 cells (NK/NKT) fails to affect the sever- few hours to differentiate into the functional ef- ity of IRI at early time points [21] , it reduces the fectors, it is unlikely that they play a decisive role hepatocellular damage in the later phase [25] . IR- in the acute IRI phase. However, in the absence triggered activation of NKT cells (comprise al- of cognate Ag stimulation, effector T cells can most 50% of liver T cells) is mediated by CD1d, become activated quickly by proinflammatory expressed by most liver cells and present glyco- mediators to either secrete cytokines or upreg- lipid Ags, released possibly by necrotic cells, to ulate the costimulatory molecule programs. NKT cell invariant TCRs. Furthermore, NKT Such a scheme fits nicely with organ IRI kinetics. cell subsets play distinctive roles in vivo. Indeed, For instance in the liver, resident CD4 T cells type II NKT cells were shown to prevent liver IRI are enriched with effector memory phenotype when activated by specific glycolipid ligand sul- CXCR3+CD62L lowCD4+ T cells. Second, the use fatide [26] . IR-triggered NK cell activation is de- of CD4-blocking Ab to prevent CD4 TCR-medi- pendent on CD39 to hydrolyze ADP to AMP. In- ated activation without concomitant cell deple- deed, CD39-deficient livers were resistant to IRI, tion has shown that de novo CD4 T cell activation and IFN- ␥ production by their NK cells was di- was not required for their function [21] . Third, minished, possibly due to P2 receptor activation mice sensitized with allo-Ag were characterized [27] . Thus, T cells, NKT cells and NK cells are all by enhanced CD4 T cell-mediated IR immune re- involved, possibly at different stages of IR innate sponses [20] . Fourth, RAG-deficient TCR trans- activation, by providing costimulatory signaling genic mice with fewer effector T cells suffered a via direct cell-cell interactions or cytokine stim- less severe IR damage [22] . ulation to KCs and/or DCs. This, in turn, pro- CD4 T cells can differentiate into Th1, Th2, motes proinflammatory phenotype by augment- Th17 or Treg effectors. Results from T cell trans- ing IL-12 and/or inhibiting IL-10. fer studies indicate a more profound effect of Stat4-dependent cells in nude mice, consistent with the pathogenic role of Th1 cells in the mech- Conclusion anism of IRI [23] . It is unknown as to how CD4 T cells function in hepatic IRI. A possible reverse This chapter focuses on innate-adaptive mecha- CD154-CD40 costimulation between CD4 T nisms of IR immune activation and cell damage. cells and macrophages in innate immune activa- The challenge is to understand how different tion has been proposed [21]. Although livers fol- pathways may function in a synergistic or coun- lowing CD154 blockade in WT mice or in nude teractive fashion. Our better appreciation of cell mice reconstituted with CD154-deficient CD4 T interactions and their molecular pathways should cells are IRI resistant, agonist anti-CD40 Ab re- result in the development of much needed novel stored organ injury in CD4 KO mice. CD40 sig- therapeutic approaches to ameliorate organ IRI naling has also been shown to synergize with and prevent its adverse consequences in trans- TLR ligands to facilitate proinflammatory phe- plant recipients. notype in DC and macrophages, in particular, the elaboration of functional IL-12p70 [24] . Thus,

Innate-Adaptive Immune Responses in Organ IRI 33

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527) References

1 http://optn.transplant.hrsa.gov/data/. 10 Hua F, Ha T, Ma J, Li Y, Kelley J, Gao X, 19 Ellett JD, Atkinson C, Evans ZP, Amani 2 Starzl TE, Fung JJ: Themes of liver Browder IW, et al: Protection against Z, Balish E, Schmidt MG, van Rooijen

transplantation. Hepatology 2010; 51: myocardial ischemia/reperfusion in- N, et al: Murine Kupffer cells are pro- 1869–1684. jury in TLR4-deficient mice is medi- tective in total hepatic ischemia/reper- 3 Lentsch AB, Kato A, Yoshidome H, Mc- ated through a phosphoinositide fusion injury with bowel congestion

Masters KM, Edwards MJ: Inflamma- 3-kinase-dependent mechanism. J through IL-10. J Immunol 2010; 184:

tory mechanisms and therapeutic strat- Immunol 2007; 178: 7317–7324. 5849–5858. egies for warm hepatic ischemia/ 11 Tsung A, Hoffman RA, Izuishi K, 20 Shen X, Reng F, Gao F, Uchida Y, Busut-

reperfusion injury. Hepatology 2000; Critchlow ND, Nakao A, Chan MH, til RW, Kupiec-Weglinski JW, Zhai Y:

32: 169–173. Lotze MT, et al: Hepatic ischemia/re- Alloimmune activation enhances in- 4 Zhai Y, Busuttil RW, Kupiec-Weglinski perfusion injury involves functional nate tissue inflammation/injury in a JW: Liver ischemia and reperfusion TLR4 signaling in nonparenchymal mouse model of liver ischemia/reperfu-

injury: new insights into mechanisms cells. J Immunol 2005; 175: 7661–7668. sion injury. Am J Transplant 2010; 10: of innate-adaptive immune-mediated 12 Hreggvidsdottir HS, Ostberg T, Waha- 1729–1737. tissue inflammation. Am J Transplant maa H, Schierbeck H, Aveberger AC, 21 Shen X, Wang Y, Gao F, Ren F, Busuttil

2011; 11: 1563–1569. Klevenvall L, Palmblad K, et al: The RW, Kupiec-Weglinski JW, Zhai Y: CD4 5 Zhai Y, Shen XD, O’Connell R, Gao F, alarmin HMGB1 acts in synergy with T cells promote tissue inflammation Lassman C, Busuttil RW, Cheng G, et endogenous and exogenous danger sig- via CD40 signaling without de novo al: Cutting edge: TLR4 activation medi- nals to promote inflammation. J Leu- activation in a murine model of liver

ates liver ischemia/reperfusion inflam- koc Biol 2009; 86: 655–662. ischemia/reperfusion injury. Hepatol-

matory response via IFN regulatory 13 Bamboat ZM, Balachandran VP, Ocuin ogy 2009; 50: 1537–1546. factor 3-dependent MyD88-indepen- LM, Obaid H, Plitas G, DeMatteo RP: 22 Kuboki S, Sakai N, Tschop J, Edwards

dent pathway. J Immunol 2004; 173: Toll-like receptor 9 inhibition confers MJ, Lentsch AB, Caldwell CC: Distinct 7115–7119. protection from liver ischemia-reperfu- contributions of CD4+ T cell subsets in

6 Aprahamian CJ, Lorenz RG, Harmon sion injury. Hepatology 2010; 51: 621–632. hepatic ischemia/reperfusion injury. CM, Dimmit RA: Toll-like receptor 2 is 14 Cavassani KA, Ishii M, Wen H, Schaller Am J Physiol Gastrointest Liver Physiol

protective of ischemia-reperfusion-me- MA, Lincoln PM, Lukacs NW, Hoga- 2009; 296:G1054–G1059. diated small-bowel injury in a murine boam CM, et al: TLR3 is an endogenous 23 Shen XD, Ke B, Zhai Y, Gao F, Anselmo

model. Pediatr Crit Care Med 2008; 9: sensor of tissue necrosis during acute D, Lassman CR, Busuttil RW, et al:

105–109. inflammatory events. J Exp Med 2008; Stat4 and Stat6 signaling in hepatic

7 Leemans JC, Stokman G, Claessen N, 205: 2609–2621. ischemia/reperfusion injury in mice: Rouschop KM, Teske GJ, Kirschning 15 McDonald B, Pittman K, Menezes GB, HO-1 dependence of Stat4 disruption- CJ, Akira S, et al: Renal-associated Hirota SA, Slaba I, Waterhouse CC, Beck mediated cytoprotection. Hepatology

TLR2 mediates ischemia/reperfusion PL, et al: Intravascular danger signals 2003; 37: 296–303.

injury in the kidney. J Clin Invest 2005; guide neutrophils to sites of sterile in- 24 Abdi K, Singh NJ: Antigen-activated T

115: 2894–2903. flammation. Science 2010; 330: 362–366. cells induce IL-12p75 production from 8 Shigeoka AA, Holscher TD, King AJ, 16 Zhu P, Duan L, Chen J, Xiong A, Xu Q, dendritic cells in an IFN-gamma-inde- Hall FW, Kiosses WB, Tobias PS, Mack- Zhang H, Zheng F, et al: Gene silencing pendent manner. Scand J Immunol

man N, et al: TLR2 is constitutively of NALP3 protects against liver isch- 2010; 72: 511–521. expressed within the kidney and par- emia-reperfusion injury in mice. Hum 25 Lappas CM, Day YJ, Marshall MA,

ticipates in ischemic renal injury Gene Ther 2011; 22: 853–864. Engelhard VH, Linden J: Adenosine through both MyD88-dependent and 17 Kato A, Okaya T, Lentsch AB: Endog- A2A receptor activation reduces hepat- -independent pathways. J Immunol enous IL-13 protects hepatocytes and ic ischemia reperfusion injury by in-

2007; 178: 6252–6258. vascular endothelial cells during isch- hibiting CD1d-dependent NKT cell ac-

9 Wu H, Chen G, Wyburn KR, Yin J, Ber- emia/reperfusion injury. Hepatology tivation. J Exp Med 2006; 203:

tolino P, Eris JM, Alexander SI, et al: 2003; 37: 304–312. 2639–2648. TLR4 activation mediates kidney isch- 18 Ji H, Shen X, Gao F, Ke B, Freitas MC, 26 Arrenberg P, Maricic I, Kumar V: Sulfa- emia/reperfusion injury. J Clin Invest Uchida Y, Busuttil RW, et al: Pro- tide-mediated activation of type II nat-

2007; 17: 2847–2859. grammed death-1/B7-H1 negative co- ural killer T cells prevents hepatic isch- stimulation protects mouse liver emic reperfusion injury in mice.

against ischemia and reperfusion in- Gastroenterology 2011; 140: 646–655.

jury. Hepatology 2010; 52: 1380–1389. 27 Beldi G, Banz Y, Kroemer A, Sun X, Wu Y, Graubardt N, Rellstab A, et al: Dele- Jerzy W. Kupiec-Weglinski, MD, PhD tion of CD39 on natural killer cells at- Dumont-UCLA Transplantation Center 77-120 CHS tenuates hepatic ischemia/reperfusion

10833 Le Conte Avenue injury in mice. Hepatology 2010; 51: Los Angeles, CA 90095 (USA) 1702–1711.

E-Mail jkupiec @ mednet.ucla.edu

34 Ji Zhai Kupiec-Weglinski

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 29–34 (DOI: 10.1159/000346527) Chapter 2: Innate Immunity and Transplantation

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507)

Tolerogenic Dendritic Cells in Transplantation: From Preclinical to Clinical Application

Aurélie Moreau Maria-Cristina Cuturi

ITUN INSERM S1064, CHU Hôtel Dieu, Nantes , France

A b s t r a c t plant patients, we are now developing tolerogenic The use of immunosuppressive drugs to treat transplant monocyte-derived DC in humans. In this review, we will recipients has markedly reduced the incidence of acute discuss the preclinical studies in transplantation and de- rejection and early graft loss. However, such treatments scribe the recent clinical trials using tolerogenic mono- have numerous adverse side effects and have failed to cyte-derived DC in humans. We will also discuss the lim- prevent chronic allograft dysfunction. In this context, its and difficulties to translate preclinical experiments to therapies based on the adoptive transfer of regulatory the clinic. Copyright © 2013 S. Karger AG, Basel cells are promising strategies to induce indefinite trans- plant survival. Use of tolerogenic dendritic cells (TolDC) appears to be a good candidate as preliminary experi- ments done in rodents showed that administration of Over the last 50 years, the success rates of trans- TolDC prolong graft survival. Recipient DC, donor DC plant surgery have greatly improved. However, or donor antigen (Ag) pulsed recipient DC were used in without treatment, the development of the im- preclinical studies, and administration of these cells mune response against the donor organ by the with suboptimal immunosuppression allows them to in- transplanted patients leads to graft destruction. crease their tolerogenic potential. We have demonstrat- To block this immunological response and pro- ed that autologous unpulsed TolDC injected in the pres- tect the transplanted organs from rejection, a ence of suboptimal immunosuppression are able to range of general immunosuppressive drugs (IS) is induce Ag-specific allograft tolerance. We derived simi- necessary. Unfortunately, the use of IS induces lar TolDC in different animal models (mice and non-hu- numerous adverse side effects increasing the man primates) and confirmed their protective ability in risks of infection and cancer [1]. The aim of re- vitro and in vivo. The mechanisms involved in the toler- search in transplantation today is to find an ap- ance induced by autologous TolDC were also investigat- proach to induce long-term acceptance of trans- ed. In a perspective to use autologous DC in kidney trans- plants in the presence of little amount of IS. Cell therapy appears as an innovative and promising In transplantation, Tol present donor Ag to re- strategy to address these problems [2]. A Euro- cipient T cells either by the direct pathway, the pean project called the ‘ONE Study’ has been de- indirect pathway or the semi-direct pathway. By veloped to test the efficacy of different immuno- direct allorecognition pathway, donor DC will regulatory cell products in organ transplantation present donor peptide/donor MHC molecules to recipients. In our center, tolerogenic dendritic T cells; this Ag presentation is mainly correlated cells (TolDC) will be injected in humans in an at- with acute graft rejection. In contrast, the indi- tempt to achieve donor-specific tolerance. rect pathway is defined by the presentation of donor peptide by recipient MHC molecules and induces chronic rejection. In the semi-direct al- Tolerogenic Dendritic Cells in Animal Models lorecognition pathway, recipient DC will be able to present donor MHC molecules (transferred DC are potent antigen (Ag)-presenting cells able from donor cells) to T cells [10, 11]. In order to to induce either immunity or tolerance. The achieve donor-specific tolerance using Tol thera- mechanisms used by DC to induce tolerance py in transplantation, both donor TolDC (direct could be either by anergy/death of effector T cells pathway) or recipient TolDC loaded with donor or by generation/expansion of regulatory T cells. peptides (indirect pathway) were injected in ro- To induce tolerance using DC in animal models dent models of transplantation. Efficacy of these of transplantation, most of the studies injected different types of DC was demonstrated in mouse TolDC generated from precursors in vitro (as de- models of transplantation [8, 9]. scribed in the next paragraph). Another approach Recently, the group of Morelli demonstrated is to deliver donor Ag to quiescent conventional that injected donor DCs are actually unable to host DC in vivo. This technique was showed to be regulate directly donor-specific T cells in vivo. feasible either using CD205+ DC or DCIR2+ DC Indeed, after injection, donor TolDC die quickly [3–5]. In this second model, targeting of do- and the donor Ag is reprocessed and presented by nor MHC molecules to DCIR2+ DC leads to in- the host DC via the indirect pathway [12]. These definite survival of MHC class I mismatch skin results indicate that injected donor TolDC act as graft [6]. ‘donor Ag-transporting cells’ which could be re- lated to the DST (donor specific transfusion) pro- Generation of Tolerogenic Dendritic Cells ex vivo tocol. DST consists of the injection of blood from In the literature, the dogma described immature the donor to the patient before transplantation DC as tolerogenic cells, and mature DC as immu- and it is still used in clinic. Some studies showed nogenic ones [7]. However, some properties of that injection of DST to patients before trans- mature cells, such as Ag presentation to T cells plantation improves graft survival and function and in vivo migration to lymphoid organs, are [13, 14]. also requested by some TolDC. Thus, TolDC In parallel, we demonstrated in a rat model of could be either immature, maturation resistant full MHC-mismatched cardiac allograft that in- or alternatively activated cells [8]. Different ma- jection of unpulsed recipient DC the day before nipulations of DC ex vivo were described to ob- the graft induces a higher transplant survival tain TolDC. Indeed, to increase their efficacy and than injection of donor DC [15]. To improve the block the maturation process, DC could be mod- system and correlate with the conditions used in ified using dexamethasone, vitamin D 3 , IL-10, clinic, recipient DC were then injected into rats TGF- ␤ , rapamycin, LPS or by gene transfer [9]. treated with a suboptimal dose of the IS LF15- 0195 [16]. This deoxyspergualin analog is known

36 Moreau Cuturi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) to inhibit DC maturation by blocking NF- ␬B ac- this step. To reinforce this hypothesis, we deplet- tivation [17]. Both recipient DC and LF15-0915 ed graft passenger leukocytes (interstitial DC) have a synergic effect, and this cotreatment in- from the donor heart by administration of cyclo- duces tolerance to the allogeneic heart transplant phosphamide to the donor rat before transplan- in 90% of the treated rats. We then investigated tation. In this context, treatment of recipient an- whether the tolerance was donor specific. To an- imal with unpulsed recipient DC and LF15-0195 swer this question, tolerant rats received synge- failed to induce any graft prolongation [unpubl. neic, donor or third party skin grafts at 100 days results]. However, the effect of recipient DC and after heart transplantation. Only the third-party LF15-0195 is rescued when donor splenic APC skin graft transplant was rejected, showing that are injected in this model. This experiment high- the tolerance induced by recipient TolDC + lights the essential role of graft passenger leuko- LF15-0195 was donor specific [16]. cytes in recipient TolDC therapy. Our studies of To confirm the efficacy of cell therapy using recipient TolDC also demonstrate their ability to recipient TolDC, we generated TolDC in mouse express tolerogenic markers such as HO-1 and [18] and in non-human primates [19]. As previ- EBI3. Expression of HO-1 was observed in rat ously showed in rats, injection of mouse recipient and macaque TolDC. Expression of this marker TolDC associated with a transient anti-CD3 treat- was demonstrated earlier to be correlated with ment led to a prolongation of graft survival in skin the state of maturation of DC [21]. Indeed, im- or pancreatic islet transplantation models [Sego- mature TolDC express high levels of HO-1, and via et al., in preparation]. In macaques, we showed this molecule allows TolDC to inhibit allogeneic that TolDC are able to expand Treg in vitro [20]. T cell proliferation. In both rat and macaque, Furthermore, in a model of immune response blockade of HO-1 in TolDC impairs their ability against transgene, injection of recipient macaque to suppress T cell proliferation in vitro. Further- TolDC decreased the development of the immune more, in our model of tolerance to heart trans- response [Moreau et al., in preparation]. plantation using both recipient TolDC and LF15- 0195, blockade of HO-1 prevents tolerance in- Mechanisms of Action of Recipient Dendritic duction [19]. EBI3+, a marker expressed by Cells TolDC, has also a crucial role. Indeed, in the rat As we proved the relevant use of unpulsed re- cardiac allotransplantation model developed in cipient TolDC to induce donor-specific tolerance the laboratory using syngeneic TolDC, an in- in different animal models, we wanted to under- crease in double-negative T cells (TCR ␣ ␤ +, stand the mechanisms of action of these cells. In CD3+, CD4-CD8– NKRP1–, DNT) was ob- contrast to most of the studies using TolDC (do- served in the spleen of tolerant mice. These DNT nor TolDC or donor pulsed recipient TolDC), re- cells produce IFN-␥ which is essential for toler- cipient TolDC are injected the day prior the graft ance induction as anti-IFN- ␥ treatment of recip- (instead of one week before the graft). Indeed, af- ient mice leads to the loss of tolerance induction ter injection recipient cells migrate rapidly to the [22]. To investigate how injection of TolDC me- spleen and they are still detectable in this organ diates IFN- ␥ production by DNT and tolerance 15 days later [15]. In parallel, donor-derived induction, we identified the possible regulato- MHC class II+ cells (OX3+) from the graft are ry cytokines produced by TolDC. Our results present in the spleen 3–5 days after transplan- showed that TolDC express EBI3. By using anti- tation and seem to interact with the injected EBI3 antibody and EBI3 siRNA, we demonstrat- TolDC. We hypothesized that injected recipient ed that expression of EBI3 by TolDC is essential TolDC could be able to process the donor Ag at for IFN- ␥ production by DNT cells. Further-

Translational Research Using Tolerogenic DC 37

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) more, in our in vivo model of tolerance induc- mice [28]. This phase I study demonstrated that tion using TolDC, anti-EBI3 treatment of the re- intradermal injections of autologous TolDC cipient mice induces graft rejection showing the (both control and immunosuppressive ones) are important role of EBI3 expressed by TolDC to well tolerated and safe in diabetic patients. No ad- induce tolerance [22]. verse effect or toxicity was observed. Interesting- ly, the authors observed a statistically increased frequency of B220+CD11– lymphocytes in pa- Tolerogenic Dendritic Cells in Humans tients treated with autologous TolDC (both con- trol and immunosuppressive ones) during the Studies performed in rodents ensured the char- Tol administration period compared to baseline acterization and the efficient use of TolDC in [27]. Other clinical trials in autoimmune diseas- vivo. The goal today is to transfer this knowledge es, and more specifically in rheumatoid arthritis to humans in order to treat patients with TolDC. (RA), will soon begin. The first one will be per- formed by the team of R. Thomas in Australia Use of Human Tolerogenic Dendritic Cells in (University of Queensland). BAY11-7082-treated Clinical Trials DC loaded with citrullinated peptides derived Even if clinical protocols of vaccination using im- from RA candidate auto-Ag will be used [29]. In- munogenic DC have been tested since 15 years to deed, in a mouse model of Ag-induced arthritis, avoid the development of tumors in cancer pa- the authors showed previously that injection of tients [23, 24], less is known about use of TolDC BAY11-7082-treated DC and loaded with Ag was in the clinic. A first study published in 2001 able to suppress DTH (delayed-type hypersensi- showed the feasibility and safety of injecting au- tivity) reaction and arthritis [30]. BAY11-7082, tologous immature TolDC in healthy volunteers an NF ␬B inhibitor, affects DC differentiation [25]. In this work, immature DC were pulsed with leading to a low expression of MHC class II and peptides and injected by subcutaneous route to 2 CD40 by these DC. In vivo injection of BAY11- volunteers. Each person received a single injec- 7082-treated DC prevents priming of immunity tion of 2 million cells. DC injections were well and induces IL-10 producing CD4+ Tregs [31]. In tolerated without signs of toxicity, and no evi- parallel, another clinical trial in RA will be per- dence of autoimmunity was detected. Injection of formed by the team of C.M.U. Hilkens and J.D. DC was associated with Ag-specific inhibition of Isaacs in the UK (University of Newcastle). In effector T cell function and induction of Ag-spe- this case, autologous DC will be generated with cific CD8 Tregs in vivo [25, 26]. The first phase I dexamethasone and vitamin D 3 and loaded with clinical trial using TolDC was reported recently synovial fluid [29]. in type 1 diabetic patients [27]. Ten patients re- So far, no report of a clinical trial using TolDC ceived four intradermal injections of 10 million in transplantation has been published. As part of autologous DC. Three patients received control a European project, we will test the safety of au- DC generated in the presence of GM-CSF and tologous monocyte-derived TolDC in kidney- IL-4, and 7 patients received immunosuppressive transplanted patients. DC generated in the presence of GM-CSF, IL-4 and antisense oligonucleotides targeting CD40, Use of Autologous Tolerogenic Dendritic Cells CD80 and C86 transcripts. Use of TolDC gener- Most of the studies in animal models of trans- ated with these antisense oligonucleotides was plantation use donor TolDC or recipient TolDC shown previously by the same team to have a pre- loaded with donor Ag. In contrast, we showed the ventive and curative effect on diabetes in NOD efficacy of unpulsed recipient TolDC to induce

38 Moreau Cuturi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) tolerance. In humans, use of autologous TolDC is facility. The advantages of using elutriation in- reinforced by the safety and the feasibility of this stead of bead selection are that the cells are un- DC origin to be applied in clinic. touched, and no extra components (no beads) Regarding safety, the major risk of donor will be injected into humans. The limit of the elu- TolDC injection in transplantation is the donor triation is the presence of contaminant cells, sensitization. Indeed maturation of TolDC after which is not a real problem when autologous cells in vivo injection or presence of a slight contami- are injected. After one week of TolDC differen- nant cell product could lead to the development tiation, the presence of T cells, NK cells or B cells of the sensitization of the recipient to the donor is less than 2%. Ag. In this case, priming or a higher immune re- Ideally, cells are cultured in a closed system as sponse against the graft could potentially happen bags and potentially in high density. Suitable at the time of transplantation. Furthermore, an- clinical grade containers are available and have to other risk of injection of allogeneic cells is the be tested. The culture medium is also an impor- non-self recognition by the host immune system. tant parameter. The medium as well as the cyto- In this context, the injected cells may be deleted kines used to generate the TolDC have to be GMP by recipient NK cells [32]. clinical grade. These reagents undergo a high Regarding the application in clinic, use of au- amount of tests before shipment; they are slightly tologous TolDC allows transplantation from ei- different from the research grade ones and have ther living or deceased donors. Autologous cell to be validated for each lot. As everything has to therapy could thus be applied to all transplanted be controlled, a basic rule to produce a cell prod- organs. Another advantage of using autologous uct for clinical application is to use the simplest cell therapy is that the cell product could be pre- protocol. Temperature and pressure of the rooms, pared as soon as the patient is waiting for a trans- people allowed to work in the GMP facility, stan- plant and preserved frozen. At the time of trans- dardization and quality controls and others pa- plantation, the cells could be thawed and injected rameters also have to be considered before doing without any preliminary preparations. Use of au- a clinical trial using cell therapy. tologous TolDC is greatly applicable to the clinic as neither the donor nor the time of transplanta- Safety Assays in Animal Models tion has to be planned in advance, in accordance Even if a lot of quality control could be per- with cadaveric transplants. formed, the modifications and migration of the cells after injection in patients are difficult to vi- sualize by in vitro assays. Experiments in animal From Bench to Bedside models are essential to understand how the cells will ‘behave’ in vivo. Application of Good Manufacturing Practice We know from our expertise in animal mod- Conditions els that injected tolerogenic bone marrow DC do In our protocol to derive TolDC, monocytes are not proliferate in vivo and are present for at least enriched from leukapheresis of peripheral blood 2 weeks after injection in rats. We are currently by elutriation (purity around 90–95%). Elutria- analyzing long-term survival of these cells in the tion is a purification technique based on sepa- mouse model of skin graft. ration of cells according to their size and den- In a kidney transplantation clinical trial, the sity [33]. The cell separator enriches untouched cell therapy will be performed in patients treated monocytes in a closed and disposable system with different IS. Previous works determined the adapted for good manufacturing practice (GMP) interaction between DC therapy and IS. Indeed,

Translational Research Using Tolerogenic DC 39

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) our experiments showed that treatment of ro- TolDC [37]. Their results showed that TolDC gen- dents with rapamycin or with cyclosporin A does erated from RA patients have a similar phenotype not improve TolDC effect in our model of trans- and in vitro function to the ones generated from plantation. This is different from the injection of healthy controls [37]. In order to develop immu- allo-Ag pulsed RAPA-DC in mice that promoted notherapy for multiple sclerosis, another team de- indefinite graft survival when treated with low scribed TolDC derived from relapsing-remitting doses of rapamycin at the time of the graft [34]. multiple sclerosis (RR-MS) patients. Their results Regarding human TolDC, some in vitro studies showed that TolDC generated with vitamin D 3 showed that rapamycin increases CCR7 expres- from RR-MS patients and healthy controls have a sion which is necessary for TolDC migration to similar differentiation and function [38]. lymphoid organs [35]. Other IS, such as calci- Prior to TolDC injection, different parameters neurin inhibitors, including cyclosporin A or ta- which could influence the immunogenicity and crolimus, block MHC-restricted Ag processing survival of the injected cells have to be defined. pathways in mouse bone marrow DC in vitro [36]. One of the parameters is the time of DC injection, In the context of the ONE Study clinical trial i.e. the cells could be injected either prior to the project, the patients will receive three IS associ- transplant, at the time of the graft or a few days ated with the cell product therapy: mycopheno- later. Depending on the mechanisms of action of late mofetil, tacrolimus and prednisolone. From these cells, the best timing of cell injection could the point of view of safety, it is necessary to vali- be variable. Furthermore, patients can receive a date that the TolDC will not interfere with these single or multiple injections of TolDC depending IS actions. To answer this question, graft survival of the survival and the efficacy of the cells in vivo. after injection of each IS in presence/absence of The choice of the amount of injected cells is an- TolDC will be monitored in our mouse skin graft other parameter to determine; it is highly corre- model. So far, we observed that injection of my- lated with the number of injections and also with cophenolate mofetil induces a prolongation of the administration route of the cell product. Our graft survival and injection of TolDC does not experiments in macaques show that intradermal impair this function. A slight increase in graft injection of autologous TolDC primes an im- survival was actually detected [Segovia et al., in mune response, while intravenous injection pro- preparation]. Similar experiments using the two motes the tolerogenic function of these TolDC other IS associated or not with DC therapy are [unpubl. results]. A study also performed in mon- ongoing. The combination of three IS in the pres- keys confirmed the fact that intravenous injec- ence or absence of cell therapy will also be tested. tion of TolDC is well tolerated [39]. Finally, as we described above, IS could either potentiate or in- Tolerogenic Dendritic Cells in the Clinic hibit the effect of TolDC in vivo. The characterization of human TolDC was first performed in blood of volunteers. For the clinical trial on kidney-transplanted patients, TolDC will Conclusion be generated using monocytes from patients with chronic renal failure. Before the beginning of the Cell therapy, e.g. TolDC, is an attractive approach clinical trial, it is essential to validate our TolDC today to minimize the use of IS in transplanta- in these patients. A comparative study of the gen- tion. Studies performed in rodent models de- eration of clinical-grade TolDC in healthy volun- scribed the feasibility and the efficacy of TolDC teers and in RA patients was reported prior to an in inducing tolerance in transplantation. In par- ongoing clinical trial in RA using autologous allel, protocols to generate human TolDC in vitro

40 Moreau Cuturi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) were defined, but most of these protocols are not Acknowledgements used in the clinic. New preclinical tools such as humanized mice or non-human primates have The authors are grateful for the funding support from the ONE Study and Foundation Progreffe. emerged and appear as new strategies to help the transfer of research in animal models to clinical application in humans.

References

1 Dantal J, Hourmant M, Cantarovich D, 8 Ezzelarab M, Thomson AW: Tolerogen- 16 Beriou G, Peche H, Guillonneau C, Giral M, Blancho G, Dreno B, Soulillou ic dendritic cells and their role in Merieau E, Cuturi MC: Donor-specific

JP: Effect of long-term immunosup- transplantation. Semin Immunol 2011; allograft tolerance by administration of

pression in kidney-graft recipients on 23: 252–263. recipient-derived immature dendritic cancer incidence: randomised compar- 9 Morelli AE, Thomson AW: Tolerogenic cells and suboptimal immunosuppres-

ison of two cyclosporin regimens. dendritic cells and the quest for trans- sion. Transplantation 2005; 79: 969–

Lancet 1998; 351: 623–628. plant tolerance. Nat Rev Immunol 972.

2 Bluestone JA, Thomson AW, Shevach 2007; 7: 610–621. 17 Yang J, Bernier SM, Ichim TE, Li M, Xia EM, Weiner HL: What does the future 10 Herrera OB, Golshayan D, Tibbott R, X, Zhou D, Huang X, Strejan GH, White hold for cell-based tolerogenic therapy? Salcido Ochoa F, James MJ, Marelli- DJ, Zhong R, Min WP: LF15-0195 gen-

Nat Rev Immunol 2007; 7: 650–654. Berg FM, Lechler RI: A novel pathway erates tolerogenic dendritic cells by 3 Hawiger D, Inaba K, Dorsett Y, Guo M, of alloantigen presentation by dendritic suppression of NF-kappaB signaling

Mahnke K, Rivera M, Ravetch JV, cells. J Immunol 2004; 173: 4828–4837. through inhibition of IKK activity. J

Steinman RM, Nussenzweig MC: Den- 11 Smyth LA, Herrera OB, Golshayan D, Leukoc Biol 2003; 74: 438–447. dritic cells induce peripheral T cell un- Lombardi G, Lechler RI: A novel path- 18 Segovia M, Cuturi MC, Hill M: Prepa- responsiveness under steady state con- way of antigen presentation by dendrit- ration of mouse bone marrow-derived

ditions in vivo. J Exp Med 2001; 194: ic and endothelial cells: implications dendritic cells with immunoregulatory

769–779. for allorecognition and infectious properties. Methods Mol Biol 2011; 677:

4 Bonifaz L, Bonnyay D, Mahnke K, Ri- diseases. Transplantation 2006; 161–168. vera M, Nussenzweig MC, Steinman 82(1 suppl):S15–S18. 19 Moreau A, Hill M, Thébault P, Des- RM: Efficient targeting of protein anti- 12 Divito SJ, Wang Z, Shufesky WJ, Liu Q, champs JY, Chiffoleau E, Chauveau C, gen to the dendritic cell receptor DEC- Tkacheva OA, Montecalvo A, Erdos G, Moullier P, Anegon I, Alliot-Licht B, 205 in the steady state leads to antigen Larregina AT, Morelli AE: Endogenous Cuturi MC: Tolerogenic dendritic cells presentation on major histocompatibil- dendritic cells mediate the effects of actively inhibit T cells through heme ity complex class I products and pe- intravenously injected therapeutic im- oxygenase-1 in rodents and in nonhu-

ripheral CD8+ T cell tolerance. J Exp munosuppressive dendritic cells in man primates. FASEB J 2009; 23: 3070–

Med 2002; 196: 1627–1638. transplantation. Blood 2010; 116: 2694– 3077. 5 Bonifaz LC, Bonnyay DP, Charalam- 2705. 20 Moreau A, Chiffoleau E, Beriou G, Des- bous A, Darguste DI, Fujii S, Soares H, 13 Sharma RK, Rai PK, Kumar A, Kumar champs JY, Heslan M, Ashton-Chess J, Brimnes MK, Moltedo B, Moran TM, P, Gupta A, Kher V, Agrawal S, Bhan- Rolling F, Josien R, Moullier P, Cuturi Steinman RM: In vivo targeting of anti- dari M: Role of preoperative donor-spe- MC, Alliot-Licht B: Superiority of bone gens to maturing dendritic cells via the cific transfusion and cyclosporine in marrow-derived dendritic cells over DEC-205 receptor improves T cell vac- haplo-identical living related renal monocyte-derived ones for the expan-

cination. J Exp Med 2004; 199: 815–824. transplant recipients. Nephron 1997; sion of regulatory T cells in the ma-

6 Tanriver Y, Ratnasothy K, Bucy RP, 75: 20–24. caque. Transplantation 2008; 85: 1351– Lombardi G, Lechler R: Targeting MHC 14 Marti HP, Henschkowski J, Laux G, 1356. class I monomers to dendritic cells Vogt B, Seiler C, Opelz G, Frey FJ: Ef- 21 Chauveau C, Remy S, Royer PJ, Hill M, inhibits the indirect pathway of allo- fect of donor-specific transfusions on Tanguy-Royer S, Hubert FX, Tesson L, recognition and the production of IgG the outcome of renal allografts in the Brion R, Beriou G, Gregoire M, Josien

alloantibodies leading to long-term cyclosporine era. Transpl Int 2006; 19: R, Cuturi MC, Anegon I: Heme oxygen-

allograft survival. J Immunol 2010; 184: 19–26. ase-1 expression inhibits dendritic cell 1757–1764. 15 Peche H, Trinite B, Martinet B, Cuturi maturation and proinflammatory 7 Probst HC, Lagnel J, Kollias G, van den MC: Prolongation of heart allograft function but conserves IL-10 expres-

Broek M: Inducible transgenic mice survival by immature dendritic cells sion. Blood 2005; 106: 1694–1702. reveal resting dendritic cells as potent generated from recipient type bone inducers of CD8+ T cell tolerance. Im- marrow progenitors. Am J Transplant

munity 2003; 18: 713–720. 2005; 5: 255–267.

Translational Research Using Tolerogenic DC 41

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) 22 Hill M, Thebault P, Segovia M, Louvet 28 Machen J, Harnaha J, Lakomy R, Styche 34 Turnquist HR, Raimondi G, Zahorchak C, Bériou G, Tilly G, Merieau E, Ane- A, Trucco M, Giannoukakis N: Anti- AF, Fischer RT, Wang Z, Thomson AW: gon I, Chiffoleau E, Cuturi MC: Cell sense oligonucleotides down-regulating Rapamycin-conditioned dendritic cells therapy with autologous tolerogenic costimulation confer diabetes-preven- are poor stimulators of allogeneic dendritic cells induces allograft toler- tive properties to nonobese diabetic CD4+ T cells, but enrich for antigen-

ance through interferon-gamma and mouse dendritic cells. J Immunol 2004; specific Foxp3+ T regulatory cells and

Epstein-Barr virus-induced gene 3. Am 173: 4331–4341. promote organ transplant tolerance. J

J Transplant 2011; 11: 2036–2045. 29 Hilkens CM, Isaacs JD, Thomson AW: Immunol 2007; 178: 7018–7031. 23 Correale P, Campoccia G, Tsang KY, Development of dendritic cell-based 35 Sordi V, Bianchi G, Buracchi C, Mer- Micheli L, Cusi MG, Sabatino M, Bruni immunotherapy for autoimmunity. Int calli A, Marchesi F, D’Amico G, Yang

G, Sestini S, Petrioli R, Pozzessere D, Rev Immunol 2010; 29: 156–183. CH, Luini W, Vecchi A, Mantovani A, Marsili S, Fanetti G, Giorgi G, Francini 30 Martin E, Capini C, Duggan E, Lutzky Allavena P, Piemonti L: Differential G: Recruitment of dendritic cells and VP, Stumbles P, Pettit AR, O'Sullivan B, effects of immunosuppressive drugs on enhanced antigen-specific immune Thomas R: Antigen-specific suppres- chemokine receptor CCR7 in human reactivity in cancer patients treated sion of established arthritis in mice by monocyte-derived dendritic cells: se- with hr-GM-CSF (Molgramostim) and dendritic cells deficient in NF-kappaB. lective upregulation by rapamycin.

hr-IL-2. Results from a phase Ib clini- Arthritis Rheum 2007; 56: 2255–2266. Transplantation 2006; 82: 826–834.

cal trial. Eur J Cancer 2001; 37: 892–902. 31 Martin E, O’Sullivan B, Low P, Thomas 36 Lee YR, Yang IH, Lee YH, Im SA, Song 24 Redman BG, Chang AE, Whitfield J, R: Antigen-specific suppression of a S, Li H, Han K, Kim K, Eo SK, Lee CK: Esper P, Jiang G, Braun T, Roessler B, primed immune response by dendritic Cyclosporin A and tacrolimus, but not Mulé JJ: Phase Ib trial assessing autolo- cells mediated by regulatory T cells rapamycin, inhibit MHC-restricted gous, tumor-pulsed dendritic cells as a secreting interleukin-10. Immunity antigen presentation pathways in den-

vaccine administered with or without 2003; 18: 155–167. dritic cells. Blood 2005; 105: 3951–3955. IL-2 in patients with metastatic mela- 32 Yu G, Xu X, Vu MD, Kilpatrick ED, Li 37 Harry RA, Anderson AE, Isaacs JD,

noma. J Immunother 2008; 31: 591–598. XC: NK cells promote transplant toler- Hilkens CM: Generation and charac- 25 Dhodapkar MV, Steinman RM, Kra- ance by killing donor antigen-present- terisation of therapeutic tolerogenic

sovsky J, Munz C, Bhardwaj N: Anti- ing cells. J Exp Med 2006; 203: 1851– dendritic cells for rheumatoid arthritis.

gen-specific inhibition of effector T cell 1858. Ann Rheum Dis 2010; 69: 2042–2050. function in humans after injection of 33 Berger TG, Strasser E, Smith R, Carste 38 Raiotach-Regue D, Grau-Lopez L, immature dendritic cells. J Exp Med C, Schuler-Thurner B, Kaempgen E, Naranjo-Gómez M, Ramo-Tello C, Pu-

2001; 193: 233–238. Schuler G: Efficient elutriation of jol-Borrell R, Martínez-Cáceres E, Bor- 26 Dhodapkar MV, Steinman RM: Anti- monocytes within a closed system (Elu- ràs FE: Stable antigen-specific T-cell gen-bearing immature dendritic cells tra) for clinical-scale generation of den- hyporesponsiveness induced by tolero-

induce peptide-specific CD8(+) regula- dritic cells. J Immunol Methods 2005; genic dendritic cells from multiple scle-

tory T cells in vivo in humans. Blood 298: 61–72. rosis patients. Eur J Immunol 2012; 42:

2002; 100: 174–177. 771–782. 27 Giannoukakis N, Phillips B, Finegold 39 Zahorchak AF, Kean LS, Tokita D, D, Harnaha J, Trucco M: Phase I (safe- Turnquist HR, Abe M, Finke J, Hamby ty) study of autologous tolerogenic den- K, Rigby MR, Larsen CP, Thomson AW: dritic cells in type 1 diabetic patients. Infusion of stably immature monocyte-

Diabetes Care 2011; 34: 2026–2032. derived dendritic cells plus CTLA4Ig modulates alloimmune reactivity in rhesus macaques. Transplantation

2007; 84: 196–206.

Dr. Maria-Cristina Cuturi INSERM S1064, CHU Hotel-Dieu 30 Bd Jean Monnet FR–44093 Nantes Cedex 1 (France)

E-Mail maria-cristina.cuturi @ univ-nantes.fr

42 Moreau Cuturi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 35–42 (DOI: 10.1159/000346507) Chapter 3: Innate Immunity and Intestinal Microbiota

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 43–47 (DOI: 10.1159/000346523)

Microbiota’s Influence on Immunity

Kenya Honda Takeshi Tanoue Yuji Nagano Koji Atarashi

Department of Immunology, Graduate School of Medicine, The University of Tokyo, Tokyo , Japan

A b s t r a c t ficile (a member of cluster XI), and C. tetani (a Clostridium species belonging to clusters XIVa and IV are member of cluster I), produce pathogenic toxins, predominant organisms in the gut microbiota. They are most of the Clostridia are commensal to the host. known to affect various aspects of host biology, includ- Members of Clostridium clusters XIVa and IV are ing the optimal breakdown of foods, competition with typically described as fusiform-shaped commen- pathogenic microorganisms, maintenance of the intesti- sal bacteria in the mouse intestine [3] . Clostridi- nal epithelial barrier, and immune system development. um cluster XIVa includes the genera Clostridium , We have recently shown that a defined mixture of 46 Lachnospira, Eubacterium , Ruminococcus , Co- strains of mouse-derived Clostridium species belonging prococcus , and Roseburia [2] . The Cl ostridium to clusters XIVa and IV affects the development and func- cluster IV group includes species belonging to the tion of colonic Foxp3+ CD4+ regulatory T (Treg) cells in Clostridium , Faecalibacterium , and Ruminococ- mice. Clostridium -induced Treg cells are likely to play a cus genera. These clusters constitute 10–40% of role in the maintenance of immune homeostasis in the the total bacteria in the gut microbiota [1, 4, 5] . gut. Copyright © 2013 S. Karger AG, Basel The Clostridium cluster XIVa and IV groups, particularly the families Lachnospiraceae and Ruminococcaceae, are concentrated in the mu- cosal folds of the murine ascending colon, in con- Gut Clostridia Affect the Immune System trast to regions of the central lumen, which are colonized with Bacteroidaceae, Enterococcaceae, Most bacterial organisms in the gut microbiota and Lactobacillaceae [6] . belong to one of two phyla of anaerobes: Bacte- Clostridia employ a wide variety of enzymes roidetes and Firmicutes [1] . Firmicutes include capable of degrading polysaccharides and oligo- Clostridia and Bacilli. The Clostridia are Gram- saccharides into short-chain fatty acids (SCFAs), positive, rod-shaped, endospore-forming bacte- such as acetic acid, propionic acid, and butyric ria. Although some species of Clostridia, such as acid. SCFAs influence aspects of host physiology Clostridium perfringens (a member of Clostridi- such as gut motility and colonocyte development. um cluster I, defined by Collins et al. [2]), C. dif- In addition, SCFAs contribute to the mainte- nance of intestinal immune homeostasis. Indeed, CD4+ T cells [15] as well as innate immunity-me- butyrate metabolism is impaired in patients with diated colitis triggered by Helicobacter hepaticus ulcerative colitis, and topical sodium butyrate [16]. Some human patients with mutations in the treatments or butyrate enemas are effective ther- FOXP3 gene locus show intestinal inflammation apies [7, 8]. Germ-free (GF) mice are highly sus- [17]. Treatment of cancer patients with cytotoxic ceptible to dextran sulfate sodium-mediated coli- T cell-associated antigen-4 antibodies, a critical tis, but treatment of GF mice with acetated drink- immune suppressive molecule expressed by Treg ing water markedly improves colitis disease in- cells, induces colitis as a side effect [18] . There- dices [9] . fore, Foxp3+ CD4+ Treg cells play a critical role The Lachnospiraceae family, which consti- in intestinal immune homeostasis and tolerance tutes the Clostridium cluster XIVa, has been re- [19, 20] . ported to be significantly less abundant in in- Intestinal Foxp3+ Treg cells exhibit distinct flammatory bowel disease (IBD) patients com- characteristics from those in other secondary pared to healthy subjects [5, 10] . It has also been lymphoid organs. Treg cells in the lamina propria reported that the ileal mucosa in Crohn’s disease of the small intestine express CCR9, CD103, kill- is enriched with Escherichia coli but relatively de- er cell lectin-like receptor G1, granzyme B pleted in a subset of Clostridia [11]. Other reports (Gzmb), interleukin (IL)-10 and IL-35 [21] . IL-10 demonstrate reduced Clostridium cluster IV bac- plays an indispensable role in the suppression of teria, particularly Faecalibacterium prausnitzii , aberrant activation of myeloid cells [22], ␥ ␦ T in the intestines of IBD patients [12, 13] . Al- cells [23] , and Th17 cells [24] in the intestine, and though it remains unclear whether the decrease Treg-specific disruption of IL-10 results in severe in Clostridia is a cause or a consequence of chron- colitis [25]. Mice with a Treg-specific deficiency ic inflammation in IBD patients, it is likely that of the Stat3 gene show reduced Treg cell expres- the maintenance of the Clostridia community is sion of Gzmb, IL-10 and IL-35 and develop spon- necessary to prevent colitis. Furthermore, neona- taneous colitis [26]. Therefore, the intestinal en- tal vancomycin treatment reduces Clostridium vironment provides specific signals and gener- clusters XIVa and IV and promotes susceptibility ates a subset of Foxp3+ Treg cells with a unique to allergic asthma in a mouse model [14] . There- sub-phenotype in which STAT3 is constitutively fore, Clostridia contribute to the maintenance of activated, leading to the expression of IL-10 and local and systemic immune homeostasis. other molecules that regulate immune homeo- stasis in the gut. In addition to having intestine-specific char- Regulatory T Cells in the Gut acteristics, it has been suggested that a popula- tion of intestinal Foxp3+ Treg cells are extrathy- Recent research has provided new insight into mically developed, ‘induced Treg cells (iTreg the mechanism underlying the immunoregula- cells)’. There are several reports demonstrating tory role of the microbiota, and in particular has that Foxp3+ Treg cells can develop extrathymi- suggested the importance of Foxp3-expressing cally under certain conditions. For example, in CD4+ regulatory T (Treg) cells. Foxp3+ Treg the presence of IL-2 and TGF- ␤, naïve CD4+ T cells are present in essentially all organs, but their cells differentiate into Foxp3+ iTreg cells in vitro frequency in the gut lamina propria is signifi- [27] . Retinoic acid can substitute for IL-2 and also cantly higher than in other anatomical sites. Treg promotes iTreg induction [28] . In vivo, oral in- cells efficiently inhibit experimental colitis in- oculation of antigens efficiently induces antigen- duced by the adoptive transfer of CD45RBhi specific iTreg cells in the intestinal mucosa [29] .

44 Honda Tanoue Nagano Atarashi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 43–47 (DOI: 10.1159/000346523) TGF- ␤ receptor signaling leads to the recruit- the colon [33] . A large number of Treg cells in- ment of Smad2/3 to the conserved non-coding duced by the Clostridium mixture are negative DNA sequence element 1 (CNS1) enhancer re- for Helios, but express high levels of IL-10 [33] . gion of the Foxp3 gene. The ablation of the CNS1 Therefore, the Clostridium mixture affects both sequence of the Foxp3 locus impairs iTreg induc- the number and function of colonic Treg cells. tion in mice, particularly in the gut lymphoid Importantly, an increased abundance of the Clos- tissues and lamina propria, but does not affect tridium clusters XIVa and IV results in resistance natural Treg (nTreg) cell differentiation [30] . Im- to allergy or intestinal inflammation in experi- portantly, CNS1-deficient mice spontaneously mental murine models [33] . develop Th2-type gut and lung pathology [31] , Consistent with these findings, Geuking et al. suggesting that iTreg cells restrain Th2-type in- [34] have shown that colonization with altered flammation at mucosal sites. Because iTreg and Schaedler flora, including C. clostridioforme , in- nTreg cells express similar cell surface markers duces the accumulation of Treg cells in the colon. and suppress immune responses in similar ways, Furthermore, Sokol et al. [12, 42] have shown that it is difficult to distinguish between them. It has F. prausnitzii , which belongs to Clostridium clus- been proposed that iTreg cells can be distin- ter IV, increases IL-10 expression in peripheral guished from nTreg cells based on the lack of He- blood mononuclear cells in vitro. These reports lios expression [32] . Importantly, Helios-negative raise the possibility that Clostridia constitutively Foxp3+ Treg cells are abundant in the intestine induce the accumulation of Treg cells in the colon [33, 34] and express high levels of IL-10 and cyto- and that the relative abundance of Clostridia in toxic T cell-associated antigen-4 [33] . Perhaps the microbiota affects the immune status of the iTreg cells expressing this unique phenotype are host. induced in the gut because it is rich in TGF-␤ and Recent analyses using high-throughput se- retinoic acid. quencing have revealed that T cell receptors (TCRs) on murine colonic Treg cells are different than Treg cells isolated from other organs [38] . Clostridia-Mediated Treg Cell Induction Many of the TCRs on colonic Treg cells recognize antigens derived from the intestinal microbiota, The number of colonic Foxp3+ Treg cells is including Clostridium species. Therefore, these reduced in GF mice [33, 35–37]. In addition, reports suggest that Clostridia induce Treg cells the percentage and number of Helios-negative in an antigen-specific manner. Foxp3+ putative iTreg cells are markedly reduced in GF mice compared with specific pathogen-free mice [33, 34, 38] . It has been shown that certain Conclusion components of the microbiota specifically affect the number and function of Treg cells in the co- Treg cells are abundantly present in the intestinal lon. Mazmanian’s group has shown that polysac- mucosa. Intestinal Foxp3+ Treg cells have unique charide A from Bacteroides fragilis activates Toll- characteristics, such as the expression of IL-10. like receptor 2 on Treg cells, facilitating their The composition of the intestinal microbiota, functional maturation and the production of particularly Clostridium clusters XIVa and IV, af- IL-10 [39–41] . We have shown that colonization fects the number and function of Treg cells pres- of GF mice with a defined mixture of 46 Clos- ent in the intestine. However, the molecular tridium strains belonging to clusters XIVa and IV mechanisms underlying the microbiota-mediat- strongly induces the accumulation of Treg cells in ed development of Treg cells in the intestine are

Microbiota’s Influence on Immunity 45

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 43–47 (DOI: 10.1159/000346523) poorly understood. A multitude of factors, in- in the induction of Treg cells. Although further cluding microbial components and their metabo- studies are needed, the results discussed here lites (including SCFAs) may be involved. Consid- suggest that increasing the abundance of Clos- ering that numerous intestinal Treg cells express tridia could underlie a treatment for diseases me- TCRs that recognize Clostridia, it seems that diated by autoinflammation, autoimmunity and peptide antigens from Clostridia may be involved allergy.

References

1 Hold GL, Pryde SE, Russell VJ, Furrie 9 Maslowski KM, Vieira AT, Ng A, 15 Asseman C, Mauze S, Leach MW, Coff- E, Flint HJ: Assessment of microbial Kranich J, Sierro F, Yu D, Schilter HC, man RL, Powrie F: An essential role for diversity in human colonic samples by Rolph MS, Mackay F, Artis D, Xavier RJ, interleukin 10 in the function of regu- 16S rDNA sequence analysis. FEMS Teixeira MM, Mackay CR: Regulation latory T cells that inhibit intestinal in-

Microbiol Ecol 2002; 39: 33–39. of inflammatory responses by gut mi- flammation. J Exp Med 1999; 190: 995– 2 Collins MD, Lawson PA, Willems A, crobiota and chemoattractant receptor 1004.

Cordoba JJ, Fernandez-Garayzabal J, GPR43. Nature 2009; 461: 1282–1286. 16 Maloy KJ, Salaun L, Cahill R, Dougan Garcia P, Cai J, Hippe H, Farrow JA: 10 Manichanh C, Rigottier-Gois L, Bon- G, Saunders NJ, Powrie F: CD4+CD25+ The phylogeny of the genus Clostridi- naud E, Gloux K, Pelletier E, Frangeul T(R) cells suppress innate immune pa- um : proposal of five new genera and L, Nalin R, Jarrin C, Chardon P, Mar- thology through cytokine-dependent

eleven new species combinations. Int J teau P, Roca J, Dore J: Reduced diversity mechanisms. J Exp Med 2003; 197: 111–

Syst Bacteriol 1994; 44: 812–826. of faecal microbiota in Crohn’s disease 119. 3 Savage DC, Dubos R, Schaedler RW: revealed by a metagenomic approach. 17 Wildin RS, Smyk-Pearson S, Filipovich

The gastrointestinal epithelium and its Gut 2006; 55: 205–211. AH: Clinical and molecular features of autochthonous bacterial flora. J Exp 11 Baumgart M, Dogan B, Rishniw M, the immunodysregulation, polyendo-

Med 1968; 127: 67–76. Weitzman G, Bosworth B, Yantiss R, crinopathy, enteropathy, X linked

4 Huffnagle GB, Noverr MC (eds): GI Mi- Orsi RH, Wiedmann M, McDonough P, (IPEX) syndrome. J Med Genet 2002;

crobiota and Regulation of the Immune Kim SG, Berg D, Schukken Y, Scherl E, 39: 537–545. System. Adv Exp Med Biol. New York, Simpson KW: Culture independent 18 Langer LF, Clay TM, Morse MA: Up- Springer, vol 635, 2008. analysis of ileal mucosa reveals a selec- date on anti-CTLA-4 antibodies in 5 Frank DN, St Amand AL, Feldman RA, tive increase in invasive Escherichia clinical trials. Expert Opin Biol Ther

Boedeker EC, Harpaz N, Pace NR: Mo- coli of novel phylogeny relative to de- 2007; 7: 1245–1256. lecular-phylogenetic characterization pletion of Clostridiales in Crohn’s dis- 19 Izcue A, Coombes JL, Powrie F: Regula-

of microbial community imbalances in ease involving the ileum. ISME J 2007; tory T cells suppress systemic and mu-

human inflammatory bowel diseases. 1: 403–418. cosal immune activation to control in-

Proc Natl Acad Sci USA 2007; 104: 12 Sokol H, Pigneur B, Watterlot L, Lakh- testinal inflammation. Immunol Rev

13780–13785. dari O, et al: Faecalibacterium praus- 2006; 212: 256–271. 6 Nava GM, Friedrichsen HJ, Stappen- nitzii is an anti-inflammatory com- 20 Barnes MJ, Powrie F: Regulatory T cells beck TS: Spatial organization of intesti- mensal bacterium identified by gut reinforce intestinal homeostasis. Im-

nal microbiota in the mouse ascending microbiota analysis of Crohn disease munity 2009; 31: 401–411.

colon. ISME J 2011; 5: 627–638. patients. Proc Natl Acad Sci USA 2008; 21 Feuerer M, Hill JA, Kretschmer K, von

7 Scheppach W, Sommer H, Kirchner T, 105: 16731–16736. Boehmer H, Mathis D, Benoist C: Ge- Paganelli GM, Bartram P, Christl S, 13 Willing B, Halfvarson J, Dicksved J, nomic definition of multiple ex vivo Richter F, Dusel G, Kasper H: Effect of Rosenquist M, Jarnerot G, Engstrand L, regulatory T cell subphenotypes. Proc

butyrate enemas on the colonic mucosa Tysk C, Jansson JK: Twin studies reveal Natl Acad Sci USA 2010; 107: 5919–5924. in distal ulcerative colitis. Gastroenter- specific imbalances in the mucosa-as- 22 Takeda K, Clausen BE, Kaisho T, Tsu-

ology 1992; 103: 51–56. sociated microbiota of patients with ile- jimura T, Terada N, Forster I, Akira S: 8 Vernia P, Annese V, Bresci G, d’Albasio al Crohn’s disease. Inflamm Bowel Dis Enhanced Th1 activity and develop-

G, D’Inca R, Giaccari S, Ingrosso M, 2009; 15: 653–660. ment of chronic enterocolitis in mice Mansi C, Riegler G, Valpiani D, Caprilli 14 Russell SL, Gold MJ, Hartmann M, devoid of Stat3 in macrophages and

R: Topical butyrate improves efficacy of Willing BP, Thorson L, Wlodarska M, neutrophils. Immunity 1999; 10: 39–49. 5-ASA in refractory distal ulcerative Gill N, Blanchet MR, Mohn WW, Mc- 23 Park SG, Mathur R, LongM, Hosh N, colitis: results of a multicentre trial. Nagny KM, Finlay BB: Early life antibi- Hao L, Hayden MS, Ghosh S: T regula-

Eur J Clin Invest 2003; 33: 244–248. otic-driven changes in microbiota en- tory cells maintain intestinal homeo- hance susceptibility to allergic asthma. stasis by suppressing gammadelta T

EMBO Rep 2012; 13: 440–447. cells. Immunity 2010; 33: 791–803.

46 Honda Tanoue Nagano Atarashi

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 43–47 (DOI: 10.1159/000346523) 24 Huber S, Gagliani N, Esplugues E, 30 Zheng Y, Josefowicz S, Chaudhry A, 36 Ostman S, Rask C, Wold AE, Hult- O’Connor W Jr, Huber FJ, Chaudhry A, Peng XP, Forbush K, Rudensky AY: krantz S, Telemo E: Impaired regula- Kamanaka M, Kobayashi Y, Booth CJ, Role of conserved non-coding DNA tory T cell function in germ-free mice.

Rudensky AY, Roncarolo MG, Battaglia elements in the Foxp3 gene in regula- Eur J Immunol 2006; 36: 2336–2346.

M, Flavell RA: Th17 cells express inter- tory T-cell fate. Nature 2010; 463: 808– 37 Strauch UG, Obermeier F, Grunwald N, leukin-10 receptor and are controlled 812. Gurster S, Dunger N, Schultz M, Griese by Foxp3 and Foxp3+ regulatory CD4+ 31 Josefowicz SZ, Niec RE, Kim HY, DP, Mahler M, Scholmerich J, Rath HC: T cells in an interleukin-10-dependent Treuting P, Chinen T, Zheng Y, Umetsu Influence of intestinal bacteria on in-

manner. Immunity 2011; 34: 554–565. DT, Rudensky AY: Extrathymically duction of regulatory T cells: lessons 25 Rubtsov YP, Rasmussen JP, Chi EY, generated regulatory T cells control from a transfer model of colitis. Gut

Fontenot J, Castelli L, Ye X, Treuting P, mucosal TH2 inflammation. Nature 2005; 54: 1546–1552.

Siewe L, Roers A, Henderson WR Jr, 2012; 482: 395–399. 38 Lathrop SK, Bloom SM, Rao SM, Muller W, Rudensky AY: Regulatory T 32 Thornton AM, Korty PE, Tran DQ, Nutsch K, Lio CW, Santacruz N, Peter- cell-derived interleukin-10 limits in- Wohlfert EA, Murray PE, Belkaid Y, son DA, Stappenbeck TS, Hsieh CS: flammation at environmental interfac- Shevach EM: Expression of Helios, an Peripheral education of the immune

es. Immunity 2008; 28: 546–558. Ikaros transcription factor family system by colonic commensal micro-

26 Chaudhry A, Rudra D, Treuting P, Sam- member, differentiates thymic-derived biota. Nature 2011; 478: 250–254. stein RM, Liang Y, Kas A, Rudensky from peripherally induced Foxp3+ T 39 Mazmanian SK, Round JL, Kasper DL:

AY: CD4+ regulatory T cells control regulatory cells. J Immunol 2010; 184: A microbial symbiosis factor prevents TH17 responses in a Stat3-dependent 3433–3441. intestinal inflammatory disease. Na-

manner. Science 2009; 326: 986–991. 33 Atarashi K, Tanoue T, Shima T, Imaoka ture 2008; 453: 620–625. 27 Chen W, Jin W, Hardegen N, Lei KJ, Li A, Kuwahara T, Momose Y, Cheng G, 40 Round JL, Mazmanian SK: Inducible L, Marinos N, McGrady G, Wahl SM: Yamasaki S, Saito T, Ohba Y, Taniguchi Foxp3+ regulatory T-cell development Conversion of peripheral CD4+CD25- T, Takeda K, Hori S, Ivanov II, Umesaki by a commensal bacterium of the intes- naive T cells to CD4+CD25+ regulatory Y, Itoh K, Honda K: Induction of colon- tinal microbiota. Proc Natl Acad Sci

T cells by TGF-beta induction of tran- ic regulatory T cells by indigenous USA 2010; 107: 12204–12209.

scription factor Foxp3. J Exp Med 2003; Clostridium species. Science 2011; 331: 41 Round JL, Lee SM, Li J, Tran G, Jabri B,

198: 1875–1886. 337–341. Chatila TA, Mazmanian SK: The Toll- 28 Mucida D, Park Y, Kim G, Turovskaya 34 Geuking MB, Cahenzli J, Lawson MA, like receptor 2 pathway establishes col- O, Scott I, Kronenberg M, Cheroutre H: Ng DC, Slack E, Hapfelmeier S, McCoy onization by a commensal of the hu-

Reciprocal TH17 and regulatory T cell KD, Macpherson AJ: Intestinal bacte- man microbiota. Science 2011; 332: differentiation mediated by retinoic rial colonization induces mutualistic 974–977.

acid. Science 2007; 317: 256–260. regulatory T cell responses. Immunity 42 Sokol H, Seksik P, Furet JP, Firmesse O,

29 Curotto de Lafaille MA, Kutchukhidze 2011; 34: 794–806. Nion-Larmurier I, Beaugerie L, Cosnes N, Shen S, Ding Y, Yee H, Lafaille JJ: 35 Ishikawa H, Tanaka K, Maeda Y, Aiba J, Corthier G, Marteau P, Dore J: Low Adaptive Foxp3+ regulatory T cell-de- Y, Hata A, Tsuji NM, Koga Y, Matsumo- counts of Faecalibacterium prausnit- pendent and -independent control of to T: Effect of intestinal microbiota on zii in colitis microbiota. Inflamm Bow-

allergic inflammation. Immunity 2008; the induction of regulatory CD25+ el Dis 2009; 15: 1183–1189.

29: 114–126. CD4+ T cells. Clin Exp Immunol 2008;

153: 127–135.

Kenya Honda Department of Immunology, Graduate School of Medicine The University of Tokyo 7-3-1 Hongo, Bunkyo-ku Tokyo 113-0033 (Japan)

E-Mail kenya @ m.u-tokyo.ac.jp

Microbiota’s Influence on Immunity 47

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 43–47 (DOI: 10.1159/000346523) Chapter 3: Innate Immunity and Intestinal Microbiota

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 48–52 (DOI: 10.1159/000346510)

Inflammasomes and Mucosal Immune Response

a a a, b Eran Elinav Jorge Henao-Mejia Richard A. Flavell a b Yale University School of Medicine, New Haven, Conn. , and Howard Hughes Medical Institute, Chevy Chase, Md., USA

A b s t r a c t associated molecular patterns (DAMPs) that are Inflammasomes are a group of protein complexes built released as a result of perturbations of tissue ho- around several proteins including NLRP3, NLRC4, AIM2 meostasis caused by microbial or non-microbial and NLRP6. Recognition of a diverse range of exogenous insults enabling a general sensing of stressed tis- and endogenous microbial, stress and damage signals sue. Inflammasomes are protein complexes that by inflammasomes results in direct activation of caspase assemble following recognition by a diverse set of 1, which subsequently induces secretion of potent in- PAMPs and DAMPs that lead to the production flammatory cytokines and a form of cell death called of important proinflammatory cytokines such as pyroptosis. The molecular characterization of the inflam- interleukin (IL)-1 ␤ and IL-18 [1, 2] . Furthermore, masomes has led to a richer understanding of innate they have been found to regulate other important immune responses in the context of infection and sterile aspects of inflammation and tissue repair such inflammation. Inflammasome-mediated responses are as pyroptosis, a form of cell death. Distinct mo- now recognized as key components of immune respons- lecular activation mechanisms have identified a es to microbial infections and also as critical regulators multitude of ligands of both endogenous and ex- of metabolic disorders. ogenous origins that lead to inflammasome acti- Copyright © 2012 S. Karger AG, Basel vation [2] . Of the inflammasome effector mecha- nism, IL-1␤ and IL-18 are central proinflamma- tory cytokines that broadly affect inflammatory Inflammasome Activation and Function processes; they are synthesized as pro-proteins without a typical signal sequence that would en- Inflammation is a process involving the sensing able their secretion, and instead their activation of tissue distress, damage or other disturbances and cellular release are controlled by the cysteine of steady state. Host-derived sensing systems are protease caspase 1. Other inflammasome-regu- aimed at distinguishing between homeostasis lated effector proteins are IL-1␣ and fibroblast and threats to the host by recognition of patho- growth factor-2 [3] . Like other caspases, caspase gen-associated molecular patterns (PAMPs) in 1 is synthesized as an inactive zymogen (procas- microbes or host-derived signals called damage- pase 1) and only becomes proteolytically active after controlled dimerization in inflammasomes single layer of epithelial cells [8] . To avoid an un- that are built around one of several different mol- wanted inflammatory reaction to naturally re- ecules. These include the NOD-like receptor siding commensal microbes and food antigens (NLR) family and the pyrin and HIN200 do- while preserving the ability to react to pathogen- main-containing protein (PYHIN) family. The ic insults, epithelial, stromal and hematopoietic NLR family members include NLRP1, NLRP2, cells belonging to a complex mucosal immune NLRP3, NLRP6, NLRC4 and, potentially, system interact closely with each other and with NLRP12, which typically contain a pyrin domain the surrounding microenvironment to ensure (all except NLRC4) or a caspase activation and proper sensing and differentiation between recruitment domain (CARD; in NLRC4), a nu- harmless and potentially pathogenic signals. cleotide-binding domain (NBD) and a carboxy- When such normal regulation is perturbed, auto- terminal leucine-rich repeat (LRR) domain [1] . inflammation may develop, potentially leading The LRR domain is involved in autoinhibition at to diseases such as celiac or inflammatory bowel times when sensing does not take place, while the disease (IBD) [8]. The ability of inflammasomes NBD domains are involved in the regulation of to recognize exogenous and endogenous signals homo-oligomerization or hetero-oligomeriza- has stimulated several studies characterizing tion, which is required for inflammasome assem- their role during animal models of intestinal au- bly. Recently identified non-NLR inflammasome toinflammation such as IBD. Several groups re- sensors include the PYHIN family members ported decreased disease severity in mice defi- AIM2 and IFI16 that are characterized by a cient in caspase 1 or NLRP3, which correlated HIN200 domain, which is involved in ligand with lower IL-1␤ production during disease [9, binding [4] . On receiving an activating signal, in- 10]. Using the same model, other groups found flammasome sensors recruit procaspase 1, which that mice deficient in NLRP3, ASC and caspase 1 has a CARD domain, either directly through ho- show an exacerbated disease severity [11–14] . In motypic binding of CARD domains or indirectly the later papers, a role was suggested for the through a pyrin domain by means of the adaptor NLRP3 inflammasome in the promotion of tis- apoptosis-associated speck-like protein contain- sue regeneration in response to injury. ing a CARD (ASC). Our recent work offered an alternative expla- nation for these discrepancies. We have identified a new inflammasome, the NLRP6 inflamma- Inflammasomes and the Intestinal Mucosal some, which participates in the steady-state regu- Immune Response lation of the commensal microflora [5]. Deficien- cy in components of the NLRP6 inflammasome In the last decade, inflammasomes have been was found to elicit a dysbiotic colitogenic mi- shown to play central roles in diverse mammali- croflora, dominated by the presence of the taxa an physiological processes such as the antimicro- Prevotellaceae and TM7, which in turn were as- bial response against bacteria, viruses and para- sociated with enhanced susceptibility for colon- sites, the pathogenesis of metabolic disease and ic autoinflammation in NLRP6 inflammasome- tumorigenesis [1, 5–7] . Different and often con- deficient mice [5]. Our results demonstrated that flicting roles for different inflammasomes have regulation of the microflora by an epithelial in- been suggested in the mucosal immune response, flammasome was mediated, at least in part, and in particular in the gastrointestinal tract. In through the induction of basal secretion of IL-18 the gastrointestinal mucosa, the host is separated by epithelial cells. Our results suggested an in- from an intricate microbial ecosystem by only a ability of NLRP6 inflammasome-deficient mice

Inflammasomes and Mucosal Immune Response 49

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 48–52 (DOI: 10.1159/000346510) (Nlrp6 –/– , Asc –/– , caspase 1 –/– ) to regulate the mi- enhanced inflammation-induced tumorigenesis croflora in any given animal facility, thereby pro- through impaired mucosal self-renewal and pro- viding a plausible explanation for the different liferation mediated by alterations in the intesti- results obtained at different facilities. The IL- nal stem cell niche, or alternatively by an im- 18-dependent and -independent mechanism by paired hematopoietic immune response [15, 16] . which different inflammasomes participate in It remains to be determined whether the effect of the regulation of the mucosal homeostasis, in- NLRP6 on tumorigenesis is a direct consequence cluding the composition of the microbiome, tis- of NLRP6 deficiency in one or more intestinal sue regeneration and repair, remain to be fully cell types or whether it is indirectly induced by elucidated. Intriguingly, wild-type mice housed the documented alterations in the regulation of with NLRP6-inflammasome-deficient mice ac- microflora composition of these mice. quire their colitogenic microflora and develop We recently reported that the epithelial phenotypes that may be profoundly different NLRC4 inflammasome might also participate from that of wild-type mice housed alone [5] . So in the prevention of intestinal inflammation- far, the factors inducing or regulating the forma- induced tumorigenesis by demonstrating that tion of the NLRP6 inflammasome as well as the NLRC4-deficient mice are more prone to the de- precise effector mechanisms for regulation of velopment of colonic neoplasms after the induc- the microflora remain unknown. In addition to tion of chronic inflammation [6] . This effect was NLRP6, other inflammasomes that are expressed independent of inflammation or the composition mainly within the hematopoietic compartment, of the intestinal microflora, and was suggested to such as the NLRP3 inflammasome, may also involve enhanced proliferation and impaired function by regulating the microflora as well as apoptosis of colonic epithelial cells [17] . Here the autoinflammatory process itself. Thus, dif- again, results from the different reports suggest ferent inflammasomes may cooperate to main- that different inflammasomes at different cellu- tain tolerance towards commensal microbes and lar compartments may cooperate to produce net to initiate a potent immune response towards regulatory effects on mucosal inflammation and pathogens. Distinct inflammasomes expressed on the resultant tendency for tumor formation. in different cell lineages may orchestrate these seemingly opposite functions during acute mu- cosal inflammation. Inflammasomes as Regulators of Metabolic Disorders

The Inflammasome and Carcinogenesis Results obtained from our study enabled the study of another intriguing and largely unknown Chronic intestinal autoinflammation is a signifi- phenomena – effects of the microbiome on seem- cant predisposing factor for the formation of in- ingly unrelated distal processes such as systemic flammation-induced tumorigenesis. This is espe- inflammation and metabolic events. The meta- cially evident in human IBD, in which long- bolic syndrome is associated with multiorgan in- standing inflammation leads to dramatically flammatory abnormalities in organs such as liv- increased risk for a diffuse and aggressive form of er, adipose tissue and the pancreas, and repre- colorectal carcinoma. Inflammasomes have been sents a major disease burden in the developed suggested to participate in inflammation-in- world. Of the different manifestations of the met- duced tumorigenesis. Two recent reports suggest abolic syndrome, non-alcoholic fatty liver dis- that NLRP6 deficiency may be associated with ease (NAFLD) represents a major hepatic pathol-

50 Elinav Henao-Mejia Flavell

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 48–52 (DOI: 10.1159/000346510) ogy in western populations, with up to 100% of secretion was recently associated with increased morbidly obese individuals developing fatty ac- size of atherosclerotic lesions [20] . In addition, cumulation within hepatocytes [18] . Important- caspase 1-deficient precursors differentiate more ly, 25% of individuals with NAFLD develop a efficiently into mature adipocytes and have an progressive inflammatory form of liver disease increased oxidation rate, suggesting that inflam- termed non-alcoholic steatohepatitis (NASH) masomes might regulate developmental pro- that is associated with liver cirrhosis, hepatocel- grams in adipocytes and directly control cellular lular carcinoma, and increased mortality [19] . energy metabolism pathways [26] . Factors mediating progression from NAFLD to We recently reported that the inflammasome- NASH remain largely unknown. deficient aberrant microbiota has profound effects Inflammasomes have been recently suggested on the progression from NALFD to NASH [27]. In to participate in different physiopathological three mouse models of NASH, we demonstrated processes of the metabolic syndrome. IL-1␤ has that NLRP3 and NLRP6 inflammasome-deficient been shown to play a negative role in the patho- mice develop more severe inflammatory manifes- genesis of disorders associated with metabolic tation of NASH, and that this exacerbation is fully abnormalities such as ␤-cell death and athero- transferable to cohoused WT mice upon pro- sclerosis [20, 21]. In contrast, IL-18-deficient mice longed cohousing. The enhanced liver inflamma- are prone to developing obesity, hyperphagia and tion is induced by CCL5-dependent local gut in- insulin resistance [22] . The opposite roles of these flammation resulting in efflux of Toll-like recep- proinflammatory cytokines may again suggest tor (TLR) agonists into the portal circulation. that inflammasomes are sensing and responding Activation of TLR signaling pathways in the liver to the metabolic status of an organism through in susceptible mice leads to massive secretion of cooperative effects of different NLRs in different TNF- ␣ which promotes NASH progression. In tissues and cell types. Indeed, in adipose tissue, addition to demonstrating a direct causative link increased levels of saturated fatty acids such as between alterations of the microbiota composi- palmitate and ceramides decrease the activity of tion and the metabolic syndrome, our results sug- AMPK, leading to defective autophagy of mito- gest a unique cooperative activity of two major chondria and therefore to accumulation of ROS sensing systems, namely NLRs and TLRs in the in the cytosol, which in turn promotes NLRP3 pathogenesis of a complex multifactorial disease. activation and IL-1␤ secretion [23, 24] . Elevated In our model, these results suggest that genetic levels of IL-1 ␤ promote insulin resistance and an (inflammasome deficiency) and environmental effector Th1 proinflammatory cellular infiltrate factors (dietary and microbiota changes) cooper- in adipose tissue. In the pancreas, hyperglyce- ate to enhance a metabolic-inflammatory disor- mia-induced islet amyloid polypeptide and ROS der. Furthermore, we identify a bacterial family accumulation activates NLRP3 inflammasome (Porphyromonadaceae) that is significantly ex- and promotes IL-1␤ release from macrophages panded as a result of these combined host-envi- and ␤ -cells, leading to ␤-cell dysfunction and ronmental changes, e.g. inflammasome-deficient death [25]. Consistent with this, Nlrp3–/– mice mice fed a diet that induces either NASH (methio- exhibit improved glucose tolerance, insulin sen- nine choline deficient diet) or the metabolic syn- sitivity, and seem to be protected from HFD-in- drome (high-fat diet) [27] . The mechanisms by duced obesity. Finally, activation of the NLRP3 which this bacterial family or others induce meta- inflammasome by cholesterol crystals and IL-1␤ bolic changes remain to be explored.

Inflammasomes and Mucosal Immune Response 51

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 48–52 (DOI: 10.1159/000346510) References

1 Elinav E, Strowig T, Henao-Mejia J, 11 Allen IC, TeKippe EM, Woodford RM, 19 Caldwell SH, Oelsner DH, Iezzoni JC, Flavell RA: Regulation of the antimi- Uronis JM, Holl EK, Rogers AB, Her- Hespenheide EE, Battle EH, Driscoll CJ, crobial response by NLR proteins. Im- farth HH, Jobin C, Ting JP l: The et al: Cryptogenic cirrhosis: clinical

munity 2011; 34: 665–679. NLRP3 inflammasome functions as a characterization and risk factors for

2 Strowig T, Henao-Mejia J, Elinav E, negative regulator of tumorigenesis underlying disease. Hepatology 1999;

Flavell R: Inflammasomes in health during colitis-associated cancer. J Exp 29: 664–669.

and disease. Nature 2012; 481: 278–286. Med 2010; 207: 1045–1056. 20 Duewell P, Kono H, Rayner KJ, Sirois 3 Gross O, Yazdi AS, Thomas CJ, Masin 12 Hirota SA, Ng J, Lueng A, Khajah M, et CM, et al: NLRP3 inflammasomes are M, Heinz LX, Guarda G, Quadroni M, al: NLRP3 inflammasome plays a key required for atherogenesis and activat-

Drexler SK, Tschopp J: Inflammasome role in the regulation of intestinal ho- ed by cholesterol crystals. Nature 2010;

activators induce interleukin-1alpha meostasis. Inflamm Bowel Dis 2011; 17: 464: 1357–1361. secretion via distinct pathways with 1359–1372. 21 Larsen CM, Faulenbach M, Vaag A, differential requirement for the prote- 13 Zaki MH, Boyd KL, Vogel P, Kastan Vølund A, Ehses JA, Seifert B, Man- ase function of caspase-1. Immunity MB, Lamkanfi M, Kanneganti TD: The drup-Poulsen T, Donath MY: Interleu-

2012; 36: 388–400. NLRP3 inflammasome protects against kin-1-receptor antagonist in type 2 dia-

4 Jin T, Perry A, Jiang J, Smith P, et al: loss of epithelial integrity and mortal- betes mellitus. N Engl J Med 2007; 356: Structures of the HIN domain:DNA ity during experimental colitis. Immu- 1517–1526.

complexes reveal ligand binding and nity 2010; 32: 379–391. 22 Netea MG, Joosten LA, Lewis E, Jensen activation mechanisms of the AIM2 14 Dupaul-Chicoine J, Yeretssian G, DR, et al: Deficiency of interleukin-18 inflammasome and IFI16 receptor. Doiron K, Bergstrom KS, et al: Control in mice leads to hyperphagia, obesity

Immunity 2012; 36: 561–571. of intestinal homeostasis, colitis, and and insulin resistance. Nat Med 2006;

5 Elinav E, Strowig T, Kau AL, Henao- colitis-associated colorectal cancer by 12: 650–656. Mejia J, et al: NLRP6 inflammasome the inflammatory caspases. Immunity 23 Vandanmagsar B, Youm YH, Ravussin

regulates colonic microbial ecology and 2010; 32: 367–378. A, Galgani JE, Stadler K, Mynatt RL,

risk for colitis. Cell 2011; 145: 745–757. 15 Normand S, Delanoye-Crespin A, Bres- Ravussin E, Stephens JM, Dixit VD: 6 Hu B, Elinav E, Huber S, Booth CJ, senot A, Huot L, et al: Nod-like receptor The NLRP3 inflammasome instigates Strowig T, Jin C, Eisenbarth SC, Flavell pyrin domain-containing protein 6 obesity-induced inflammation and

RA: Inflammation-induced tumorigen- (NLRP6) controls epithelial self-renew- insulin resistance. Nat Med 2011; 17: esis in the colon is regulated by cas- al and colorectal carcinogenesis upon 179–188.

pase-1 and NLRC4. Proc Natl Acad Sci injury. Proc Natl Acad Sci USA 2011; 24 Wen H, Gris D, Lei Y, Jha S, Zhang L,

USA 2010; 107: 21635–21640. 108: 9601–9606. Huang MT, Brickey WJ, Ting JP: Fatty 7 Henao-Mejia J, Elinav E, Jin C, Hao L, 16 Chen GY, Liu M, Wang F, Bertin J, acid-induced NLRP3-ASC inflamma- et al: Inflammasome-mediated dysbio- Núñez G: A functional role for Nlrp6 in some activation interferes with insulin

sis regulates progression of NAFLD and intestinal inflammation and tumori- signaling. Nat Immunol 2011; 12: 408–415.

obesity. Nature 2012; 482: 179–185. genesis. J Immunol 2011; 186: 7187– 25 Masters SL, Dunne A, Subramanian SL, 8 Maloy KJ, Powrie F: Intestinal homeosta- 7194. Hull RL, et al: Activation of the NLRP3 sis and its breakdown in inflammatory 17 Hu B, Elinav E, Flavell RA: Inflamma- inflammasome by islet amyloid poly-

bowel disease. Nature 2011; 474: 298–306. some-mediated suppression of inflam- peptide provides a mechanism for en- 9 Siegmund B, Lehr HA, Fantuzzi G, Di- mation-induced colorectal cancer pro- hanced IL-1beta in type 2 diabetes. Nat

narello CA: IL-1 beta-converting en- gression is mediated by direct Immunol 2010; 11: 897–904. zyme (caspase-1) in intestinal inflam- regulation of epithelial cell prolifera- 26 Stienstra R, Joosten LA, Koenen T, van

mation. Proc Natl Acad Sci USA 2001; tion. Cell Cycle 2011; 10: 1936–1939. Tits B, et al: The inflammasome-medi-

98: 13249–13254. 18 Sheth SG, Gordon FD, Chopra S: Nonal- ated caspase-1 activation controls adi- 10 Bauer C, Duewell P, Mayer C, Lehr HA, coholic steatohepatitis. Ann Intern pocyte differentiation and insulin sen-

Fitzgerald KA, Dauer M, Tschopp J, En- Med 1997; 126: 137–145. sitivity. Cell Metab 2010; 12: 593–605. dres S, Latz E, Schnurr M: Colitis in- 27 Henao-Mejia J, Elinav E, Jin C, Hao L, duced in mice with dextran sulfate so- et al: Inflammasome-mediated dysbio- dium (DSS) is mediated by the NLRP3 sis regulates progression of NAFLD and

inflammasome. Gut 2010; 59: 1192–1199. obesity. Nature 2012; 482: 179–185.

Richard Flavell Yale School of Medicine, The Anlyan Center Department of Immunology 300 Ceda Street, Suite 569A New Haven, CT 06519 (USA)

E-Mail richard.flavell @ yale.edu

52 Elinav Henao-Mejia Flavell

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 48–52 (DOI: 10.1159/000346510) Chapter 3: Innate Immunity and Intestinal Microbiota

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532)

Microbial Endocrinology: An Evolution-Based Shared Mechanism Determining Microbiota’s Influence on Health and Disease

Mark Lyte

Texas Tech University Health Sciences Center, Abilene, Tex. , USA

A b s t r a c t convergence of seemingly disparate fields, microbiolo- The ability of bacteria to both produce and respond to gy, endocrinology and neurophysiology, is the emerging the panoply of neuroendocrine hormones that are more translational medicine discipline known as microbial en- commonly associated with mammalian organisms is be- docrinology. Given that immune cell function can be coming increasingly recognized as playing a pivotal role modulated by host neuroendocrine hormones, one of in both disease pathogenesis as well as the maintenance the areas in which microbial endocrinology has been of homeostasis. The mammalian host consists of 1 90% proposed to have particular application is that of probi- prokaryotic microorganisms with the scientific and clini- otics in which selection of a probiotic based on its neu- cal perception, and understanding, of host-microorgan- rochemical profile to specifically target host immuno- ism interaction mainly from the perspective of a ‘dumb logical processes may provide for a rationale design in bug’, in other words an organism which simply replicates the selection of probiotics for both inflammatory-medi- by binary fission but otherwise has limited interaction ated disease and homeostasis. with its host. The range of interactions between host and Copyright © 2013 S. Karger AG, Basel microorganism are widely recognized to have both pos- itive and negative consequences for the host, but rarely are these interactions considered to have as a common The range of neuroendocrine hormones that mechanism a shared use and recognition of neuroen- have been described in microorganisms is ex- docrine hormones. The recognition of such a common tremely large and include corticotropin [1] , soma- pathway carries important implications to our under- tostatin [2] and progesterone [3] to name but a standing of how the vast majority of microorganisms few. A recent comprehensive analysis by Roshchi- both in our gut and in the general environment influence na [4] of the wide spectrum of neurohormones both the pathogenesis of disease and homeostasis. This and related cognate receptors that have been iso- lated from microorganisms highlights the pres- rect stimulation of microbial growth by stress- ence in microorganisms of what are otherwise related neurochemicals in both in vitro as well thought to be more commonly associated with as in vivo systems [10–21] . However, it should be mammalian systems [4] . In general, the precise noted that not all biogenic amines have similar role of these neuroendocrine hormones in bacte- effects on bacterial growth. For example, nor- rial physiology is largely unknown. The diverse epinephrine and dopamine, but not the indole- nature of these neurohormones strongly suggests amine serotonin, increased the growth of a that from an evolutionary perspective the posses- number of Gram-negative enteric pathogens [7, sion of what are normally considered to be spe- 22] . cific to vertebrates implies that microorganisms The first mechanism that was identified to ac- have a means to recognize neurohormones with- count for these experimental observations was in a vertebrate host and initiate changes in physi- iron mediated. The ability of catecholamines to ology that would prove advantageous to its sur- act in a siderophore-like capacity to remove iron vival. This ability of bacteria to recognize neuro- from the iron-sequestering glycoproteins trans- endocrine hormones represents what has been ferrin and lactoferrin and enable its uptake by termed interkingdom signaling by a number of bacteria was shown for a number of bacterial spe- investigators. cies [11] . More recently, results from other re- To date, one of the most potent neurophysi- searchers have greatly expanded the mechanisms ological events that have been shown to influ- by which catecholamines, as the prototypical ence the outcome of infectious disease episodes bacterial neuroendocrine hormone, may affect is that of stress. Although the concept of stress is microbial physiology. One of the more recent the most ambiguous in biology, for the purposes studies that have now extended the role of cate- of this paper it is defined as a physical event that cholamines beyond just growth has been that of results in the release of a number of neuroendo- Peterson et al. [23] who have demonstrated the crine hormones, most specifically those belong- ability of catecholamines to increase the rate of ing to the catecholamine family. Immediately conjugative gene transfer between enteric bac- following a physical stressor, such as a severe in- teria. Mechanistically, Peterson et al. [23] also jury as that encountered in a motor vehicle ac- showed that the use of an adrenergic antagonist cident, the catecholamines norepinephrine and blocked the ability of norepinephrine to increase epinephrine are released. These catecholamines the horizontal gene transfer efficiency of a conju- are also known as ‘flight or fight’ hormones. In gative plasmid from one strain to another. This vitro experiments with pathogenic bacteria such study strongly suggests that the neurohormonal as Escherichia coli have shown that micromolar environment within the gut, such as that encoun- exposure to norepinephrine or epinephrine re- tered in the elaboration of norepinephrine, may sults in explosive growth that is several log or- contribute to the exchange of genetic material be- ders higher than controls [5–8] . Further, in- tween various bacterial species, and may there- creases in growth are also accompanied by an fore represent not only a mechanism by which increase in elaboration of virulence factors such antibiotic resistance could be increased, but more as toxins [6] and the upregulation of attachment importantly a neurophysiological-based mecha- factors such as adhesion which enable infectious nism influencing the adaptation of bacteria to the microorganisms such as enterotoxigenic bacte- host. ria to attach onto the intestinal mucosa [9] . More Crucial to an understanding of the role that recent studies by a number of groups have con- bacterial production and recognition of verte- firmed and extended these early reports of di- brate neuroendocrine hormones has as regards

54 Lyte

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532) the pathophysiology of infectious disease as well probiotics and their well-recognized ability to as the maintenance of normal homeostasis is produce a neuroendocrine hormone to more ful- that the gastrointestinal tract is a rich source of ly discuss the concept of microbial endocrinology neuroendocrine hormones. Results from numer- and how it applies to health and infectious dis- ous laboratories utilizing rodent, porcine and ease progression. human in vivo models have shown that large Recently, I have proposed that probiotics can amounts of neuroendocrine hormones extend- act as drug-like delivery vehicles due to their ing from dopamine to norepinephrine to sero- production of neuroactive compounds such as tonin are found in the luminal space [24–26] . In the biogenic amine ␥-amino butyric acid (GABA) fact, the stomach has even been shown to be a [29]. This hypothesis represents a novel, innova- dopaminergic organ producing a significant tive interdisciplinary approach that challenges proportion of the dopamine found in the body the current understanding of how probiotics [27] . Considering, then, that within the microen- function in ameliorating gastrointestinal-relat- vironment of the tissue adequate levels of neuro- ed disease, especially those with an inflammato- hormones may be available to an infectious mi- ry component. Currently, the mechanism(s) by croorganism upon entrance into the host, could which probiotics are believed to function within the pathogenesis of the resulting infection some- the gastrointestinal tract is due to the direct how related to the neurophysiological response modulation of immune cell activity by either of the host [28] ? Can the infectious agent active- direct bacterial-immune cell interaction (where ly utilize the hormonal products of the host’s the probiotic can either be alive or dead) or by a neurophysiological response to stress, such as non-neurological product produced by the pro- the elaboration of norepinephrine, to its own ad- biotic. vantage? Critically, a microbial endocrinology-based While little is known why these hormones are hypothesis also provides for a unifying model found in the luminal space, whether produced by that can guide the selection of probiotics based the host or the bacteria themselves, the environ- on a matching of the specific probiotic organ- ment within the intestinal tract represents a re- ism’s capacity to produce a specific neurochemi- gion where there is ample opportunity for these cal and the disease condition that needs to be secreted bacterial and host neuroendocrine hor- treated. For example, treatment of inflammato- mones to affect the physiology of both bacteria ry bowel disease with probiotic bacteria that and host as will be discussed in the following sec- produce an anti-inflammatory neurotransmit- tion. ter/immune modulator may be proposed since many of the cellular-based mechanisms which participate in the inflammatory condition are Discussion amenable to intervention with an inhibitory neurochemical such as GABA. As mentioned As discussed above, the experimental results above with regard to the general role of micro- which have demonstrated that bacteria are able to bial endocrinology in disease and homeostasis, both produce and recognize what are otherwise it is expressly not the intent of a microbial endo- regarded as vertebrate neuroendocrine hormones crinology-based action of probiotics to obviate a implies that interkingdom signaling holds the role for direct modulation of immune respon- potential to affect infectious disease processes as siveness within the gut by probiotics which has well as homeostasis. In order to more completely been well documented by numerous studies. In- expand on this concept, I will use the example of deed, the model proposes a mechanism by which

Microbial Endocrinology 55

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532) probiotic bacteria modulate immune respon- uals. Testing of a neurochemical-based hypoth- siveness [29] . esis of probiotic action would allow the identifi- GABA is the predominant inhibitory neu- cation of not only a novel mechanism by which rotransmitter in the nervous system, and exerts probiotics function in the host, but also the mo- anti-inflammatory actions in the immune sys- lecular and cellular targets for probiotic-pro- tem as well. Because GABA can inhibit proin- duced neurochemicals. flammatory peptide release from viscerosenso- In support of this hypothesis has been the re- ry neurons, and inhibits activity of inflamma- cent publication by Thomas et al. [18] who have tory immune cells, the production of GABA by shown that the ability of the probiotic Lactobacil- gut probiotic bacteria could constitute a power- lus reuteri ATCC PTA 6475 to inhibit TNF-me- ful mechanism for prophylaxis of gastrointesti- diated aspects of host mucosal immunity is due nal inflammatory conditions. It should be noted to the production by L. reuteri of histamine. that GABA receptors have been localized on When histamine-containing fractions produced proinflammatory immune cells, which function during in vitro growth of L. reuteri were added to to downregulate inflammatory responses such Toll-like receptor 2-activated human monocyt- as cytokine release [30]. Thus, production of oid cells, the production of TNF was suppressed GABA by probiotic bacteria could reduce in- through the activation of the H(2) receptor [18] . flammation in colitis via several different mech- While the authors concluded that production of anisms. such bioactive metabolites like the neurochemi- That a microorganism, such as a probiotic cal histamine offers a new anti-inflammatory bacterium, should be able to produce a neuro- strategy with which to treat chronic immune-re- chemical that is exactly the same as found in lated diseases, it still must be shown that probiot- mammalian systems may seem surprising; how- ics will in fact produce the desired neurochemi- ever, what is commonly considered to be ex- cal in vivo. This aspect, which is integral to the clusively vertebrate neurotransmitters, neuro- proof that probiotics can serve as drug delivery hormones, and related receptors, are in fact, vehicles, constitutes only the first step in a step- as already discussed above, widely dispersed wise sequential design to evaluate the ability of throughout nature from bacteria to plants to in- neurochemical-producing probiotics to influ- sects [10] . In fact, contamination of a distilled wa- ence disease pathobiology that has been previ- ter apparatus led to the discovery that bacteria ously discussed [29] . possess a high-affinity receptor for GABA, and subsequently one of the first bioassays for GABA that was entirely based on bacteria as a GABA Conclusions biosensor [31] . Thus, neurochemicals such as GABA should be viewed as a common shared The ability of microorganisms to both produce language enabling interkingdom signaling be- and recognize through receptor-mediated pro- tween prokaryotes (e.g. probiotic bacteria) and cesses the exact same neuroendocrine hormones eukaryotes (e.g. vertebrates). By proposing that that mammalian hosts (as well as plants and in- GABA, or other neurochemical, production by sects) produce has been known for a number of probiotic organisms constitutes a hitherto un- decades [10] . Critically, at no time was it proposed known mechanism by which probiotics act in the that such a common evolutionary-shared mecha- host, this hypothesis represents a paradigm- nism could possibly play a role in both disease shifting approach to elucidating the mechanisms and homeostasis. This convergence of seeming- by which a natural product functions in individ- ly disparate fields, microbiology, endocrinology

56 Lyte

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532) and neurophysiology, is the emerging evolution- Acknowledgements based translational medicine discipline known as microbial endocrinology which is increasingly I would like to thank all my students and collaborators who over the years I have had the pleasure to work with. finding application to understanding both the pathophysiology of infectious disease as well as maintenance of homeostasis.

References

1 LeRoith D, Liotta AS, Roth J, Shiloach J, 9 Lyte M, Erickson AK, Arulanandam 16 Burton CL, Chhabra SR, Swift S, Bald- Lewis ME, Pert CB, Krieger D: Cortico- BP, Frank CD, Crawford MA, Francis win TJ, Withers H, Hill SJ, Williams P: tropin and ␤ -endorphin-like materials DH: Norepinephrine-induced expres- The growth response of Escherichia are native to unicellular organisms. sion of the K99 pilus adhesin of entero- coli to neurotransmitters and related

Proc Natl Acad Sci USA 1982; 79: 2086– toxigenic Escherichia coli . Biochem catecholamine drugs requires a func-

2090. Biophys Res Commun 1997; 232: 682– tional enterobactin biosynthesis and

2 LeRoith D, Pickens W, Vinik AI, Shi- 686. uptake system. Infect Immun 2002; 70: loach J: Bacillus subtilis contains mul- 10 Lyte M: Microbial endocrinology and 5913–5923. tiple forms of somatostatin-like mate- infectious disease in the 21st century. 17 Bearson BL, Bearson SM: The role of

rial. Biochem Biophys Res Commun Trends Microbiol 2004; 12: 14–20. the QseC quorum-sensing sensor ki-

1985; 127: 713–719. 11 Freestone PP, Lyte M, Neal CP, Maggs nase in colonization and norepineph- 3 Schar G, Stover EP, Clemons KV, Feld- AF, Haigh RD, Williams PH: The mam- rine-enhanced motility of Salmonella man D, Stevens DA: Progesterone bind- malian neuroendocrine hormone nor- enterica serovar Typhimurium. Mi-

ing and inhibition of growth in Tricho- epinephrine supplies iron for bacterial crob Pathog 2008; 44: 271–278. phyton mentagrophytes. Infect growth in the presence of transferrin or 18 Thomas CM, Hong T, van Pijkeren JP,

Immun 1986; 52: 763–767. lactoferrin. J Bacteriol 2000; 182: 6091– Hemarajata P, Trinh DV, Hu W, Britton 4 Roshchina VV: Evolutionary consider- 6098. RA, Kalkum M, Versalovic J: Hista- ations of neurotransmitters in micro- 12 Sperandio V, Torres AG, Jarvis B, Na- mine derived from probiotic Lactoba- bial, plant, and animal cells; in Lyte M, taro JP, Kaper JB: Bacteria-host com- cillus reuteri suppresses TNF via mod- Freestone PPE (eds): Microbial Endo- munication: the language of hormones. ulation of PKA and ERK signaling.

crinology: Interkingdom Signaling in Proc Natl Acad Sci USA 2003; 100: 8951– PLoS One 2012; 7:e31951. Infectious Disease and Health. New 8956. 19 Vlisidou I, Lyte M, van Diemen PM, York, Springer, 2010, pp 17–52. 13 Reissbrodt R, Rienaecker I, Romanova Hawes P, Monaghan P, Wallis TS, Ste- 5 Lyte M: The microbial organ in the gut JM, Freestone PP, Haigh RD, Lyte M, vens MP: The neuroendocrine stress as a driver of homeostasis and disease. Tschape H, Williams PH: Resuscitation hormone norepinephrine augments

Med Hypotheses 2010; 74: 634–638. of Salmonella enterica serovar ty- Escherichia coli O157:H7-induced en- 6 Lyte M, Arulanandam BP, Frank CD: phimurium and enterohemorrhagic teritis and adherence in a bovine ligat- Production of Shiga-like toxins by Escherichia coli from the viable but ed ileal loop model of infection. Infect

Escherichia coli O157: H7 can be influ- nonculturable state by heat-stable en- Immun 2004; 72: 5446–5451. enced by the neuroendocrine hormone terobacterial autoinducer. Appl Envi- 20 Pullinger GD, Carnell SC, Sharaff FF,

norepinephrine. J Lab Clin Med 1996; ron Microbiol 2002; 68: 4788–4794. van Diemen PM, Dziva F, Morgan E,

128: 392–398. 14 Rahman H, Reissbrodt R, Tschape H: Lyte M, Freestone PP, Stevens MP: Nor- 7 Lyte M, Ernst S: Catecholamine in- Effect of norepinephrine on growth of epinephrine augments Salmonella duced growth of gram negative bacte- Salmonella and its enterotoxin pro- enterica -induced enteritis in a manner

ria. Life Sci 1992; 50: 203–212. duction. Indian J Exp Biol 2000; 38: associated with increased net replica- 8 Lyte M, Freestone PP, Neal CP, Olson 285–286. tion but independent of the putative BA, Haigh RD, Bayston R, Williams 15 Kinney KS, Austin CE, Morton DS, adrenergic sensor kinases QseC and

PH: Stimulation of Staphylococcus Sonnenfeld G: Norepinephrine as a QseE. Infect Immun 2010; 78: 372–380. epidermidis growth and biofilm for- growth stimulating factor in bacteria- 21 Pullinger GD, van Diemen PM, Carnell

mation by catecholamine inotropes. mechanistic studies. Life Sci 2000; 67: SC, Davies H, Lyte M, Stevens MP:

Lancet 2003; 361: 130–135. 3075–3085. 6-hydroxydopamine-mediated release of norepinephrine increases faecal ex- cretion of Salmonella enterica serovar

Typhimurium in pigs. Vet Res 2010; 41: 68.

Microbial Endocrinology 57

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532) 22 Lyte M: Induction of gram-negative 25 Ruckebusch Y, Meirieu O, von Ritter C, 29 Lyte M: Probiotics function mechanis- bacterial growth by neurochemical Blum AL: Measurement of mono- tically as delivery vehicles for neuroac- containing banana (Musa x paradi- amines and their metabolites in the tive compounds: microbial endocrinol- siaca) extracts. FEMS Microbiol Lett interstitial fluid of the gut. J Pharmacol ogy in the design and use of probiotics.

1997; 154: 245–250. Methods 1985; 13: 167–177. Bioessays 2011; 33: 574–581. 23 Peterson G, Kumar A, Gart E, Naray- 26 Eldrup E, Richter EA: DOPA, dopa- 30 Bjurstom H, Wang J, Ericsson I, anan S: Catecholamines increase conju- mine, and DOPAC concentrations in Bengtsson M, Liu Y, Kumar-Mendu S, gative gene transfer between enteric the rat gastrointestinal tract decrease Issazadeh-Navikas S, Birnir B: GABA, a

bacteria. Microb Pathog 2011; 51: 1–8. during fasting. Am J Physiol Endocri- natural immunomodulator of T lym-

24 Meirieu O, Pairet M, Sutra J, Rucke- nol Metab 2000; 279:E815–E822. phocytes. J Neuroimmunol 2008; 205: busch M: Local release of monoamines 27 Mezey E, Eisenhofer G, Hansson S, 44–50. in the gastrointestinal tract: an in vivo Hunyady B, Hoffman BJ: Dopamine 31 Balcar VJ: Presence of a highly efficient

study in rabbits. Life Sci 1986; 38: 827– produced by the stomach may act as a ‘binding’ to bacterial contamination 834. paracrine/autocrine hormone in the can distort data from binding studies.

rat. Neuroendocrinology 1998; 67: 336– Neurochem Res 1990; 15: 1237–1238. 348. 28 Lyte M: The role of microbial endocri- nology in infectious disease. J Endocri-

nol 1993; 137: 343–345.

Mark Lyte, PhD, MS, MT (ASCP) Texas Tech University Health Sciences Center 1718 Pine Street Abilene, TX 79601 (USA)

E-Mail mark.lyte @ ttuhsc.edu

58 Lyte

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 53–58 (DOI: 10.1159/000346532) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504)

Standing Guard: Innate Lymphoid Cells Promote Immunity and Tissue Homeostasis at Barrier Surfaces

a a, b Laurel A. Monticelli David Artis a Department of Microbiology and Institute for Immunology, Perelman School of Medicine, and b Department of Pathobiology, School of Veterinary Medicine, University of Pennsylvania, Philadelphia, Pa. , USA

A b s t r a c t populations share morphological, developmental Maintenance of epithelial barrier function and restora- and functional similarities with CD4+ T helper tion of tissue homeostasis after injury are essential to cells, but lack antigen receptors. Specifically, limit exposure to commensal and pathogenic microbes members of the ILC family lack expression of cell at barrier sites such as the lung, skin and intestine. Innate surface makers associated with granulocytes, lymphoid cells (ILCs) are a recently recognized family of dendritic cells, macrophages and conventional B innate immune cells that perform critical roles in anti- and T lymphocytes but can be positively identified pathogen immunity, regulation of inflammation and by expression of CD90 (Thy1), CD25 (IL-2R ␣ ), promotion of wound healing and tissue repair at barrier CD127 (IL-7R ␣) and c-Kit [1–3] . surfaces. In this review, we discuss recent advances in the While the cell lineage relationships between understanding of how ILC populations in the intestine heterogeneous ILC populations are still being and respiratory tract impact immunity, inflammation elucidated, all ILCs are thought to develop from and tissue homeostasis. a common bone marrow-derived lymphoid pre- Copyright © 2013 S. Karger AG, Basel cursor that is dependent on the transcription factor inhibitor of DNA-binding 2 (Id2) [1] (fig. 1 ). Heterogeneity in cell surface marker ex- Recent studies have revealed the existence of a pression, transcriptional regulation and effector previously unrecognized population of immune molecule expression allow ILCs to be broadly cells that appear to represent the innate counter- classified into 3 main populations referred to as part to the T cell lineage of the adaptive immune group 1, group 2 or group 3 ILCs [1] (fig. 1 ). The system. Located at multiple anatomical sites, these group 1 ILC population is composed of classical heterogeneous populations of innate lymphoid NK cells and may include other ILC populations cells (ILCs) serve as crucial regulators of immu- that can express IFN␥ [4] (fig. 1 ). In contrast, ILC nity and inflammation at barrier surfaces such as populations found in group 2 are composed of the skin, intestine and respiratory tract [1–3]. ILC cells that express the T helper type 2-associated Id2+ ILC precursor(s)

IL-7 T-bet IL-2 ROR␣ IL-2 ROR␥t IL-15 IL-7 GATA3? IL-7 Fig. 1. The ILC family is composed of three main groups that all derive from Id2+ ILC precursors. Group 1 ILCs (or- ange) include classical NK cells and other ILCs that re- Group 1 Group 2 Group 3 quire T-bet, IL-7 and IL-15 for development and respond LTi- ␥ NH like to IL-12 to produce IFN . Group 2 ILCs (green) are com- Nuo Ih2 NK ILC ILC22 ILC17 prised of nuocytes (Nuo), natural helper (NH) cells, in- Lung NCR- ILC 22 nate type 2 helper (Ih2) cells and lung-resident ILCs that ? express GATA3 and depend upon ROR ␣ for develop- IL-25 TSLP? IL-23 AhR ment. Upon IL-25, IL-33 and/or thymic stromal lympho- IL-12 IL-33 IL-1␤ ligands? poietin (TSLP) stimulation, group 2 ILCs can produce IL-5, IL-13, IL-9 and amphiregulin (Areg). Group 3 ILCs (blue) IL-5 IL-13 consist of a heterogeneous population of ROR ␥ t-de- ␥ pendent ILCs that respond to IL-23, IL-1 ␤ and AhR ligands IFN IL-9 Areg IL-17A IL-22 to express IL-22 and/or IL-17A.

cytokines IL-5, IL-9 and IL-13 as well as the populations found within group 3 include lym- growth factor amphiregulin [3, 5, 6]. These cell phoid tissue inducer (LTi)-like cells, ILC17, populations, which were originally termed natu- ILC22 and NK cell receptor-expressing NCR-22 ral helper cells, nuocytes or innate helper type 2 cells [12–16] ( fig. 1 ). These phenotypically dispa- cells, also express the Th2 cell-associated tran- rate ILC populations are linked by their com- scription factor GATA3 [7–9], although the po- mon dependence on ROR␥ t for development and tential influence of GATA3 on ILC development their production of the Th17 cell-associated cy- or activation is currently unknown (fig. 1 ). tokines IL-17A and IL-22 [1] . Group 2 ILCs are activated by the epithelial-de- The factors regulating development and func- rived cytokines IL-25 and IL-33, but whether ad- tion of classical NK cells have been extensively ditional epithelial cell-derived factors, such as reviewed elsewhere [4, 17]. Therefore, this review thymic stromal lymphopoietin, can influence will focus on discussing recent studies that high- ILC proliferation or effector function has not light newly defined roles for group 2 and group 3 been fully addressed (fig. 1 ). Interestingly, IL-25 ILCs in regulating immunity, inflammation and also elicits an innate multi-potent progenitor tissue homeostasis at barrier sites. population (termed MPP type2 cells) that gives rise to myeloid lineage cells that contribute to type 2 anti-helminth immunity [10], but this Identification of Group 2 Innate Lymphoid progenitor population appears to be distinct Cells from IL-25-elicited group 2 ILCs. The develop- ment of group 2 ILCs is dependent on expression Expression of the Th2-cell-associated cytokines of the transcriptional regulator ROR␣ , but is in- IL-5 and IL-13 is a classic hallmark of both patho- dependent of ROR ␥t [1, 7, 11] . In contrast, ILC logic allergic inflammation and host-protective

60 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) anti-parasitic immune responses at mucosal sites files of ILCs, raising the possibility that some [3, 18–20] . Group 2 ILCs provide a critical, early ILCs (or their less differentiated progenitors) may innate source of these cytokines before the onset circulate throughout the bloodstream or lymph, of adaptive CD4+ Th2 cell-dependent immune migrating into different anatomical sites in re- responses. ROR␥ t-independent ILCs were first sponse to infectious or inflammatory cues, while described by Moro et al. [7] as natural helper cells other ILCs may remain tissue-resident sentinels located in fat-associated lymphoid clusters sur- at barrier sites such as the lung, intestine and rounding the gut mesentery that provide aid to skin. Further studies are needed to determine B1 cells. Subsequent studies by two independent whether these anatomically disparate popula- laboratories further identified group 2 ILC popu- tions belong to one or more discrete ILC subsets. lations at multiple tissue sites (termed nuocytes or innate helper type 2 cells) that were critical for optimal protective immunity against intestinal Identification of Innate Lymphoid Cells in the helminth parasites [8, 9] . These group 2 ILCs ex- Lungs of Mice and Humans press the receptors for the epithelial cell-derived cytokines IL-25 (IL-17Rb) and IL-33 (T1/ST2) A number of recent studies have identified a previ- and produce IL-5 and IL-13 upon stimulation ously unrecognized population of ILC cells in with IL-25 and/or IL-33 in combination with IL-2 murine lung parenchyma [21–28] that are pheno- and IL-7 [6, 8, 9] . In parallel with these studies, a typically analogous to the group 2 ILC popula- separate report identified a population of IL- tions initially reported in the gut-associated lym- 25-elicited, multi-potent progenitor type 2 cells phoid tissue, fat-associated lymphoid clusters and (MPPtype2 cells) that also augment anti-helminth spleen [7–9]. The lung contains a constitutive pop- immunity, but are functionally distinct from ulation of Id2-dependent lineage-negative (Lin–) ILCs by their ability to differentiate into myeloid cells that express CD90, c-Kit, CD127 (IL-7R␣ ), cells [10]. The potential lineage relationship, if CD25 (IL-2R ␣), CD44, ICOS and T1/ST2 (IL- any, between ILCs and MPP type2 cells has yet to 33R) but lack expression of NK cell markers and be examined. ROR␥ t, a phenotype consistent with group 2 ILCs Although the biological distinction between [22]. Lung-resident ILCs are responsive to IL-33 natural helper cells, nuocytes and innate helper stimulation and produce IL-5 and IL-13, but these type 2 cells remains controversial, the striking cells do not express large amounts of IL-22, IL-17A similarities in cell surface phenotype, develop- or IFN␥ [21, 22], indicating that the ILC popula- mental requirements and effector cytokine ex- tion in the lung is part of the larger group 2 ILC pression between the three populations suggests subset, distinct from ROR␥ t-dependent ILCs. that these cells may be part of a single subset that Recent work has identified a population of exists in multiple anatomical locations and ex- ILCs in human lung and intestine that is pheno- hibits some degree of functional heterogeneity. typically and functionally analogous to murine Transient differences in cell surface expression of lung-resident ILC [22, 29] . Examination of healthy activation markers (such as c-Kit or Sca-1) be- human lung tissue or bronchoalveolar lavage fluid tween these three cell populations could be the from lung transplant patients revealed a constitu- result of varying degrees of maturation/differen- tive population of Lin– cells expressing CD25, tiation governed by environmental cues in dis- CD127 and ST2, a cell surface phenotype shared tinct tissue microenvironments. Furthermore, by ILCs in murine lung parenchyma [22] . In a par- there is very little known about the trafficking allel study, Mjosberg et al. [29] identified ILC pop- patterns and chemokine receptor expression pro- ulations at multiple tissue sites, including fetal and

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 61

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) Airway repair Airway inflammation AHR

Allergen Fig. 2. Group 2 ILCs promote AHR and tissue repair. Con- Influenza tact with allergens, viruses or parasitic helminth worms Helminth causes inflammation of the airway epithelium, resulting in production of IL-33 and/or IL-25 (red arrow) that acti- vates group 2 ILCs to express a variety of effector cyto- IL-33 MMC Epithelial cell kines that can drive either pathologic AHR responses IL-25? and (blue arrow) or can promote beneficial tissue-protective IL-9 goblet cell ? proliferation responses to repair the damaged epithelium (green ar- ILC row). IL-5-mediated recruitment of eosinophils and IL- IL-5 13-mediated epithelial cell/goblet cell proliferation can coordinately drive AHR responses. IL-9 acts in an auto- crine manner to promote optimal IL-5/IL-13 expression Areg in ILCs and could potentially impact MMC responses. ILC-derived IL-13 may also influence differentiation of al- AAMac ternatively-activated macrophages (AAMac). Addition- IL-13 ally, ILC expression of amphiregulin (Areg) can promote Eosinophil ? epithelial cell proliferation to repair injured airway epi- recruitment +IL-4 thelia that have been damaged by virus infection.

adult intestine and lung that were characterized immunity, inflammation or tissue homeostasis by expression of CD161 and the chemoattractant in the lung. Multiple reports have emerged over receptor CRTH2. CRTH2+ ILCs isolated from the past year demonstrating that group 2 ILCs fetal intestine were responsive to IL-33 or IL-25 serve as crucial mediators of inflammatory air- to produce IL-13, similar to murine ILC popula- way disease. tions [29]. CRTH2+ ILCs were also detectable in Airway inflammation is commonly triggered human blood and displayed some level of plastic- by a wide variety of allergic stimuli, such as pol- ity not observed in tissue-resident ILC popula- len or animal dander, and is characterized by tions. Blood CRTH2+ ILCs could produce IL-22 high levels of IgE, mucus production and eosino- and IL-13 [29] , suggesting that the blood may con- philia [19, 30–32] . Additionally, a second type of tain a progenitor ILC population capable of differ- non-allergic inflammation can arise following entiating into either IL-22-producing ILCs or IL- respiratory viral infection [19, 30] , but the cellu- 13-producing ILCs depending upon the cytokine lar and molecular mechanisms underlying this signals present in the tissue microenvironment. type of airway hyperreactivity (AHR) response are poorly understood. In a recent study, Chang et al. [21] observed that ILCs in the respiratory Group 2 Innate Lymphoid Cells Regulate tract played a pathologic role in promoting AHR Airway Inflammation following viral infection. In this model, infection with influenza A virus subtype H3N1 induced The presence of a population of ILCs that are acute airway inflammation independently of constitutive in respiratory tissue raises questions adaptive immunity. Macrophage-derived IL-33 about their potential involvement in regulation of induced the activation and population expansion

62 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) of Lin– CD90+ ST2+ Sca-1+ ILCs following infec- disorders like COPD [34, 35]. However, the po- tion. Notably, development of AHR was strictly tential ability of ILC-derived IL-13 to impact dependent on the IL-33/IL-33R pathway and AAMac differentiation during allergic responses ILCs as genetic disruption of IL-33R-signaling or has not yet been examined ( fig. 2 ). antibody-mediated depletion of CD90+ group 2 IL-9 has been linked to a wide array of allergic ILCs was sufficient to provide protection from disorders, although the cellular sources and reg- H3N1-induced AHR [21] . Collectively, these da- ulation of IL-9 expression remains controversial ta implicate group 2 ILCs as critical players in [36, 37] . Using a model of papain-induced airway promoting virus-induced airway inflammation inflammation, Wilhelm et al. [38] recently re- ( fig. 2 ). IL-13 has been shown to promote epithe- ported that ILCs in the lung are a predominant lial cell proliferation in multiple settings of air- early source of IL-9. Although IL-9 production by way injury or inflammation [19, 30, 32, 33] , and ILCs was transient and dissipated soon after al- IL-13 expression by ILCs appeared to be essential lergen challenge, antibody-mediated blockade of for infection-induced AHR responses, as adop- IL-9 activity demonstrated a crucial role for auto- tive transfer of wild-type, but not IL-13-deficient crine IL-9 in promoting optimal IL-5 and IL-13 ILCs, was sufficient to drive AHR responses in expression ( fig. 2 ). ILC-intrinsic IL-9 expression IL-13-deficient hosts [21] . was strikingly dependent on IL-2 produced by T In addition to regulating virus-induced AHR, cells, revealing a previously unappreciated role further studies have now defined roles for ILCs in for adaptive immunity in regulation of ILC effec- promoting allergic inflammation in response to tor function. In addition to promoting optimal a wide variety of allergens. Elegant studies by Th2 cytokine expression, the authors also specu- Barlow et al. [24] demonstrated that intranasal late that IL-9 may be important in the survival or challenge with either OVA antigen, recombinant maintenance of ILC cells [36] . IL-9 can also elicit IL-25 or IL-33 protein resulted in the accumula- mucosal-associated mast cells (MMCs), a crucial tion of IL-13-expressing ILCs in the lung. Similar player in allergic responses [37, 39], but the poten- to the requirement for ILC-intrinsic IL-13 during tial influence of ILC-intrinsic IL-9 on MMCs is influenza virus-induced AHR, adoptive transfer unknown (fig. 2 ). Further studies are required to of IL-13-sufficient ILCs, but not IL-13-deficient define the differential contributions of IL-5, IL- ILCs, promoted AHR following IL-25 intranasal 13 and IL-9 in ILC-dependent regulation of aller- challenge. A similar dependence on ILC-derived gic and non-allergic virally induced airway in- IL-13 was also observed in an AHR model using flammation. NKT glycolipid antigen [23] . An additional study examining papain allergen-induced inflamma- tion implicated group 2 ILCs as essential media- Innate Lymphoid Cells Promote Airway tors of eosinophilic infiltration and mucus pro- Epithelial Repair duction in the airways [28] , providing further evidence that ILC-derived cytokines are central While it is becoming increasingly clear that ILCs to the promotion of airway inf lammation in mul- act as essential mediators in progression of tiple models ( fig. 2 ). Notably, expression of IL-13 chronic allergic airway inflammation, new evi- in cooperation with IL-4 drives alternative acti- dence suggests that group 2 ILCs can also serve vation of macrophages (termed AAMacs) that beneficial host-protective roles by promoting tis- have been implicated as important regulators of sue repair in injured respiratory tissue. Genome- lung tissue fibrosis in settings of allergic airway wide transcriptional profiling of naïve lung-res- inflammation and chronic pulmonary fibrosis ident ILCs revealed a transcriptional signature

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 63

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) strongly enriched in genes that regulate wound A Role for Lung Innate Lymphoid Cells in healing pathways, including the gene encoding Anti-Helminth Immunity? the epidermal growth factor family member am- phiregulin [22]. As part of the epidermal growth Group 2 ILCs were first described as previously factor family, amphiregulin is involved in tissue unrecognized mediators of anti-parasitic immu- remodeling and repair in diverse settings rang- nity in the intestine, promoting expulsion of hel- ing from acute airway epithelial injury to chron- minth parasites via IL-5 and IL-13-induced epi- ic allergic airway inflammation [32, 33, 40, 41] . thelial turnover and mucus production [7–9] . Al- Despite advances in understanding the influ- though adult helminth worms reside in the gut, ence of epithelial cell-derived amphiregulin, the the larvae must first migrate through the lung hematopoietic sources of amphiregulin have re- parenchyma before being coughed up and swal- mained poorly defined. Remarkably, IL-33 stim- lowed to complete maturation within the intes- ulation induced amphiregulin production in tine [20, 44] . Larval passage through the lung re- lung ILCs, and amphiregulin expression was el- sults in severe acute tissue damage and triggers evated in the lung following exposure to H1N1 development of chronic fibrosis that persists for subtype of influenza A virus, which causes sub- weeks after the parasite leaves the respiratory stantial damage to the respiratory epithelium tract [20, 44]. The cells and effector mechanisms [42] . Unlike the H3N1 subtype used by Chang et involved in regulating lung tissue remodeling fol- al. [21] , the H1N1 strain is not reported to cause lowing helminth-induced respiratory damage re- AHR but instead more closely recapitulates the main poorly defined. In two recent studies, Liang cytopathic activity of many human influenza A et al. [45] and Yasuda et al. [25] observed that IL- strains [42] . ILCs accumulated in the lung paren- 13-producing ILCs accumulated in the lung in chyma following H1N1 infection and depletion response to two helminth parasites, Nippostron- of ILCs in influenza virus-infected Rag1–/– hosts gylus brasiliensis or Strongyloides venezuelensis . using anti-CD90 monoclonal antibody treat- Although the precise role of ILCs in the lung ment resulted in severely decreased lung func- phase of helminth infection is not yet clear, it is tion, compromised lung epithelial barrier integ- conceivable that IL-5 produced by the activated rity and increased host mortality, revealing a ILCs may aid in either parasite killing or expul- crucial role for group 2 ILCs in regulating lung sion of the parasite from the lung via recruitment epithelial repair and remodeling [22]. Therapeu- of eosinophils [25] (fig. 2 ). Furthermore, since IL- tic treatment with exogenous amphiregulin ef- 13 is a potent inducer of epithelial hyperplasia fectively restored lung function and epithelial characteristic of both injury repair responses and repair responses in ILC-depleted influenza vi- chronic pulmonary fibrosis [19, 32, 33] , it is pos- rus-infected mice, while administration of re- sible that ILC-derived IL-13 may regulate repair combinant IL-13 was not sufficient, suggesting and remodeling of the injured lung. that ILC-derived amphiregulin is one mecha- nism by which ILCs can regulate lung tissue homeostasis following acute tissue damage Implications for Targeting Group 2 Innate (fig. 2 ). Notably, amphiregulin is widely ex- Lymphoid Cells in Human Disease pressed throughout multiple soft tissue sites [43] , raising the possibility that ILC-intrinsic amphi- Allergic airway diseases affect over 300 million regulin expression may influence tissue homeo- people worldwide [30, 31] , representing a signifi- stasis at extrapulmonary sites. cant global public health burden. While there is some clinical evidence showing an association be-

64 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) tween elevated levels of IL-5 and IL-13 and inci- expression profile reminiscent of CD4+ Th17 dence of allergic airway disorders in human pa- cells. A variety of names have been used to de- tients [19, 30, 31, 46], the potential contribution of scribe IL-17A/IL-22-producing ILCs, including ILCs to the development and/or progression of al- LTi-like cells, ILC22 and ILC17 [1, 12, 16, 48] . As lergic inflammation in humans is unclear. The their name suggests, fetal LTi cells orchestrate ability of ILCs to secret high levels of Th2 cell-as- the development of secondary lymphoid organs sociated cytokines, and their role in murine mod- through a lymphotoxin-dependent interaction els of AHR indicate that ILCs could be an attrac- with neighboring stromal cells to recruit hemato- tive potential target for improved immunothera- poietic cells to the site of organogenesis [49] . pies. Supporting a role for ILCs in human allergic Postnatally, a population of CD90+ CD25+ airway disorders, Mjosberg et al. [29] found elevat- CD127+ ROR␥ t-dependent ILCs phenotypically ed numbers of CRTH2+ ILCs in the nasal polyps of analogous to their fetal counterparts persists in patients with chronic rhinosinusitis, an allergic the adult lymphoid tissue and intestinal mucosa type 2 inflammatory disease characterized by high [12, 16] . Similar to fetal LTi cells, these ROR ␥ t + levels of circulating IgE and eosinophilia. Based on adult ILCs are also capable of producing lympho- these observations, it is possible that high amounts toxin, and ILC-derived LT ␣ ␤ has been shown of IL-5 produced by CRTH2+ ILCs within the to be essential for the development of intestinal nasal polyps can contribute to the characteristic cryptopatches and infection-induced isolated eosinophilia observed in this disease. Consistent lymphoid follicles [50, 51]. Furthermore, adult with this hypothesis, patients with rhinosinusitis ‘LTi-like’ ILCs are distinguished by their ability polyps have higher levels of IL-5 and IL-13 tran- to produce high amounts of IL-22 or IL-17A in scripts than patients without polyps [29] , suggest- response to IL-23 and IL-1 ␤ stimulation ( fig. 1 ). ing that CRTH2+ ILCs may be an important target Additionally, a separate population of IL-22-pro- in the development of new therapeutic strategies to ducing ILCs expresses the NK cell cytotoxicity ameliorate allergic airway inflammation. It is re- receptor NKp46, and therefore has been termed markable that elevated levels of amphiregulin ex- NCR22 or NKR-LTi [13–15, 52] . While the ex- pression have been reported in the sputum of pa- pression of NCRs suggests a possible lineage rela- tients undergoing an acute asthma attack [40] , tionship with NK cells, fate-mapping studies in- suggesting that ILC-derived amphiregulin could dicate that NKp46+ group 3 ILCs develop from also be involved in airway remodeling in human ROR␥ t + precursors and are not a subpopulation disease [47] . Furthermore, given the ability of am- of NK cells [52–55] . phiregulin to promote epithelial cell proliferation Several studies have identified analogous and mucus production [30, 32] , it is possible that ROR␥ t + ILCs in human adult tonsil or intestinal ILC may also be involved in the pathology associ- tissue that produce IL-22 or IL-17A [15, 53, 56– ated with chronic inflammatory lung disorders 58]. NCR-expressing ILCs are also present in hu- such as sarcoidosis or COPD. mans, although they are characterized by expres- sion of NKp44 instead of NKp46 as reported in mice [15, 53, 58]. Most recently, a constitutive Identification of Group 3 Innate Lymphoid population of IL-22-expressing ROR ␥ t + ILCs was Cells identified under steady-state conditions in the in- testine of multiple mammalian species, including The third ILC subset is comprised of several dis- humans, non-human primates and mice [59] , tinct cell populations that share a common devel- suggesting that ILCs have evolved to influence in- opmental requirement for ROR ␥ t and a cytokine testinal homeostasis across multiple species.

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 65

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) R O R t + Innate Lymphoid Cells Promote Group 3 Innate Lymphoid Cells Promote Protective Immunity against Enteric Extraintestinal Tissue Regeneration Pathogens In addition to playing an essential role in protec- ILCs maintain a close dialogue with the neigh- tive immunity to enteric pathogens, IL-22-pro- boring intestinal epithelial cells (IEC), as ILC- ducing ROR␥ t + ILCs have also been implicated in derived IL-22 promotes IEC expression of ␤-de- maintaining tissue homeostasis at sites outside fensins and antimicrobial proteins like RegIII ␤ the intestine. and RegIII ␥ [12, 13, 60, 61]. Production of these Whole-body irradiation accompanied by bone antimicrobial factors is critical for maintaining marrow transplantation has proven to be an ef- IEC barrier integrity to protect against patho- fective treatment for combating leukemia in genic microbes, such as the murine enteric bac- many human patients [65] . However, the process terium Citrobacter rodentium , an attaching and can also result in acute damage to multiple organ effacing bacterium similar to the human patho- systems, including disruption of thymic function gen Escherichia coli [12, 13, 62] . Several studies that limits the ability of patients to mount suc- have shown that the ILC-IL-22 pathway is criti- cessful adaptive immune responses long after ir- cal for immunity to C. rodentium infection, as radiation therapy has ended [65]. The cellular antibody-mediated depletion of ILCs or block- and molecular factors that regulate thymic re- ade of IL-22-IL-22R signaling resulted in severe generation after tissue injury remain poorly bacterial dissemination and increased host mor- understood. Recent work by Dudakov et al. [66] tality [12, 13, 62] . IL-23 is crucial for induction of identified a novel role for IL-22-producing IL-22 expression in ROR ␥ t + ILCs and subse- ROR␥ t + ILCs in thymic tissue homeostasis. Fol- quent bacterial clearance, as inhibition of IL-23 lowing irradiation or steroid-induced injury, IL- signaling also resulted in greater host morbidity 22 expression was elevated in the thymus, and ge- and mortality [12, 62] . Recently, studies from netic ablation of IL-22-IL-22R signaling resulted three independent research groups identified an in substantial impairment of thymic tissue regen- important role for the transcriptional regulator eration. Surprisingly, the dominant source of IL- aryl hydrocarbon receptor (AhR) in driving 22 after irradiation was not conventional T cells, ILC-intrinsic IL-22 production in the steady but instead was comprised of a radioresistant state and during C. rodentium infection [48, 63, population of intrathymic CD3– ROR ␥ t + ILCs 64] . AhR is thought to act as an environmental that phenotypically resembled NKp46– LTi-like sensor, and many naturally occurring AhR li- ILCs [66]. Therapeutic administration of recom- gands are found in plant-based foods [63] . AhR- binant IL-22 increased thymic regeneration by deficient mice have fewer IL-22-expressing ILCs promoting proliferation and survival of thymic prior to infection and rapidly succumbed to C. epithelial cells [66], demonstrating that the ILC- rodentium infection at similar kinetics as ob- IL-22 pathway can operate as a central mediator served for IL-23-deficient animals [48, 63, 64] . of epithelial cell homeostasis and tissue repair One study also showed that administration of outside the intestinal environment. certain dietary ligands led to upregulation of AhR and the accumulation of ROR␥ t + ILCs in the intestine [63] , suggesting that diet may be an important influence on ILC function in the in- testinal microenvironment.

66 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) Innate Lymphoid Cells Regulate sues of humans, non-human primates and mice Containment of Commensal Bacteria to [68, 69]. Strikingly, administration of Al caligenes Maintain Homeostasis spp. was sufficient to induce systemic immune ac- tivation to a similar degree as observed in ILC- Maintenance of intestinal barrier integrity is es- depleted mice [59] , supporting the notion that sential to preserve anatomical containment of the ILC-regulated containment of commensal bacte- trillions of commensal microbes that inhabit the ria within lymphoid tissue is essential to maintain gut microenvironment [67]. It is becoming in- homeostasis ( fig. 3 ). Remarkably, Alcaligenes -spe- creasingly clear that in addition to the vast array of cific IgG responses were detected in human pa- microbial species inhabiting the intestinal lumen, tients suffering from progressive hepatitis C virus some commensal bacterial species are anatomi- (HCV) infection, linking these observations to cally sequestered in gut-associated lymphoid tis- human disease and raising the possibility that tar- sue such as the Peyer’s patches [68, 69]. While the geting the ILC-IL-22 pathway could be used ther- symbiotic relationship between commensal bacte- apeutically to limit disease progression. ria and the mammalian host is essential for nor- mal digestion and proper nutrition, unrestrained microbial dissemination has severe consequences Targeting Group 3 Innate Lymphoid Cells in for the host, resulting in systemic immune activa- Human Disease tion that has been associated with progression of multiple chronic inflammatory diseases, such as Given the presence of phenotypically and func- HIV and hepatitis infections [70–76] . tionally analogous ILC populations in multiple In a recent study, a novel role for IL-22-ex- mammalian species, it is possible that develop- pressing ILCs in containment of lymphoid-resi- ment of new therapeutic strategies to target group dent commensal bacteria to limit systemic im- 3 ILC function may offer improvement in the mune activation was reported. Following anti- speed of tissue regeneration after injury or may body-mediated depletion of CD90+ ILCs, multiple help limit systemic immune activation in chronic hallmarks of systemic immune activation, includ- disease. ing elevated proinflammatory cytokines IL-6 Specifically, the identification of IL-22-pro- and TNF␣ , LPS in the serum, splenomegaly and ducing ILCs as novel players in thymic regenera- spread of bacteria to peripheral organs were ob- tion opens the possibility that manipulation of served [59]. This effect appeared to be mediated the IL-22-IL-22R pathway or selective enhance- by IL-22+ ILCs, as administration of recombinant ment of ILC function may aid in restoring adap- IL-22 was able to induce IEC expression of mul- tive immune function in patients undergoing tiple antimicrobial factors and limited systemic bone marrow irradiation therapy. Reconstitution immune activation [59] . Surprisingly, depletion of of the hematopoietic immune system from the ILCs did not appear to impair global intestinal donor bone marrow requires at least 6 months barrier function, suggesting that the immune ac- [65], leaving patients vulnerable to deadly infec- tivation observed in ILC-depleted mice was not tion with opportunistic pathogens that would due to spread of intestinal luminal bacteria as a normally be easily defeated by a fully mature im- result of impaired IEC barrier integrity. Pyrose- mune system. Addition of recombinant IL-22 quencing revealed that the majority of dissemi- protein to the drug regimens given to patients re- nating bacteria were Alcaligenes species, a com- covering from irradiation treatment could poten- mensal organism previously shown to reside in tially enhance the rate of thymic epithelial cell Peyer’s patches and gut-associated lymphoid tis- regeneration and subsequently shorten the time-

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 67

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) Homeostasis ILC dysregulation

Anti-microbial proteins

PP PP DC DC Fig. 3. Group 3 ILCs regulate containment of commensal bacteria. Under homeostatic conditions, group 3 ILCs ex- IL-22 press IL-22 and possibly other factors that coordinately ILC ILC promote barrier integrity by inducing epithelial cells to proliferate and produce anti-microbial proteins. Howev- IL-22 + Alcaligenes er, upon dysregulation of group 3 ILC responses and loss other factors? of IL-22, commensal bacteria such as Alcaligenes species that were previously sequestered within lymphoid tis- sue like the Peyer’s patches (PP) are released systemical- Systemic ly within the host. This microbial translocation leads to immune activation multiple hallmarks of systemic immune activation, in- IL-6 TNF␣ cluding enhanced levels of IL-6 and TNF␣ cytokines, which may impact progression of chronic diseases.

frame needed to develop functional T cell immu- SIV infection [77, 78], supporting the hypothesis nity. that ILC dysfunction could play a role in progres- Multiple studies have provided evidence that sive HIV infection. microbial translocation and systemic immune In addition to impacting development of activation correlates with poor prognosis and chronic infections, group 3 ILCs have also been disease progression in human patients suffering implicated as central players in driving intestinal from HIV or chronic HCV [70–76]. However, the autoimmune disorders. A recent study in human identities of the cellular players and molecular patients suggests that ROR ␥ t + ILCs are an impor- factors involved in regulating the containment of tant contributor to intestinal pathology observed commensal microbiota have remained largely during inflammatory bowel disease [79] , sug- unknown. The recent identification of the ILC- gesting that targeting group 3 ILCs may be a use- IL-22 pathway as a novel regulator of commensal ful therapeutic strategy to combat intestinal in- bacterial containment is a crucial first step that flammation. has the potential to contribute toward the devel- The potential therapeutic benefits of targeting opment of novel therapeutics to prevent micro- ILC responses in human disease may not be lim- bial translocation. Further studies are needed to ited to the intestine or respiratory tract. New re- examine the functionality of human IL-22+ ILCs search suggests that ROR ␥ t + group 3 ILCs are during progressive HIV or HCV infection to help also found in the skin and may contribute to the determine whether manipulating the ILC-IL-22 development of psoriasis [80] . Whether group 2 pathway may be a viable therapeutic strategy. Re- ILC populations are also located in the skin and cent studies have provided evidence that IL-22+ could possibly play a role in development or pro- IL-7A+ ILCs are greatly diminished in the intesti- gression of Th2 cytokine-associated skin diseas- nal mucosa of non-human primates following es remains unknown.

68 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) Conclusions unclear whether ILCs exhibit functional plastic- ity that would enable them to acquire effector Recent work over the past several years has shed functions that are characteristic of each of the light on the development, regulation and effector three major groups of ILCs depending on differ- function of this novel family of immune cells. ent inflammatory cues they receive in the tissue ILC family members play vital roles in regulating environment. Delineating the lineage relation- protective immunity against pathogens and in ships and functional plasticity between these het- contributing to allergic inflammation and tissue erogeneous cell populations will be essential for homeostasis at barrier surfaces. While these designing novel therapeutic strategies to effec- studies demonstrate the critical impact of these tively manipulate human ILC function in the rare immune cells on multiple aspects of disease, context of infection, inflammation, tissue repair the study of ILC development and function is still and vaccination. in its infancy and many questions remain to be addressed. The ILC family is remarkably hetero- geneous, and the lineage relationships between Acknowledgements members of each of the three ILC groups are not We would like to thank members of the Artis laboratory well defined. In particular, the question of wheth- for helpful discussions and critical reading of the man- er natural helper cells, nuocytes and innate help- uscript. Research in the Artis lab is supported by the er type 2 cells represent distinct effector cell types National Institutes of Health (AI061570, AI087990, AI074878, AI083480, AI095466, AI095608 and AI097333 or alternatively belong to the same group 2 ILC to D.A.; T32-AI007532 to L.A.M.) and the Burroughs population existing at multiple anatomical loca- Wellcome Fund Investigator in Pathogenesis of Infec- tions remains controversial. Furthermore, it is tious Disease Award (D.A.).

References

1 Spits H, Cupedo T: Innate lymphoid 6 Koyasu S, Moro K, Tanabe M, Takeuchi 10 Saenz SA, Siracusa MC, Perrigoue JG, cells: emerging insights in develop- T: Natural helper cells: a new player in Spencer SP, Urban JF Jr, Tocker JE, Bu- ment, lineage relationships, and func- the innate immune response against delsky AL, Kleinschek MA, Kastelein

tion. Annu Rev Immunol 2012; 30: 647– helminth infection. Adv Immunol RA, Kambayashi T, Bhandoola A, Artis

675. 2010; 108: 21–44. D: IL25 elicits a multipotent progenitor 2 Spits H, Di Santo JP: The expanding 7 Moro K, Yamada T, Tanabe M, Takeuchi cell population that promotes T(H)2

family of innate lymphoid cells: regula- T, Ikawa T, Kawamoto H, Furusawa J, cytokine responses. Nature 2010; 464: tors and effectors of immunity and tis- Ohtani M, Fujii H, Koyasu S: Innate pro- 1362–1366.

sue remodeling. Nat Immunol 2011; 12: duction of T(H)2 cytokines by adipose 11 Wong SH, Walker JA, Jolin HE, Drynan 21–27. tissue-associated c-Kit(+)Sca-1(+) lym- LF, Hams E, Camelo A, Barlow JL, Neill

3 Saenz SA, Noti M, Artis D: Innate im- phoid cells. Nature 2010; 463: 540–544. DR, Panova V, Koch U, Radtke F, Hard- mune cell populations function as ini- 8 Neill DR, Wong SH, Bellosi A, Flynn RJ, man CS, Hwang YY, Fallon PG, McKen- tiators and effectors in Th2 cytokine Daly M, Langford TK, Bucks C, Kane zie AN: Transcription factor RORalpha

responses. Trends Immunol 2010; 31: CM, Fallon PG, Pannell R, Jolin HE, is critical for nuocyte development. Nat 407–413. McKenzie AN: Nuocytes represent a Immunol 2012;13:229–236. 4 Sun JC, Lanier LL: NK cell develop- new innate effector leukocyte that me- 12 Sonnenberg GF, Monticelli LA, Elloso

ment, homeostasis and function: paral- diates type-2 immunity. Nature 2010; MM, Fouser LA, Artis D: CD4(+) lym-

lels with CD8 T cells. Nat Rev Immunol 464: 1367–1370. phoid tissue-inducer cells promote in-

2011; 11: 645–657. 9 Price AE, Liang HE, Sullivan BM, Rein- nate immunity in the gut. Immunity

5 Monticelli LA, Sonnenberg GF, Artis D: hardt RL, Eisley CJ, Erle DJ, Locksley 2011; 34: 122–134. Innate lymphoid cells: critical regula- RM: Systemically dispersed innate IL- tors of allergic inflammation and tissue 13-expressing cells in type 2 immunity.

repair in the lung. Curr Opin Immunol Proc Natl Acad Sci USA 2010; 107:

2012; 24: 284–289. 11489–11494.

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 69

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) 13 Satoh-Takayama N, Vosshenrich CA, 23 Kim HY, Chang YJ, Subramanian S, 31 Corren J: Cytokine inhibition in severe Lesjean-Pottier S, Sawa S, Lochner M, Lee HH, Albacker LA, Matangkasom- asthma: current knowledge and future

Rattis F, Mention JJ, Thiam K, Cerf- but P, Savage PB, McKenzie AN, Smith directions. Curr Opin Pulm Med 2011;

Bensussan N, Mandelboim O, Eberl G, DE, Rottman JB, DeKruyff RH, Umetsu 17: 29–33. Di Santo JP: Microbial flora drives in- DT: Innate lymphoid cells responding 32 Crosby LM, Waters CM: Epithelial re- terleukin 22 production in intestinal to IL-33 mediate airway hyperreactiv- pair mechanisms in the lung. Am J

NKp46+ cells that provide innate mu- ity independently of adaptive immu- Physiol Lung Cell Mol Physiol 2010;

cosal immune defense. Immunity 2008; nity. J Allergy Clin Immunol 2012; 129: 298:L715–L731.

29: 958–970. 216–227:e211–e216. 33 Rock JR, Hogan BL: Epithelial progeni- 14 Sanos SL, Bui VL, Mortha A, Oberle K, 24 Barlow JL, Bellosi A, Hardman CS, tor cells in lung development, mainte- Heners C, Johner C, Diefenbach A: Drynan LF, Wong SH, Cruickshank JP, nance, repair, and disease. Annu Rev

RORgammat and commensal micro- McKenzie AN: Innate IL-13-producing Cell Dev Biol 2011; 27: 493–512. flora are required for the differentia- nuocytes arise during allergic lung in- 34 Byers DE, Holtzman MJ: Alternatively tion of mucosal interleukin 22-produc- flammation and contribute to airways activated macrophages and airway dis-

ing NKp46+ cells. Nat Immunol 2009; hyperreactivity. J Allergy Clin Immu- ease. Chest 2011; 140: 768–774.

10: 83–91. nol 2012; 129: 191–198.e1–e4. 35 Moreira AP, Hogaboam CM: Macro- 15 Cella M, Fuchs A, Vermi W, Facchetti F, 25 Yasuda K, Muto T, Kawagoe T, Matsu- phages in allergic asthma: fine-tuning Otero K, Lennerz JK, Doherty JM, Mills moto M, Sasaki Y, Matsushita K, Taki their pro- and anti-inflammatory ac- JC, Colonna M: A human natural killer Y, Futatsugi-Yumikura S, Tsutsui H, tions for disease resolution. J Interferon

cell subset provides an innate source of Ishii KJ, Yoshimoto T, Akira S, Nakani- Cytokine Res 2011; 31: 485–491. IL-22 for mucosal immunity. Nature shi K: Contribution of IL-33-activated 36 Wilhelm C, Turner JE, Van Snick J,

2009; 457: 722–725. type II innate lymphoid cells to pulmo- Stockinger B: The many lives of IL-9: a 16 Takatori H, Kanno Y, Watford WT, Tato nary eosinophilia in intestinal nema- question of survival? Nat Immunol

CM, Weiss G, Ivanov II, Littman DR, tode-infected mice. Proc Natl Acad Sci 2012; 13: 637–641.

O’Shea JJ: Lymphoid tissue inducer-like USA 2012; 109: 3451–3456. 37 Noelle RJ, Nowak EC: Cellular sources cells are an innate source of IL-17 and 26 Bartemes KR, Iijima K, Kobayashi T, and immune functions of interleu-

IL-22. J Exp Med 2009; 206: 35–41. Kephart GM, McKenzie AN, Kita H: kin-9. Nat Rev Immunol 2010; 10: 683– 17 Mortha A, Diefenbach A: Natural killer IL-33-responsive lineage- 687. cell receptor-expressing innate lym- CD25+CD44hi lymphoid cells mediate 38 Wilhelm C, Hirota K, Stieglitz B, Van phocytes: more than just NK cells. Cell innate type 2 immunity and allergic Snick J, Tolaini M, Lahl K, Sparwasser

Mol Life Sci 2011; 68: 3541–3555. inflammation in the lungs. J Immunol T, Helmby H, Stockinger B: An IL-9 fate

18 Oliphant CJ, Barlow JL, McKenzie AN: 2012; 188: 1503–1513. reporter demonstrates the induction of Insights into the initiation of type 2 im- 27 Yang Q, Saenz SA, Zlotoff DA, Artis D, an innate IL-9 response in lung inflam-

mune responses. Immunology 2011; Bhandoola A: Cutting edge: natural mation. Nat Immunol 2011; 12: 1071–

134: 378–385. helper cells derive from lymphoid pro- 1077.

19 Doherty T, Broide D: Cytokines and genitors. J Immunol 2011; 187: 5505– 39 Oh CK, Raible D, Geba GP, Molfino NA: growth factors in airway remodeling in 5509. Biology of the interleukin-9 pathway

asthma. Curr Opin Immunol 2007; 19: 28 Halim TY, Krauss RH, Sun AC, Takei F: and its therapeutic potential for the 676–680. Lung natural helper cells are a critical treatment of asthma. Inflamm Allergy

20 Allen JE, Maizels RM: Diversity and source of Th2 cell-type cytokines in Drug Targets 2011; 10: 180–186. dialogue in immunity to helminths. protease allergen-induced airway in- 40 Enomoto Y, Orihara K, Takamasu T,

Nat Rev Immunol 2011; 11: 375–388. flammation. Immunity 2012; 36: 451– Matsuda A, Gon Y, Saito H, Ra C, 21 Chang YJ, Kim HY, Albacker LA, 463. Okayama Y: Tissue remodeling induced Baumgarth N, McKenzie AN, Smith 29 Mjosberg JM, Trifari S, Crellin NK, Pe- by hypersecreted epidermal growth DE, Dekruyff RH, Umetsu DT: Innate ters CP, van Drunen CM, Piet B, Fok- factor and amphiregulin in the airway lymphoid cells mediate influenza-in- kens WJ, Cupedo T, Spits H: Human after an acute asthma attack. J Allergy

duced airway hyper-reactivity indepen- IL-25- and IL-33-responsive type 2 in- Clin Immunol 2009; 124: 913–920.e1–e7. dently of adaptive immunity. Nat Im- nate lymphoid cells are defined by ex- 41 Bierman A, Yerrapureddy A, Reddy

munol 2011; 12: 631–638. pression of CRTH2 and CD161. Nat NM, Hassoun PM, Reddy SP: Epider-

22 Monticelli LA, Sonnenberg GF, Abt Immunol 2011; 12: 1055–1062. mal growth factor receptor (EGFR) reg- MC, Alenghat T, Ziegler CG, Doering 30 Broide DH: Immunologic and inflam- ulates mechanical ventilation-induced

TA, Angelosanto JM, Laidlaw BJ, Yang matory mechanisms that drive asthma lung injury in mice. Transl Res 2008;

CY, Sathaliyawala T, Kubota M, Turner progression to remodeling. J Allergy 152: 265–272.

D, Diamond JM, Goldrath AW, Farber Clin Immunol 2008; 121: 560–570, quiz 42 Fukushi M, Ito T, Oka T, Kitazawa T, DL, Collman RG, Wherry EJ, Artis D: 571–572. Miyoshi-Akiyama T, Kirikae T, Ya- Innate lymphoid cells promote lung- mashita M, Kudo K: Serial histopatho- tissue homeostasis after infection with logical examination of the lungs of

influenza virus. Nat Immunol 2011; 12: mice infected with influenza A virus

1045–1054. PR8 strain. PLoS One 2011; 6:e21207.

70 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) 43 Busser B, Sancey L, Brambilla E, Coll 53 Crellin NK, Trifari S, Kaplan CD, Cupe- 62 Zheng Y, Valdez PA, Danilenko DM, Hu JL, Hurbin A: The multiple roles of am- do T, Spits H: Human NKp44+IL-22+ Y, Sa SM, Gong Q, Abbas AR, Modru- phiregulin in human cancer. Biochim cells and LTi-like cells constitute a sta- san Z, Ghilardi N, de Sauvage FJ,

Biophys Acta 2011; 1816: 119–131. ble RORC+ lineage distinct from con- Ouyang W: Interleukin-22 mediates 44 Anthony RM, Rutitzky LI, Urban JF Jr, ventional natural killer cells. J Exp Med early host defense against attaching

Stadecker MJ, Gause WC: Protective 2010; 207: 281–290. and effacing bacterial pathogens. Nat

immune mechanisms in helminth in- 54 Sawa S, Cherrier M, Lochner M, Satoh- Med 2008; 14: 282–289.

fection. Nat Rev Immunol 2007; 7: 975– Takayama N, Fehling HJ, Langa F, Di 63 Kiss EA, Vonarbourg C, Kopfmann S, 987. Santo JP, Eberl G: Lineage relationship Hobeika E, Finke D, Esser C, Diefen- 45 Liang HE, Reinhardt RL, Bando JK, analysis of RORgammat+ innate lym- bach A: Natural aryl hydrocarbon re-

Sullivan BM, Ho IC, Locksley RM: Di- phoid cells. Science 2010; 330: 665–669. ceptor ligands control organogenesis of vergent expression patterns of IL-4 and 55 Vonarbourg C, Mortha A, Bui VL, Her- intestinal lymphoid follicles. Science

IL-13 define unique functions in aller- nandez PP, Kiss EA, Hoyler T, Flach M, 2011; 334: 1561–1565.

gic immunity. Nat Immunol 2012; 13: Bengsch B, Thimme R, Holscher C, 64 Qiu J, Heller JJ, Guo X, Chen ZM, Fish 58–66. Honig M, Pannicke U, Schwarz K, Ware K, Fu YX, Zhou L: The aryl hydrocar- 46 Wilson MS, Wynn TA: Pulmonary fi- CF, Finke D, Diefenbach A: Regulated bon receptor regulates gut immunity brosis: pathogenesis, etiology and regu- expression of nuclear receptor ROR- through modulation of innate lym-

lation. Mucosal Immunol 2009; 2: 103– gammat confers distinct functional phoid cells. Immunity 2012; 36: 92–104. 121. fates to NK cell receptor-expressing 65 Legrand N, Dontje W, van Lent AU, 47 Wills-Karp M, Finkelman FD: Innate RORgammat(+) innate lymphocytes. Spits H, Blom B: Human thymus regen-

lymphoid cells wield a double-edged Immunity 2010; 33: 736–751. eration and T cell reconstitution. Se-

sword. Nat Immunol 2011; 12: 1025– 56 Crellin NK, Trifari S, Kaplan CD, Satoh- min Immunol 2007; 19: 280–288. 1027. Takayama N, Di Santo JP, Spits H: Reg- 66 Dudakov JA, Hanash AM, Jenq RR, 48 Lee JS, Cella M, McDonald KG, Garlan- ulation of cytokine secretion in human Young LF, Ghosh A, Singer NV, West da C, Kennedy GD, Nukaya M, Man- CD127(+) LTi-like innate lymphoid ML, Smith OM, Holland AM, Tsai JJ, tovani A, Kopan R, Bradfield CA, New- cells by Toll-like receptor 2. Immunity Boyd RL, van den Brink MR: Interleu-

berry RD, Colonna M: AHR drives the 2010; 33: 752–764. kin-22 drives endogenous thymic re-

development of gut ILC22 cells and 57 Cupedo T, Crellin NK, Papazian N, generation in mice. Science 2012; 336: postnatal lymphoid tissues via path- Rombouts EJ, Weijer K, Grogan JL, 91–95. ways dependent on and independent of Fibbe WE, Cornelissen JJ, Spits H: Hu- 67 Artis D: Epithelial-cell recognition of

Notch. Nat Immunol 2011; 13: 144–151. man fetal lymphoid tissue-inducer cells commensal bacteria and maintenance 49 Eberl G, Marmon S, Sunshine MJ, are interleukin 17-producing precur- of immune homeostasis in the gut. Nat

Rennert PD, Choi Y, Littman DR: An sors to RORC+ CD127+ natural killer- Rev Immunol 2008; 8: 411–420.

essential function for the nuclear re- like cells. Nat Immunol 2009; 10: 66–74. 68 Kunisawa J, Kiyono H: Alcaligenes is ceptor RORgamma(t) in the generation 58 Cella M, Otero K, Colonna M: Expan- commensal bacteria habituating in the of fetal lymphoid tissue inducer cells. sion of human NK-22 cells with IL-7, gut-associated lymphoid tissue for the

Nat Immunol 2004; 5: 64–73. IL-2, and IL-1beta reveals intrinsic regulation of intestinal IgA responses.

50 Lochner M, Ohnmacht C, Presley L, functional plasticity. Proc Natl Acad Front Immunol 2012; 3: 65.

Bruhns P, Si-Tahar M, Sawa S, Eberl G: Sci USA 2010; 107: 10961–10966. 69 Obata T, Goto Y, Kunisawa J, Sato S, et Microbiota-induced tertiary lymphoid 59 Sonnenberg GF, Monticelli LA, Alen- al: Indigenous opportunistic bacteria tissues aggravate inflammatory disease ghat T, Fung TC, et al: Innate lymphoid inhabit mammalian gut-associated in the absence of RORgamma t and LTi cells promote anatomical containment lymphoid tissues and share a mucosal

cells. J Exp Med 2011; 208: 125–134. of lymphoid-resident commensal bac- antibody-mediated symbiosis. Proc

51 Marchesi F, Martin AP, Thirunarayan- teria. Science 2012; 336: 1321–1325. Natl Acad Sci USA 2010; 107: 7419–7424. an N, Devany E, Mayer L, Grisotto MG, 60 Sanos SL, Vonarbourg C, Mortha A, 70 Brenchley JM, Price DA, Schacker TW, Furtado GC, Lira SA: CXCL13 expres- Diefenbach A: Control of epithelial cell Asher TE, et al: Microbial translocation sion in the gut promotes accumulation function by interleukin-22-producing is a cause of systemic immune activa- of IL-22-producing lymphoid tissue- RORgammat(+) innate lymphoid cells. tion in chronic HIV infection. Nat Med

inducer cells, and formation of isolated Immunology 2011; 132: 453–465. 2006; 12: 1365–1371. lymphoid follicles. Mucosal Immunol 61 Sonnenberg GF, Fouser LA, Artis D: 71 d’Ettorre G, Paiardini M, Zaffiri L, An-

2009; 2: 486–494. Functional biology of the IL-22-IL-22R dreotti M, Ceccarelli G, Rizza C, Indin- 52 Satoh-Takayama N, Lesjean-Pottier S, pathway in regulating immunity and nimeo M, Vella S, Mastroianni CM, Vieira P, Sawa S, Eberl G, Vosshenrich inflammation at barrier surfaces. Adv Silvestri G, Vullo V: HIV persistence in

CA, Di Santo JP: IL-7 and IL-15 inde- Immunol 2010; 107: 1–29. the gut mucosa of HIV-infected sub- pendently program the differentiation jects undergoing antiretroviral therapy of intestinal CD3-NKp46+ cell subsets correlates with immune activation and from Id2-dependent precursors. J Exp increased levels of LPS. Curr HIV Res

Med 2010; 207: 273–280. 2011; 9: 148–153.

ILCs Promote Immunity and Tissue Homeostasis at Barrier Surfaces 71

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) 72 Douek D: HIV disease progression: im- 78 Reeves RK, Rajakumar PA, Evans TI, Reference mune activation, microbes, and a leaky Connole M, Gillis J, Wong FE, Kuzmi- 1 Hoyler T, Klose CS, Souabni A, Tur-

gut. Top HIV Med 2007; 15: 114–117. chev YV, Carville A, Johnson RP: Gut queti-Neves A, Pfeifer D, Rawlins EL, 73 Page EE, Nelson M, Kelleher P: HIV inflammation and indoleamine deoxy- Voehringer D, Busslinger M, Diefen- and hepatitis C coinfection: pathogen- genase inhibit IL-17 production and bach A: The transcription factor esis and microbial translocation. Curr promote cytotoxic potential in NKp44+ GATA-3 controls cell fate and main-

Opin HIV AIDS 2011; 6: 472–477. mucosal NK cells during SIV infection. tenance of type 2 innate lymphoid

74 Paiardini M, Frank I, Pandrea I, Blood 2011; 118: 3321–3330. cells. Immunity 2012;37:634–648. Apetrei C, Silvestri G: Mucosal immune 79 Geremia A, Arancibia-Carcamo CV, 2 Mjösberg J, Bernink J, Golebski K, dysfunction in AIDS pathogenesis. Fleming MP, Rust N, Singh B, Karrich JJ, Peters CP, Blom B, te Vel-

AIDS Rev 2008; 10: 36–46. Mortensen NJ, Travis SP, Powrie F: IL- de AA, Fokkens WJ, van Drunen 75 Sandler NG, Koh C, Roque A, Eccleston 23-responsive innate lymphoid cells are CM, Spits H: The transcription factor JL, Siegel RB, Demino M, Kleiner DE, increased in inflammatory bowel dis- GATA3 is essential for the function

Deeks SG, Liang TJ, Heller T, Douek ease. J Exp Med 2011; 208: 1127–1133. of human type 2 innate lymphoid DC: Host response to translocated mi- 80 Pantelyushin S, Haak S, Ingold B, Kulig cells. Immunity 2012;37:649–659. crobial products predicts outcomes of P, Heppner FL, Navarini AA, Becher B: 3 Sciumé G, Hirahara K, Takahashi H, patients with HBV or HCV infection. Rorgammat+ innate lymphocytes and Laurence A, Villarino AV, Singleton

Gastroenterology 2011; 141: 1220–1230, gammadelta T cells initiate psoriasi- KL, Spencer SP, Wilhelm C, Poholek 1230.e1–3. form plaque formation in mice. J Clin AC, Vahedi G, Kanno Y, Belkaid Y,

76 Marchetti G, Cozzi-Lepri A, Merlini Invest 2012; 122: 2252–2256. O’Shea JJ: Distinct requirements for E, Bellistri GM, Castagna A, Galli M, T-bet in gut innate lymphoid cells. J Verucchi G, Antinori A, Costantini A, Exp Med 2012;209:2331–2338. Giacometti A, di Caro A, D’Arminio 4 Klose CS, Kiss EA, Schwierzeck V, Monforte A: Microbial translocation Note Added in Proof Ebert K, Hoyler T, d’Hargues Y, Göp- predicts disease progression of HIV- pert N, Croxford AL, Waisman A, infected antiretroviral-naive patients Two recent studies have demon- Tanriver Y, Diefenbach A: A T-bet

with high CD4+ cell count. AIDS 2011; strated that the transcription factor gradient controls the fate and func-

25: 1385–1394. GATA3 regulates the development, tion of CCR6-RORγt+ innate lym- 77 Xu H, Wang X, Liu DX, Moroney-Ras- maintenance and function of group phoid cells. Nature 2013;494:261– mussen T, Lackner AA, Veazey RS: IL- 2 ILCs in mice and humans [1, 2]. 265. 17-producing innate lymphoid cells are Additionally, the transcription fac- restricted to mucosal tissues and are depleted in SIV-infected macaques. tor T-bet has recently been shown to influence the development and Mucosal Immunol 2012; 5: 658–669. functionality of intestinal RORγt+ NKp46+ ILCs [3, 4].

David Artis, PhD Department of Microbiology, Institute for Immunology Perelman School of Medicine, University of Pennsylvania 421 Curie Boulevard, BRB II/III, Room 356 Philadelphia, PA 19104-6160 (USA)

E-Mail dartis @ mail.med.upenn.edu

72 Monticelli Artis

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 59–72 (DOI: 10.1159/000346504) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535)

miRNAs That Shape the Innate Immune System: Regulation through Toll-Like Receptor Signaling

Anne M. Curtis Luke A.J. O’Neill

Department of Biochemistry and Immunology, Trinity Biomedical Science Institute, Trinity College Dublin, Dublin, Ireland

Abstract fore, the question was raised whether other short miRNAs are a large class of small non-coding RNA that RNAs could regulate genes in this manner and if regulate gene expression mainly by transcriptional sup- so how. In 1998, Andrew Fire and Craig Mello pression. miRNAs are integral to the normal functioning worked out the cellular components and mecha- of a cell, play an important role in modulating the innate nism by which miRNAs can bind to and degrade immune response, and are known to enhance or repress target mRNAs. For this, they were jointly award- TLR activity. These microRNAs maintain a number of ed the Nobel Prize in Physiology in 2006. feedback loops within the TLR pathway, forming a fine To date, over 18,000 miRNAs entries are listed balance of control. Dysregulation of TLR-regulated on the online sequence repository miRbase, and miRNAs within immune cells often leads to disease, un- there are over 16,000 publications on PubMed derscoring their importance within the TLR pathway. related to microRNAs. miRNAs can bind to 3 Copyright © 2013 S. Karger AG, Basel UTR, coding sequences or 5 UTR leading to mRNA degradation or inhibition of translation. miRNAs appear to be able to bind to many re- miRNAs are short sequences of non-coding RNA gions of the gene. It is now accepted that the pre- that are 20–23 nucleotides in length. The first re- dominant mechanism of miRNAs is through ported miRNA came from studies in Caenorhab- destabilization of target mRNA levels [4] ; howev- ditis elegans in 1993. The combined work of two er, a variety of additional modes of regulation groups led to the discovery that the small non- by miRNAs have also been uncovered. Depend- protein-coding transcript lin-4 negatively regu- ing on the miRNA:target interaction, certain lated lin-14 through binding to its 3 UTR region miRNAs can stabilize RNA transcripts leading to [1, 2] . Whether this type of gene suppression rep- greater protein expression. miR-369-3 can bind resented a broader phenomenon was not appar- to the AU-rich element critical proinflammatory ent until 7 years later when let-7 , a 21-nt RNA, was cytokine TNF ␣ to mediate translational upregu- found to regulate lin-41 through two closely lation [5] . miRNAs can also target gene promot- spaced target sites in the lin-41 3 UTR [3] . Let-7 ers. miR-373 binding sites were discovered in the is a conserved RNA from flies to human. There- promoters of E-cadherin and CSDC2, and over- expression of the miR was shown to induce ex- ␬B) and interferon regulatory factors (IRFs; pression of those genes [6]. miRNAs can also act fig. 1 ). These transcription factors, through RNA as decoy molecules through interfering with the polymerase II activation, drive a plethora of cyto- function of regulatory RNA-binding proteins [7], kines, chemokines and antimicrobial agents that and RNA-binding proteins can shield miRNA clear the pathogen and prime specific adaptive target sequences. Therefore, miRNAs have the immune responses. A number of steps are in- ability to orchestrate the molecular phenotype of volved in mounting this response, including the cell by modulating mRNA levels, affecting physical interactions at the extracellular junction protein translation and stability, and even gene of the receptor, conformational changes, signal- expression. Exact sequence homology between ing via posttranslational modifications through a the miRNA and its target sequence is not essen- range of adaptor and signaling molecules within tial for binding to occur, and web-based algo- the cytoplasm and cytoplasmic bodies, proteoso- rithms can predict the likelihood of a miRNA mal degradation, activation and recruitment of binding to a gene of interest. As each miRNA is transcription factors to nuclear DNA, gene ex- predicted to bind to approximately 200 genes, the pression and processing to mature mRNA and scope by which miRNAs can control the RNA translation into protein. and protein content of the cell is significant, and miRNAs appear to exert control over the it is estimated that miRNAs regulate more than whole family of TLRs, and this has been reviewed 30% of the human protein-coding genome [8] . extensively [13, 14]. In this report, we will focus Dysregulation of miRNAs can impact on a wide predominantly on the role of miRNAs within variety of conditions, from cancer [9, 10] to car- TLR4 signaling and function. diovascular disease [10], and inflammatory and autoimmune diseases [11, 12] . Toll-like receptors, of which TLR1–9 have R e s u l t s been characterized extensively in mammalian cells, are at the front line of defense against in- Firstly, activation of TLR4, which can either sig- vading pathogens. TLRs are expressed predomi- nal through the membrane-associated MyD88/ nantly on sentinel cells, like macrophages, den- Mal pathway or an endosomal TRIF/TRAM dritic, epithelial and endothelial cells. Each re- pathway, ultimately ends in transcription factor- ceptor singly spans the membrane and is equipped mediated activation of RNA polymerase II (fig. 1 ). to sense the presence of distinct structural com- This causes the concomitant transcription of pri- ponents of microbes, signal through the cell to miRNAs leading into the processing of mature mount a rapid immune response, and clear the miRNAs. NF- ␬B appears dominant in activating noxious agent. TLRs are a type of pattern recog- many of these miRNAs. The majority of miRNAs nition receptors that recognize molecules that are are induced in response to TLR4, whereas a sig- broadly shared by pathogens known as pathogen nificantly smaller number are repressed [13] . The associated molecular patterns but distinguish- targets for these regulated miRNAs include many able from host molecules. For example, TLR3 of the molecules involved in the signal transduc- recognizes viral double-stranded RNA, and tion and transcriptional pathway (fig. 1 ). These TLR4 senses lipopolysaccharide from Gram-neg- miRNAs can feed back and modulate the magni- ative bacteria. TLRs are one of the prominent ef- tude and duration of the TLR response. Overac- fectors of the innate immune system. The most tive TLR signaling can be extremely detrimental common downstream responses are activation of to an organism. Therefore, miRNAs are em- the transcription factors, nuclear factor- ␬B (NF- ployed as regulators to ensure both the appropri-

74 Curtis O’Neill

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) Proinflammatory TLR4 cytokines and interferons Extracellular

Cytoplasm

MMyD88y Mal miR-155 SOCS1 Endosome

IRAK4RAK4 TLR4

SHIP1SHIP1 IRAK1K1 IIRAK2R UbUb Ubb A20 miR-155 TRIF Ubc13/Ubc13 miR-146a TRAF6TRAF UbcH5c TRAM

TRADD TANK Pellino-1Pe TAB33 TTAB2AB2 TAK1 RIP1RIP1

TRAF3 NEMONEMO MAPKK IKKα IKK␤ Ubb

IIκBακBα TBK1TBK1 miR-155 NFNF␬B IKKε

AP-1A NFNF␬B IRF3IRF3

Nucleus

Fig. 1. The TLR4 signaling pathway. Activation of TLR4 causes a cascade of signaling changes leading to transcrip- tional activation of inflammatory genes. miRNAs that are induced by this signaling pathway are able to target pro- teins within the pathway, thus shaping the magnitude and duration of the response.

ate magnitude and duration of the signaling re- miR-146a induction by LPS is NF-␬ B depen- sponse and return to homeostasis. The control of dent, and miR-146a targets the TNF receptor-as- miRNAs by TLRs is also cell type specific, indi- sociated family (TRAF) 6 and interleukin-1 re- cating further regulation at the individual cellu- ceptor-associated kinase (IRAK) 1 and IRAK2 lar level. [15, 16]. miR-146a is induced to counter bacterial By microarray, Taganov et al. [15] identified a but not viral responses. It is upregulated through number of miRNAs that are induced with LPS. TLR2, TLR4 and TLR5 receptors that are mem- miR-132, miR-146a and miR-155 were found to brane associated, and not those, TLR3, TLR7 and increase substantially with LPS treatment. Al- TLR9, that sense viral nucleic acid and localize though not much work has been performed since internally. Even within TLR4 signaling, miR- on miR-132 and TLRs, both miR-146a and miR- 146a targeting TRAF6, IRAK1 and -2 will still 155 have now been studied extensively. leave the TRIF/TRAM antiviral pathway intact

miRNAs That Shape the Innate Immune System 75

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) [15] . IL-1 ␤ signaling also induces miR-146a, duction protein TAB2, is part of a negative feed- which inhibits IL-8 and RANTES production; back loop controlling IL-1 ␤ and other inflamma- however, this is not through targeting of IRAK tory cytokines [28]. Many of the B cell lympho- or TRAF6, but via direct regulation at the trans- mas with enhanced miR-155 expression are lational level of those two proteins [17] . This latently infected with Epstein-Barr virus (EBV). microRNA has been associated with a number of miR-155 targets IKK ␧, and downregulates in- immunological conditions such as osteoarthritis terferon signaling [29] . EBV uses host levels of [18] , rheumatoid arthritis [19] , and systemic lu- miR-155 to attenuate NF- ␬B- and IRF3-directed pus erythematosus [20] . A polymorphism in the transcription, which limits the innate immune 3 UTR of IRAK1 is associated with susceptibil- response to viral infection and allows for stabili- ity to rheumatoid arthritis and psoriatic arthritis zation of the latent virus. [21, 22] , adding further weight to the importance miR-155 can enhance the TLR4 response of miRNA-mediated regulation of this signaling through a variety of mechanisms. Suppressor of molecule. miR-146a and IRAK1 have been linked cytokine signaling 1 (SOCS1), a key inhibitor of to intestinal epithelial innate immune tolerance the inflammatory process that mediates Mal deg- [23]. After birth, the intestinal epithelial cells radation [30] , is repressed by mir-155 [31] . constantly turn over IRAK1 in order to maintain miR-155 induced by H. pylori activation of a level of tolerance as the intestinal mucosa un- TLR2/4 in macrophages appears to target pro- dergoes the dramatic transition from a sterile site apoptotic genes. This may be a protective host de- to a permanently colonized surface. Induction of fense mechanism to enhance macrophage resis- miR-146a during this period is critical for the tance to apoptosis induced by the DNA damage repression of IRAK1 and maintenance of the during H. pylori infection [32]. Regulus, a bio- ongoing signal transduction in tolerant cells. technology company that entered a strategic alli- LPS tolerance, the state of hyporesponsiveness ance with the pharmaceutical giant GlaxoSmith- to subsequent LPS challenge, to prevent the over- Kline, is pursuing miR-155 as its lead target for production of harmful cytokines like TNF␣ is inflammatory conditions, possibly due to the dependent on miR-146a targeting IRAK1 and substantial actions of miR-155 on the key inflam- TRAF6 and NF-␬ B activation [24] . matory cytokine TNF ␣ . In resting macrophages, miR-155 is one of the most highly abundant TNF ␣ message is constantly turning over as the and intensely studied miRNA. Initially discov- 3 UTR of the gene induces self-repression. In- ered to have a role in cancer, miR-155 has been duction and binding of miR-155 to the 3 UTR implicated in B cell lymphomas [25] and is known releases that repression and stabilizes the TNF␣ to play a role in B cell development. More recent- transcript and protein [33] . The potent inflam- ly, it has also been shown that miR-155 plays a matory effects of miR-155 are evident given that major role in inflammation. This microRNA is two anti-inflammatory molecules, IL-10 [34] and induced rapidly in macrophages with TLR2, -3, progesterone [35] repress miR-155 expression. -4 and -9 agonists, occurring through MyD88- or The inhibitory effect of IL-10 on miR-155 leads to TRIF-dependent signaling pathways [26] . miR- increased Src homology 2 domain containing 155 can target both positive and negative regula- inositol 5 -phosphatase 1 (SHIP1), a negative reg- tors of TLR4 signaling. It negatively regulates He- ulator of TLR-induced responses. Progesterone- licobacter pylori , induced inflammation through mediated miR-155 repression appears to enhance targeting MyD88 and reducing IL-8 levels [27] . SOCS1 expression. During dendritic cell activation by LPS, miR-155, Although targeting miR-155 may have the de- through its repressive action on the signal trans- sired consequences of reducing inflammation,

76 Curtis O’Neill

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) we should be wary as host defense pathways may is upregulated by TLR4/9, targets p300 to down- also be compromised [32] . regulate mRNA transcription [39]. miR-19, an A recent study suggested the role of miR-146a LPS-responsive miRNA, suppresses levels of the and miR-155 in modulating the immune re- nuclear receptor corepressor silencing mediator sponse in neonatal versus adult cells [36] . miR- for retinoid and thyroid hormone receptor. This 146a and miR-155 are higher in plasmacytoid allows for enhanced NF-␬ B transactivation of in- dendritic cells from human umbilical cord blood flammatory genes [40] . versus adult cells. This may explain the impaired immune response of a neonate to pathogens, and also illustrates the interconnection between Conclusions miRNAs, innate immunity and development. Another miRNA worth mentioning is miR-21. microRNAs have become an exciting and grow- miR-21, like miR-146a and -155, is upregulated ing area of discovery for TLR biologists. These with TLR4 signaling. However, it feeds back on TLR-expressed miRNAs control many aspects of TLR signaling through a slightly more indirect TLR signaling from signaling within the cyto- mechanism. miR-21 targets the mRNA encoding plasm, to transcription factor activation of genes. programmed cell death 4 (PDCD4) that is a Most profiling screens to identify novel translational inhibitor of IL-10 [37] . With lower miRNAs within the TLR pathway have been per- levels of PDCD4, IL-10 is upregulated and can formed in vitro and with high concentrations of inhibit mir-155, which allows SHIP1 to repress TLR activators. Therefore, the possibility exists TLR4 [13] . More recently, it has been reported that a number of miRNAs that regulate TLRs in that decorin, a small proteoglycan with anti-tu- vivo have yet to be discovered. miRNA expres- morigenic properties, signals through TLR2/4, sion is quite cell specific; therefore, methods upregulates PDCD4 and downregulates miR-21 which can isolate pure populations of cells such to produce a proinflammatory, antitumorigenic as laser capture technology or cell sorting may environment [38] . be required. Knocking down or overexpressing The ability of miRNA to control TLR signal- miRNAs for therapeutic inflammatory control is ing also extends into the nucleus. Transcription currently under evaluation. As delivery systems factors are controlled by a number of positive and for oligonucleotides and indeed the chemistry negative regulators. behind oligonucleotide generation develop, ther- The histone acetyltransferase, p300, which ex- apeutic products that manipulate these miRNAs poses DNA by acetylating histone tails, is a ubiq- within the TLR pathway are almost certainly on uitous transcriptional activator. miR-132, which the horizon.

References 1 Lee RC, Feinbaum RL, Ambros V: The 3 Reinhart BJ, Slack FJ, Basson M, Pas- C. elegans heterochronic gene lin-4 quinelli AE, Bettinger JC, Rougvie AE, encodes small RNAs with antisense Horvitz HR, Ruvkun G: The 21-nucleo-

complementarity to lin-14. Cell 1993; tide let-7 RNA regulates developmental

75: 843–854. timing in Caenorhabditis elegans .

2 Wightman B, Ha I, Ruvkun G: Post- Nature 2000; 403: 901–906. transcriptional regulation of the heter- 4 Guo H, Ingolia NT, Weissman JS, Bartel ochronic gene lin-14 by lin-4 mediates DP: Mammalian microRNAs predomi- temporal pattern formation in C. nantly act to decrease target mRNA

elegans . Cell 1993; 75: 855–862. levels. Nature 2000; 466: 835–840.

miRNAs That Shape the Innate Immune System 77

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) 5 Vasudevan S, Tong Y, Steitz JA: Switch- 17 Perry MM, Moschos SA, Williams AE, 27 Tang B, Xiao B, Liu Z, Li N, Zhu ED, Li ing from repression to activation: mi- Shepherd NJ, Larner-Svensson HM, BS, Xie QH, Zhuang Y, Zou QM, Mao croRNAs can up-regulate translation. Lindsay MA: Rapid changes in micro- XH: Identification of MyD88 as a novel

Science 2007; 318: 1931–1934. RNA-146a expression negatively regu- target of miR-155, involved in negative 6 Place RF, Li LC, Pookot D, Noonan EJ, late the IL-1beta-induced inflamma- regulation of Helicobacter pylori-in-

Dahiya R: MicroRNA-373 induces ex- tory response in human lung alveolar duced inflammation. FEBS Lett 2010;

pression of genes with complementary epithelial cells. J Immunol 2008; 180: 584: 1481–1486. promoter sequences. Proc Natl Acad Sci 5689–5698. 28 Ceppi M, Pereira PM, Dunand-Sauthier

USA 2008; 105: 1608–1613. 18 Okuhara A, Nakasa T, Shibuya H, Ni- I, Barras E, Reith W, Santos MA, Pierre 7 Eiring AM, Harb JG, Neviani P, Garton imoto T, Adachi N, Deie M, Ochi M: P: MicroRNA-155 modulates the inter- C, et al: miR-328 functions as an RNA Changes in microRNA expression in leukin-1 signaling pathway in activated decoy to modulate hnRNP E2 regula- peripheral mononuclear cells according human monocyte-derived dendritic

tion of mRNA translation in leukemic to the progression of osteoarthritis. cells. Proc Natl Acad Sci USA 2009; 106:

blasts. Cell 2010; 140: 652–665. Mod Rheumatol 2012; 22: 446–457. 2735–2740. 8 Filipowicz W, Bhattacharyya SN, 19 Nakasa T, Miyaki S, Okubo A, Hashi- 29 Lu F, Weidmer A, Liu CG, Volinia S, Sonenberg N: Mechanisms of post- moto M, Nishida K, Ochi M, Asahara Croce CM, Lieberman PM: Epstein- transcriptional regulation by micro- H: Expression of microRNA-146 in Barr virus-induced miR-155 attenuates RNAs: are the answers in sight? Nat rheumatoid arthritis synovial tissue. NF-kappaB signaling and stabilizes

Rev Genet 2008; 9: 102–114. Arthritis Rheum 2008; 58: 1284–1292. latent virus persistence. J Virol 2008;

9 Lujambio A, Lowe SW: The microcos- 20 Dai Y, Huang YS, Tang M, Lv TY, Hu 82: 10436–10443.

mos of cancer. Nature 2012; 482: 347– CX, Tan YH, Xu ZM, Yin YB: Microar- 30 Mansell A, Smith R, Doyle SL, Gray P, 355. ray analysis of microRNA expression in Fenner JE, Crack PJ, Nicholson SE, Hil- 10 Zampetaki A, Mayr M: MicroRNAs in peripheral blood cells of systemic lupus ton DJ, O’Neill LA, Hertzog PJ: Sup-

vascular and metabolic disease. Circ erythematosus patients. Lupus 2007; 16: pressor of cytokine signaling 1 nega-

Res 2012; 110: 508–522. 939–946. tively regulates Toll-like receptor 11 Davidson-Moncada J, Papavasiliou FN, 21 Chatzikyriakidou A, Voulgari PV, signaling by mediating Mal degrada-

Tam W: MicroRNAs of the immune Georgiou I, Drosos AA: A polymor- tion. Nat Immunol 2006; 7: 148–155. system: roles in inflammation and can- phism in the 3 -UTR of interleukin-1 31 Lu C, Huang X, Zhang X, Roensch K,

cer. Ann NY Acad Sci 2010; 1183: 183– receptor-associated kinase (IRAK1), a Cao Q, Nakayama KI, Blazar BR, Zeng 194. target gene of miR-146a, is associated Y, Zhou X: miR-221 and miR-155 regu- 12 Sonkoly E, Pivarcsi A: microRNAs in with rheumatoid arthritis susceptibil- late human dendritic cell development,

inflammation. Int Rev Immunol 2009; ity. Joint Bone Spine 2010; 77: 411–413. apoptosis, and IL-12 production

28: 535–561. 22 Chatzikyriakidou A, Voulgari PV, through targeting of p27kip1, KPC1,

13 O’Neill LA, Sheedy FJ, McCoy CE: Mi- Georgiou I, Drosos AA: The role of mi- and SOCS-1. Blood 2011; 117: 4293– croRNAs: the fine-tuners of Toll-like croRNA-146a (miR-146a) and its target 4303. receptor signalling. Nat Rev Immunol IL-1R-associated kinase (IRAK1) in 32 Koch M, Mollenkopf HJ, Klemm U,

2011; 11: 163–175. psoriatic arthritis susceptibility. Scand Meyer TF: Induction of microRNA-155

14 Quinn SR, O’Neill LA: A trio of mi- J Immunol 2010; 71: 382–385. is TLR- and type IV secretion system- croRNAs that control Toll-like receptor 23 Chassin C, Kocur M, Pott J, Duerr CU, dependent in macrophages and inhibits

signalling. Int Immunol 2011; 23: 421– Gutle D, Lotz M, Hornef MW: miR-146a DNA-damage induced apoptosis. Proc

425. mediates protective innate immune Natl Acad Sci USA 2012; 109:E1153– 15 Taganov KD, Boldin MP, Chang KJ, tolerance in the neonate intestine. Cell E1162.

Baltimore D: NF-kappaB-dependent Host Microbe 2010; 8: 358–368. 33 Bala S, Marcos M, Kodys K, Csak T, induction of microRNA miR-146, an 24 Nahid MA, Pauley KM, Satoh M, Chan Catalano D, Mandrekar P, Szabo G: inhibitor targeted to signaling proteins EK: miR-146a is critical for endotoxin- Up-regulation of microRNA-155 in of innate immune responses. Proc Natl induced tolerance: implication in in- macrophages contributes to increased ␣ ␣ Acad Sci USA 2006; 103: 12481–12486. nate immunity. J Biol Chem 2009; 284: tumor necrosis factor (TNF ) pro- 16 Hou J, Wang P, Lin L, Liu X, Ma F, An 34590–34599. duction via increased mRNA half-life H, Wang Z, Cao X: MicroRNA-146a 25 Eis PS, Tam W, Sun L, Chadburn A, Li in alcoholic liver disease. J Biol Chem

feedback inhibits RIG-I-dependent Z, Gomez MF, Lund E, Dahlberg JE: Ac- 2011; 286: 1436–1444. Type I IFN production in macrophages cumulation of miR-155 and BIC RNA in 34 McCoy CE, Sheedy FJ, Qualls JE, Doyle by targeting TRAF6, IRAK1, and human B cell lymphomas. Proc Natl SL, Quinn SR, Murray PJ, O’Neill LA:

IRAK2. J Immunol 2009; 183: 2150– Acad Sci USA 2005; 102: 3627–3632. IL-10 inhibits miR-155 induction by

2158. 26 O’Connell RM, Taganov KD, Boldin toll-like receptors. J Biol Chem 2010;

MP, Cheng G, Baltimore D: Micro- 285: 20492–20498. RNA-155 is induced during the macro- phage inflammatory response. Proc

Natl Acad Sci USA 2007; 104: 1604– 1609.

78 Curtis O’Neill

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) 35 Sun Y, Cai J, Ma F, Lu P, Huang H, Zhou 37 Sheedy FJ, Palsson-McDermott E, Hen- 39 Lagos D, Pollara G, Henderson S, Gra- J: miR-155 mediates suppressive effect nessy EJ, Martin C, O’Leary JJ, Ruan Q, trix F, Fabani M, Milne RS, Gotch F, of progesterone on TLR3, TLR4-trig- Johnson DS, Chen Y, O’Neill LA: Nega- Boshoff C: miR-132 regulates antiviral gered immune response. Immunol Lett tive regulation of TLR4 via targeting of innate immunity through suppression

2012; 146: 25–30. the proinflammatory tumor suppressor of the p300 transcriptional co-activa-

36 Charrier E, Cordeiro P, Cordeau M, PDCD4 by the microRNA miR-21. Nat tor. Nat Cell Biol 2010; 12: 513–519.

Dardari R, Michaud A, Harnois M, Immunol 2010; 11: 141–147. 40 Zhou R, Li X, Hu G, Gong AY, Drescher Merindol N, Herblot S, Duval M: Post- 38 Merline R, Moreth K, Beckmann J, KM, Chen XM: miR-16 targets tran- transcriptional down-regulation of Nastase MV, Zeng-Brouwers J, Tralhao scriptional corepressor SMRT and Toll-like receptor signaling pathway in JG, Lemarchand P, Pfeilschifter J, modulates NF-kappaB-regulated trans- umbilical cord blood plasmacytoid Schaefer RM, Iozzo RV, Schaefer L: Sig- activation of interleukin-8 gene. PLoS

dendritic cells. Cell Immunol naling by the matrix proteoglycan One 2012; 7:e30772. 2012;276:114–121. decorin controls inflammation and cancer through PDCD4 and Micro-

RNA-21. Sci Signal 2011; 4:ra75.

Anne M. Curtis Department of Biochemistry and Immunology Trinity Biomedical Science Institute (TBSI) Trinity College Dublin, 152-160 Pearse Street Dublin 2 (Ireland)

E-Mail acurtis @ tcd.ie

miRNAs That Shape the Innate Immune System 79

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 73–79 (DOI: 10.1159/000346535) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508)

Type 2 Diabetes and Islet Inflammation

Marc Y. Donath

Clinic for Endocrinology, Diabetes and Metabolism, University Hospital Basel, Basel , Switzerland

A b s t r a c t mined ability of the ␤ -cell to adapt, and the se- Type 2 diabetes occurs when the pancreatic islet ␤ -cell verity of the resistance to insulin. The reasons for fails to adapt to the increased insulin demand caused by this failure to maintain sufficient insulin secre- obesity-associated insulin resistance. Islets of patients tion are a combined decrease in ␤ -cell mass and with type 2 diabetes display an inflammatory process defective insulin secretion. Several mechanisms characterized by the presence of cytokines, immune have been proposed to explain this failure, in- cells, ␤ -cell apoptosis, amyloid deposits and fibrosis. This cluding endoplasmic reticulum stress, oxidative insulitis is due to a pathological activation of the innate stress, amyloid deposition, lipotoxicity and glu- immune system by metabolic stress and governed by cotoxicity [1–5] . Of note, all these factors may in- IL-1 signaling. Therefore, impaired insulin secretion ob- duce an inflammatory response [6] . Initially, an served in patients with type 2 diabetes can be treated by inflammatory response is probably deployed to anti-inflammatory approaches. promote ␤-cell repair and regeneration. Howev- Copyright © 2013 S. Karger AG, Basel er, as it becomes chronic, the inflammatory pro- cesses may become deleterious. Interestingly, similar inflammatory processes occur in insulin- Major progress has been achieved in our under- sensitive tissues. Therefore, type 2 diabetes can standing of the pathogenesis of type 2 diabetes. be considered an inflammatory disease. Conse- Overnutrition and inactivity promote insulin re- quently, it can be treated by agents blocking this sistance in genetically predisposed individuals. vicious cycle such as IL-1 antagonists, or more To cope with this increased demand of insulin downstream by NF-␬ B modulation via salsalate, secretion, the pancreatic islet ␤-cell needs to or similar compounds. enhance its secretory activity. While successful In this chapter, we will review the evidence for adaptation of the ␤ -cell permits maintenance of insulitis in type 2 diabetes, the mechanisms in- normal metabolism throughout life in most sub- ducing this inflammatory process, its physiolog- jects, this adaptation eventually fails in some in- ical and pathological role and the therapeutic dividuals, depending on the genetically deter- consequences. Evidence for an Islet Inflammatory Process in tients with type 2 diabetes, characterized by amy- ␤-Cell Failure during Type 2 Diabetes loid deposits. Indeed, fibrosis is a hallmark of the end stage of a chronic inflammatory process. At It is now evident that impaired insulin secretion the latter, activation of the NLRP3 inflamma- is not only an important etiologic factor in the some by islet amyloid polypeptide may enhance pathogenesis of the disease, but also the driving IL-1␤ production [16, 17] . Once IL-1␤ is upregu- force dictating the dynamics of the disease. In- lated in islets of patients with type 2 diabetes [9, deed, while insulin resistance is present at all 14] , it may induce numerous cytokines and also stages from pre-diabetes to overt diabetes, it re- chemokines [18, 19] . Thereby, it contributes to the mains constant in a single individual as long as recruitment of immune cells implementing a his bodyweight remains unchanged. In contrast, broad inflammatory response. Of note is that IL- the onset of diabetes and its progression is largely 1 ␤ will also induce itself in ␤ -cells, engendering determined by the progressive failure of the pan- a vicious cycle [14] . Thus, IL-1-driven insulitis creatic islet. At a pre-diabetic stage, insulin pro- may be considered as an integral component of duction will increase to adapt to the enhanced the pathology observed in type 2 diabetes. demand. When this adaptation fails, diabetes oc- curs. This failure then continues to progress, and affected individuals will require increasing anti- Regulation of IL-1 ␤ Expression by Nutrients diabetic treatment, until eventually all type 2 di- in Human Islets abetics require exogenous insulin to control their glycemia. In an attempt to understand the un- The first demonstration that a cell nutrient can derlying cause of this progressive failure, several directly induce an inflammatory process was the mechanisms have been described. It appears that observation that a high glucose concentration in- in all instances they constitute various compo- duces IL-1␤ release in human islets [9] , with sim- nents of an inflammatory process. The first evi- ilar observations following in rat islets [20] , reti- dence for an inflammatory process in the pancre- nal cells [21] and human monocytes [22] . Mod- atic islet arose from the observation that hyper- erately elevated glucose concentrations (11 mM ) glycemia induces ␤-cell apoptosis and impaired were sufficient to induce transcriptional activa- function via glucose-induced IL-1 ␤ and subse- tion of IL-1 ␤ expression in human islets [14] . quent activation of Fas [7–10]. Similarly, recent More recently, it was also found that long- evidence shows that fatty acids also promote an chain free fatty acids (FFAs) induce several cyto- inflammatory response [11, 12] . More down- and chemokines in human and rodent islets [11] . stream, endoplasmic reticulum stress and oxida- Oleate, palmitate and stearate, which are the tive stress also appear linked to inflammatory most abundant FFAs in human nutrition, and events. In support of insulitis in type 2 diabetes, hence in the circulation, stimulate IL-1 ␤ expres- elevated numbers of immune cells have been de- sion when added individually or as mixtures. tected in islets of patients with type 2 diabetes in Similarly, FFAs induced the IL-1-dependent conjunction with increased levels of cytokines cyto- and chemokines IL-6 and IL-8 in human and chemokines [12–14] . Of note, every animal islets, and CXCL1 (also known as chemokine KC) model of diabetes investigated to date displays in mouse islets. These stimulatory effects of FFAs islet immune cell infiltration [12, 15] . Further- on proinflammatory mediators are not restricted more, a strong argument for the occurrence of an to islet cells, but were also observed in numerous inflammatory process in islets is the well-de- other cell types such as muscle [23], macrophage scribed fibrosis observed in tissue sections of pa- and adipocyte cell lines [24–26] , and in coronary

Type 2 Diabetes and Islet Inflammation 81

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508) artery endothelial cells [27] . The combination of els of type 2 diabetes have now been character- FFAs with elevated glucose concentrations fur- ized to various degrees regarding the presence of ther increases IL-1 ␤ expression [11] and the re- an islet inflammatory signature, including the lease of various cyto- and chemokines [12] . GK rat, the high-fat diet-fed mouse, the db/db mouse, the Psammomys obesus gerbil, the fatty Zucker rat, and the Cohen rat [12, 15, 39, 40] . In The Inflammasome as a Sensor of Metabolic a number of these models, increased islet IL-1␤ Stress expression, and increased numbers of islet mac- rophages are an underlying feature. An impor- The question arises on how metabolic stress is tant property of IL-1 ␤ in inflammatory diseases sensed by the IL-1 system. The inflammasome is is to increase the local expression of chemokines a key factor mediating activation of innate im- and adhesion molecules [18] . Accordingly, treat- munity [28] . Interestingly, the inflammasome ment of an animal model of diabetes with IL-1Ra can be activated by a variety of metabolic distur- reduced peripheral and islet inflammation along bance including uric acid during gout arthritis with improved insulin sensitivity and insulin se- [29] and cholesterol in atherogenesis [30] . In the cretion [41] . context of diabetes, it has been shown that glu- cose and human islet amyloid polypeptide acti- vate the inflammasome in pancreatic islets [16, Role of Insulitis in Obesity and Type 2 17, 31] and LPS, FFAs and ceramides in adipose Diabetes: From Islet Adaptation to Failure tissues [32–34] . Recently, it was shown that lack of the NLPR3 inflammasome protects from the Inflammation is not in itself a disease, but a man- development of islet fibrosis in mice on a high-fat ifestation of a disease. Initially, it has beneficial diet [35] . All of this assigns to the inflammasome effects such as preventing spread of infection or a role as a sensor of metabolic stress. promoting regeneration. However, if prolonged Thus, unraveling the molecular mechanisms or excessive, it may exacerbate disease by tissue underlying induction of an inflammatory re- destruction. It is likely that in the case of islet in- sponse during obesity and diabetes assigns to the flammation in type 2 diabetes similar phenom- innate immune system a new role in detecting ena occur. Indeed, the endocrine pancreas has a metabolic dangers. The inflammasome appears remarkable capacity to adapt to conditions of in- as the sentinel sensing metabolic stress and creased insulin demand as encountered in obe- alarming the immune defense in pancreatic is- sity and pregnancy by increasing its functional lets, insulin-sensitive tissues and blood vessels. mass. This may be triggered by limited hypergly- cemic events which would provoke ␤-cell pro- duction of very low concentrations of IL-1 ␤ fol- IL-1 Is a Master Regulator of Tissue and Islet lowed by Fas upregulation [8, 42]. At low concen- Inflammation in Type 2 Diabetes trations of IL-1 ␤ and in the presence of FLIP, Fas engagement would lead to ␤ -cell proliferation Tissue inflammation is increasingly recognized and enhanced function via NF-␬ B and PDX1 in the pathology of both obesity-associated insu- [43]. Indeed, NF-␬ B has been reported to have lin resistance, and ␤ -cell failure in human type 2 beneficial effects on both insulin secretion and diabetes [36, 37] . In all cases, this inflammation ␤-cell proliferation [44, 45]. If this response be- is characterized by increased tissue-infiltrating comes excessive due to prolonged or repetitive macrophages [12, 38]. A number of animal mod- exposure to nutrients, this initially adaptive pro-

82 Donath

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508) cess may become deleterious. IL-1␤ will then de- Clinical Validation and Therapeutic crease FLIP and Fas engagement and switch to Consequences promote deleterious effects. In addition, IL-1␤ initiates the vicious cycle of inducing itself and Based on the above-described predominant role promoting an enhanced release of chemokines, of IL-1 ␤ , clinical trials of IL-1 antagonism in type which leads to recruitment of macrophages [14] . 2 diabetes were initiated. In a proof-of-concept Possibly, these macrophages then produce a high study, the naturally occurring antagonist of IL- amount of IL-1␤ [39] and other cytotoxic factors 1 ␤ IL-1Ra was tested in a placebo-controlled which then impair the function of the ␤ -cell. It is study of 70 patients [46] . At 13 weeks, glycated important to note that these mechanisms may be hemoglobin was significantly improved due to responsible for both a decrease in ␤ -cell mass and enhanced ␤ -cell secretory function. Remarkably, impaired function. Depending on duration and the improvement promoted by IL-1 blockade magnitude of the effect as well as the individual lasted for at least 39 weeks following treatment regenerative capacity, the functional impairment withdrawal [47], thus reflecting the disease-mod- may predominate over the more definitive de- ifying potential of this therapy. Similar results crease in ␤ -cell mass. Interestingly, such an in- were obtained in follow-up studies [48–51] and by flammatory process occurs also in the insulin- modulation of NF-␬ B via salsalate [52–54] . Rath- sensitive tissues and in end organs such as kid- er than just palliating hyperglycemia, anti-in- ney, eye and the vasculature. Therefore, tissue flammatory treatment may represent a novel inflammation may not only underlie ␤ -cell fail- treatment principle directed against the patho- ure and insulin resistance but also participate di- genesis that underlies diabetes mellitus, whereby rectly in the complications of diabetes such as ne- the progressive decline in functional ␤-cell mass phropathy, retinopathy and cardiovascular dis- could be prevented or even reversed. Due to the ease. autoinflammatory nature of the whole metabolic syndrome and based on preclinical studies, it is expected that this will also enhance insulin sen- sitivity and prevent complications like blindness, cardiovascular events and nephropathy. Ongoing clinical trials are in phase 3.

References

1 Weir GC, Bonner-Weir S: Five stages of 4 Hull RL, Westermark GT, Westermark 7 Donath MY, Gross DJ, Cerasi E, Kaiser evolving beta-cell dysfunction during P, Kahn SE: Islet amyloid: a critical en- N: Hyperglycemia-induced beta-cell

progression to diabetes. Diabetes 2004; tity in the pathogenesis of type 2 diabe- apoptosis in pancreatic islets of Psam-

53(suppl 3):S16–S21. tes. J Clin Endocrinol Metab 2004; 89: momys obesus during development of

2 Prentki M, Nolan CJ: Islet beta cell fail- 3629–3643. diabetes. Diabetes 1999; 48: 738–744. ure in type 2 diabetes. J Clin Invest 5 Harding HP, Ron D: Endoplasmic re- 8 Maedler K, Fontana A, Ris F, Sergeev P,

2006; 116: 1802–1812. ticulum stress and the development of Toso C, Oberholzer J, Lehmann R,

3 Robertson RP, Harmon J, Tran PO, diabetes: a review. Diabetes 2002; Bachmann F, Tasinato A, Spinas GA, Poitout V: Beta-cell glucose toxicity, 51(suppl 3):S455–S461. Halban PA, Donath MY: FLIP switches lipotoxicity, and chronic oxidative 6 Donath MY, Shoelson SE: Type 2 diabe- Fas-mediated glucose signaling in hu- ␤ stress in type 2 diabetes. Diabetes 2004; tes as an inflammatory disease. Nat man pancreatic cells from apoptosis

53(suppl 1):S119–S124. Rev Immunol 2011; 11: 98–107. to cell replication. Proc Natl Acad Sci

USA 2002; 99: 8236–8241.

Type 2 Diabetes and Islet Inflammation 83

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508) 9 Maedler K, Sergeev P, Ris F, Oberholzer 17 Westwell-Roper C, Dai DL, 27 Staiger H, Staiger K, Stefan N, Wahl J, Joller-Jemelka HI, Spinas GA, Kaiser Soukhatcheva G, Potter KJ, van Rooijen HG, Machicao F, Kellerer M, Haring N, Halban PA, Donath MY: Glucose- N, Ehses JA, Verchere CB: IL-1 blockade HU: Palmitate-induced interleukin-6 induced beta-cell production of inter- attenuates islet amyloid polypeptide- expression in human coronary artery

leukin-1beta contributes to glucotoxic- induced proinflammatory cytokine endothelial cells. Diabetes 2004; 53: ity in human pancreatic islets. J Clin release and pancreatic islet graft dys- 3209–3216.

Invest 2002; 110: 851–860. function. J Immunol 2011; 187: 2755– 28 Martinon F, Tschopp J: Inflammatory 10 Maedler K, Spinas GA, Lehmann R, 2765. caspases and inflammasomes: master Sergeev P, Weber M, Fontana A, Kaiser 18 Dinarello CA: Immunological and in- switches of inflammation. Cell Death

N, Donath MY: Glucose induces beta- flammatory functions of the interleu- Differ 2007; 14: 10–22.

cell apoptosis via upregulation of the kin-1 family. Annu Rev Immunol 2009; 29 Martinon F, Petrilli V, Mayor A, Tardi-

Fas-receptor in human islets. Diabetes 27: 519–550. vel A, Tschopp J: Gout-associated uric

2001; 50: 1683–1690. 19 Ehses JA, lacraz G, Giroix MH, Schmid- acid crystals activate the NALP3 in-

11 Boni-Schnetzler M, Boller S, Debray S, lin F, Coulaud J, Kassis N, Irminger JC, flammasome. Nature 2006; 440: 237– Bouzakri K, Meier DT, Prazak R, Kerr- Kergoat M, Portha B, Homo-Delarche 241. Conte J, Pattou F, Ehses JA, Schuit FC, F, Donath MY: IL-1 antagonism reduces 30 Duewell P, Kono H, Rayner KJ, Sirois Donath MY: Free fatty acids induce a hyperglycemia and tissue inflamma- CM, Vladimer G, Bauernfeind FG, Abe- proinflammatory response in islets via tion in the type 2 diabetic GK rat. Proc la GS, Franchi L, Nunez G, Schnurr M,

the abundantly expressed interleukin-1 Natl Acad Sci USA 2009; 106: 13998– Espevik T, Lien E, Fitzgerald KA, Rock

receptor I. Endocrinology 2009; 150: 14003. KL, Moore KJ, Wright SD, Hornung V, 5218–5229. 20 Fei H, Zhao B, Zhao S, Wang Q: Re- Latz E: NLRP3 inflammasomes are 12 Ehses JA, Perren A, Eppler E, Ribaux P, quirements of calcium fluxes and ERK required for atherogenesis and activat-

Pospisilik JA, Maor-Cahn R, Gueripel kinase activation for glucose- and inter- ed by cholesterol crystals. Nature 2010;

X, Ellingsgaard H, Schneider MK, leukin-1beta-induced beta-cell apopto- 464: 1357–1361.

Biollaz G, Fontana A, Reinecke M, Ho- sis. Mol Cell Biochem 2008; 315: 75–84. 31 Zhou R, Tardivel A, Thorens B, Choi I, mo-Delarche F, Donath MY: Increased 21 Yego EC, Vincent JA, Sarthy V, Busik Tschopp J: Thioredoxin-interacting number of islet-associated macro- JV, Mohr S: Differential regulation of protein links oxidative stress to inflam-

phages in type 2 diabetes. Diabetes high glucose-induced glyceraldehyde- masome activation. Nat Immunol 2010;

2007; 56: 2356–2370. 3-phosphate dehydrogenase nuclear 11: 136–140. 13 Richardson SJ, Willcox A, Bone AJ, accumulation in Muller cells by IL-1be- 32 Stienstra R, Joosten LA, Koenen T, van Foulis AK, Morgan NG: Islet-associated ta and IL-6. Invest Ophthalmol Vis Sci Tits B, et al: The inflammasome-medi-

macrophages in type 2 diabetes. Diabe- 2009; 50: 1920–1928. ated caspase-1 activation controls adi-

tologia 2009; 52: 1686–1688. 22 Dasu MR, Devaraj S, Jialal I: High glu- pocyte differentiation and insulin sen-

14 Boni-Schnetzler M, Thorne J, Parnaud cose induces IL-1beta expression in sitivity. Cell Metab 2010; 12: 593–605. G, Marselli L, Ehses JA, Kerr-Conte J, human monocytes: mechanistic in- 33 Vandanmagsar B, Youm YH, Ravussin Pattou F, Halban PA, Weir GC, Donath sights. Am J Physiol Endocrinol Metab A, Galgani JE, Stadler K, Mynatt RL, ␤ MY: Increased interleukin (IL)-1 mes- 2007; 293:E337–E346. Ravussin E, Stephens JM, Dixit VD: senger ribonucleic acid expression in 23 Senn JJ: Toll-like receptor-2 is essential The NLRP3 inflammasome instigates ␤ -cells of individuals with type 2 dia- for the development of palmitate-in- obesity-induced inflammation and ␤ betes and regulation of IL-1 in human duced insulin resistance in myotubes. J insulin resistance. Nat Med 2011; 17:

islets by glucose and autostimulation. J Biol Chem 2006; 281: 26865–26875. 179–188.

Clin Endocrinol Metab 2008; 93: 4065– 24 Shi H, Kokoeva MV, Inouye K, Tzameli 34 Wen H, Gris D, Lei Y, Jha S, Zhang L, 4074. I, Yin H, Flier JS: TLR4 links innate Huang MT, Brickey WJ, Ting JP: Fatty 15 Ehses JA, Ellingsgaard H, Boni- immunity and fatty acid-induced insu- acid-induced NLRP3-ASC inflamma-

Schnetzler M, Donath MY: Pancreatic lin resistance. J Clin Invest 2006; 116: some activation interferes with insulin

islet inflammation in type 2 diabetes: 3015–3025. signaling. Nat Immunol 2011; 12: 408– from alpha and beta cell compensation 25 Nguyen MT, Favelyukis S, Nguyen AK, 415. to dysfunction. Arch Physiol Biochem Reichart D, Scott PA, Jenn A, Liu-Bryan 35 Youm YH, Adijiang A, Vandanmagsar

2009; 115: 240–247. R, Glass CK, Neels JG, Olefsky JM: A B, Burk D, Ravussin A, Dixit VD: Elim- 16 Masters SL, Dunne A, Subramanian SL, subpopulation of macrophages infil- ination of the NLRP3-ASC inflamma- Hull RL, et al: Activation of the NLRP3 trates hypertrophic adipose tissue and some protects against chronic obesity- inflammasome by islet amyloid poly- is activated by free fatty acids via Toll- induced pancreatic damage.

peptide provides a mechanism for en- like receptors 2 and 4 and JNK-depen- Endocrinology 2011; 152: 4039–4045.

hanced IL-1beta in type 2 diabetes. Nat dent pathways. J Biol Chem 2007; 282: 36 Wellen KE, Hotamisligil GS: Inflam-

Immunol 2010; 11: 897–904. 35279–35292. mation, stress, and diabetes. J Clin

26 Song MJ, Kim KH, Yoon JM, Kim JB: Invest 2005; 115: 1111–1119. Activation of Toll-like receptor 4 is as- sociated with insulin resistance in adi- pocytes. Biochem Biophys Res

Commun 2006; 346: 739–745.

84 Donath

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508) 37 Donath MY, Schumann DM, Faulen- 43 Schumann DM, Maedler K, Franklin I, 49 Rissanen A, Howard CP, Botha J, Thu- bach M, Ellingsgaard H, Rütti S, Perren Konrad D, Storling J, Boni-Schnetzler ren T: Effect of anti-IL-1beta antibody A, Böni-Schnetzler M, Ehses JA: Islet M, Gjinovci A, Kurrer MO, Gauthier (canakinumab) on insulin secretion inflammation in type 2 diabetes: from BR, Bosco D, Andres A, Berney T, rates in impaired glucose tolerance or metabolic stress to therapy. Diabetes Greter M, Becher B, Chervonsky AV, type 2 diabetes: results of a random-

Care 2008; 31:S161–S164. Halban PA, Mandrup-Poulsen T, Woll- ized, placebo-controlled trial. Diabetes 38 Weisberg SP, McCann D, Desai M, heim CB, Donath MY: The Fas pathway Obesity Metab 2012, Epub ahead of Rosenbaum M, Leibel RL, Ferrante AW is involved in pancreatic beta cell secre- print. Jr: Obesity is associated with macro- tory function. Proc Natl Acad Sci USA 50 Sloan-Lancaster J, Polzer J, Miller JW,

phage accumulation in adipose tissue. J 2007; 104: 2861–2866. Scherer JC, Berg JK: Safety, tolerability

Clin Invest 2003; 112: 1796–1808. 44 Norlin S, Ahlgren U, Edlund H: Nuclear and efficacy of subcutaneous 39 Weksler-Zangen S, Raz I, Lenzen S, factor-␬ b activity in ␤ -cells is required LY2189102, a neutralizing IL-1b anti- Jorns A, Ehrenfeld S, Amir G, Oprescu for glucose-stimulated insulin secre- body, in patients with type 2 diabetes.

A, Yagil Y, Yagil C, Zangen DH, Kaiser tion. Diabetes 2005; 54: 125–132. Diabetes 2011; 60(suppl 1A):47-LB. N: Impaired glucose-stimulated insulin 45 Hammar EB, Irminger JC, Rickenbach 51 Cavelti-Weder C, Babians-Brunner A, secretion is coupled with exocrine pan- K, Parnaud G, Ribaux P, Bosco D, Keller C, Stahel MA, Kurz-Levin M, creatic lesions in the Cohen diabetic Rouiller DG, Halban PA: Activation of Zayed H, Solinger AM, Mandrup-

rat. Diabetes 2008; 57: 279–287. NF-kappab by extracellular matrix is Poulsen T, Dinarello CA, Donath MY: 40 Li X, Zhang L, Meshinchi S, Dias-Leme involved in spreading and glucose- Effects of gevokizumab on glycemia C, Raffin D, Johnson JD, Treutelaar stimulated insulin secretion of pancre- and inflammatory markers in type 2

MK, Burant CF: Islet microvasculature atic beta cells. J Biol Chem 2005; 280: diabetes. Diabetes Care 2012; 35: 1654– in islet hyperplasia and failure in a 30630–30637. 1662. model of type 2 diabetes. Diabetes 46 Larsen CM, Faulenbach M, Vaag A, 52 Fleischman A, Shoelson SE, Bernier R,

2006; 55: 2965–2973. Volund A, Ehses JA, Seifert B, Man- Goldfine AB: Salsalate improves glyce- 41 Ehses JA, Lacraz G, Giroix MH, drup-Poulsen T, Donath MY: Interleu- mia and inflammatory parameters in Schmidlin F, Coulaud J, Kassis N, kin-1-receptor antagonist in type 2 dia- obese young adults. Diabetes Care

Irminger JC, Kergoat M, Portha B, Ho- betes mellitus. N Engl J Med 2007; 356: 2008; 31: 289–294. mo-Delarche F, Donath MY: IL-1 an- 1517–1526. 53 Goldfine AB, Fonseca V, Jablonski KA, tagonism reduces hyperglycemia and 47 Larsen CM, Faulenbach M, Vaag A, Pyle L, Staten MA, Shoelson SE: The tissue inflammation in the type 2 dia- Ehses JA, Donath MY, Mandrup- effects of salsalate on glycemic control betic GK rat. Proc Natl Acad Sci USA Poulsen T: Sustained effects of interleu- in patients with type 2 diabetes: a ran-

2009; 106: 13998–14003. kin-1 receptor antagonist treatment in domized trial. Ann Intern Med 2010;

42 Maedler K, Schumann DM, Sauter N, type 2 diabetes. Diabetes Care 2009; 32: 152: 346–357. Ellingsgaard H, Bosco D, Baertschiger 1663–1668. 54 Goldfine AB, Silver R, Aldhahi W, Cai R, Iwakura Y, Oberholzer J, Wollheim 48 van Asseldonk EJ, Stienstra R, Koenen D, Tatro E, Lee J, Shoelson SE: Use of CB, Gauthier BR, Donath MY: Low con- TB, Joosten LA, Netea MG, Tack CJ: salsalate to target inflammation in the centration of interleukin-1␤ induces Treatment with Anakinra improves treatment of insulin resistance and

FLICE-inhibitory protein-mediated disposition index but not insulin sensi- type 2 diabetes. Clin Transl Sci 2008; 1: ␤ -cell proliferation in human pancre- tivity in nondiabetic subjects with the 36–43.

atic islets. Diabetes 2006; 55: 2713–2722. metabolic syndrome: a randomized, double-blind, placebo-controlled study.

J Clin Endocrinol Metab 2011; 96: 2119– 2126.

Marc Y. Donath Clinic for Endocrinology, Diabetes and Metabolism University Hospital Basel Petersgraben 4 CH–4031 Basel (Switzerland)

E-Mail marc.donath @ usb.ch

Type 2 Diabetes and Islet Inflammation 85

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 80–85 (DOI: 10.1159/000346508) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 86–90 (DOI: 10.1159/000346509)

The Innate Immune System in Alzheimer’s Disease

Hannes Beckert Annett Halle

Center of Advanced European Studies and Research, Bonn , Germany

A b s t r a c t lar space (senile plaques) or within neurons (neu- Alzheimer’s disease (AD) is a chronic neurodegenerative rofibrillary tangles). disease that afflicts a growing number of patients world- Interestingly, other important diseases, in- wide. It has become increasingly clear that innate im- cluding many neurodegenerative diseases, type 2 mune processes are involved in the development and diabetes and amyloidopathies are also linked to progression of the disease. Here, we discuss innate im- misfolded proteins and are sometimes unifyingly mune mechanisms and the role of the cytokine IL-1 ␤ in called proteinopathies [3]. AD and other diseases with protein misfolding. It has become clear that immune processes are Copyright © 2013 S. Karger AG, Basel an integral part of AD and other neurodegenera- tive diseases. Microglia, as the main immune cells of the brain, increase in number and become Alzheimer’s disease (AD) is the most common morphologically activated in AD. Furthermore, form of dementia. It is associated with a progres- cytokines and the complement system can mod- sive decline of higher cortical functions and ify progression of AD [4]. Thus, it is important to memory. In 2006, around 26.6 million people gain a better molecular understanding of how the worldwide were estimated to suffer from AD. It innate immune system promotes or inhibits neu- has been projected that the absolute number of rodegenerative processes. AD patients will have quadrupled by 2050 [1]. Even though AD was first neuropathological- ly described over 100 years ago [2], there is still Misfolded Proteins in Alzheimer’s Disease considerable debate about its cause. However, the and Other Neurodegenerative Diseases current consensus is that two histological hall- marks – senile plaques and neurofibrillary tan- Postmortem diagnosis of AD is confirmed by the gles – are involved in the pathogenesis of AD. presence of senile plaques and neurofibrillary Both histological alterations consist of misfolded tangles as key histological hallmarks [2]. Extra- and aggregated proteins either in the extracellu- cellular plaques mainly consist of ␤ -amyloid pep- tide (A␤ ) aggregated into fibrils [5, 6]. Another Parkinson’s disease, tauopathies and prion dis- protein aggregate, hyperphosphorylated tau, is eases. The initiation of protein aggregation can responsible for intraneuronal tangle formation result from mutations, conformational changes [7, 8]. According to the amyloid cascade theory within the protein itself or from currently unde- [9], A␤ acts upstream of tau, and numerous stud- fined pathophysiological changes. Formation of ies point towards A ␤ as a causing agent in disease hydrophobic ␤-sheet conformations within the development [10]. relevant proteins (i.e. superoxide dismutase 1, A ␤ is generated upon differential cleavage of SOD1, and TDP-43 in ALS, ␣ -synuclein in Par- the amyloid precursor protein (APP), which is a kinson’s disease, tau in tauopathies and prions in ubiquitously expressed type I transmembrane Creutzfeldt-Jakob disease) often promote aggre- protein. Endoproteolysis of APP by ␣ -secretases gation and insolubility. As in AD, there is a close prevents A␤ generation and therefore represents spatial and temporal connection between pro- the non-amyloidogenic pathway, while ␤-secre- gressive neuronal cell death and the formation of tase activity and subsequent ␥-secretase cleavage these aggregates [3]. produce amyloidogenic A␤ peptides of different lengths [11]. Several mutations in APP, presenil- in-1 and presenilin-2, the two latter are part of Microglia and IL-1 in Alzheimer’s Disease the ␥ -secretase complex, were identified in inher- ited forms of AD and lead to the differentiation Microglia migrate to senile plaques, change their of early-onset, familial versus sporadic AD [12]. morphology and secrete inflammatory cytokines Differential cleavage of intra-membranous [17, 18]. Microglia are the primary phagocytic residues of APP by ␥-secretases produces A ␤ of cells in the brain. They can phagocytose soluble different lengths, the most prominent variants and fibrillar A␤ in vitro [19–21]. Furthermore, consisting of 40 and 42 amino acids [13]. These microglia can internalize A␤ in vivo [22] and re- A ␤ monomers follow an aggregation process that move A ␤ upon vaccination [23]. leads to formation of different soluble and insol- Inhibition of inflammation by anti-inflam- uble A␤ aggregates, e.g. oligomers, protofibrils, matory drugs or immunization against A ␤ de- amyloid-derived diffusible ligands, fibrils, seeds creases senile plaques and delays progression of and plaques. These different conformational AD [24–26]. Indeed, these pharmacological in- states of A ␤ are simultaneously present in brains terventions are currently among the most prom- of AD patients. Similar to other amyloidogenic ising experimental treatment approaches. proteins, this aggregation is a stochastic process, Elevation of the proinflammatory cytokine which results in highly complex mixtures con- IL-1␤ is a typical reaction of the brain to acute taining several of these different species [14]. In injury, and elevated cerebral IL-1 ␤ concentra- vitro studies indicate that oligomers and soluble tions can be found in a number of chronic neuro- aggregates are the most neurotoxic species [15]. degenerative diseases [27]. Microglia are believed Furthermore, a correlation between levels of to represent the main source for IL-1␤ in the oligomeric A␤ species and the degree of cognitive brain. Increased expression of IL-1 ␤ has been de- decline in AD patients has been established [16]. tected in microglia cells surrounding A␤ plaques The association of protein aggregates and mis- in Alzheimer patients, and in animal models of folded proteins with neurodegenerative diseases AD [28, 29]. is a general principle rather than a phenomenon The inflammatory protease caspase 1 is re- unique to AD, as aggregation of proteins is also quired for the cleavage of pro-IL-1 ␤ into the bio- involved in amyotrophic lateral sclerosis (ALS), logically active and secreted mature cytokine.

The Innate Immune System in Alzheimer’s Disease 87

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 86–90 (DOI: 10.1159/000346509) Great progress has been made in the last years in specific trigger for IL-1 ␤ production in type 2 the understanding of how the activity of caspase diabetes has recently been identified with the 1 is controlled [30]. Multi-protein complexes, amyloidogenic islet amyloid polypeptide (IAPP). termed inflammasomes, assemble in the cytosol IAPP is found in amyloid depositions in the pan- upon a variety of stimuli, leading to activation of creas of type 2 diabetes patients. These pancre- caspase 1. Active caspase 1, in turn, catalytically atic amyloid deposits are in several ways compa- activates pro-IL-1 ␤ and triggers the release of the rable to A␤ plaque deposits in brains of AD pa- active cytokine. The NLRP3 inflammasome can tients. As in AD, many lines of evidence link be activated by a broad spectrum of bacterial tox- pancreatic amyloid deposition and IAPP to dis- ins, as well as by crystalline materials such as ease progression [39]. Comparable to A␤ in mi- monosodium uric acid crystals that occur in gout croglial cells, IAPP induces the release of IL-1 ␤ [31–33]. from bone marrow-derived macrophages or den- Since IL-1 ␤ had been linked to AD, we inves- dritic cells through the activation of the NLRP3 tigated whether A␤ can induce the inflamma- inflammasome and caspase 1 [39]. Interestingly, some. We showed that A␤ induces the release of discrimination between oligomeric and fibrillar IL-1␤ and the secretion of neurotoxic factors IAPP revealed IAPP oligomers as the species that from microglial cells through activation of the induced significantly higher levels of IL-1 ␤ secre- NLRP3 inflammasome [34]. This process tion. is dependent on A ␤ phagocytosis and accom- Another amyloidogenic protein that has been panied by lysosomal damage in microglia. Us- linked to caspase 1 activation and IL-1 ␤ release ing specific pharmacological intervention and from macrophages is SOD1. Dominant muta- knockout cells, we showed that the lysosomal tions of SOD1 are the most common monogenet- protease cathepsin B is involved in this cellular ic cause of ALS. This neurodegenerative disease pathway [34, 35]. is characterized by a progressive loss of motor neurons in the motor cortex and the spinal cord [40]. Typical symptoms are progressive paralysis Misfolded Proteins as Activators of the and muscle atrophy. Death occurs most often due Inflammasome to respiratory failure. Mutations of SOD1 cause aggregation and misfolding of the protein, which The precise mechanism of NLRP3 inflamma- may ultimately cause neuronal cell death. Addi- some activation by A␤ remains to be solved, but tionally, similar to AD, there is an inflammatory there may be a common molecular process shared component in ALS. Inflammation mediated by by A ␤ and other amyloidogenic proteins. To date, glial cells has shown to accelerate the course of protein aggregates relevant for ALS, prion diseas- the disease [40]. In cell culture experiments with es and type 2 diabetes have been shown to acti- microglia and macrophages, mutated SOD1 pro- vate caspase 1, IL-1␤ and the inflammasome [36, tein activated caspase 1 and induced the release 37]. of IL-1 ␤ . Interestingly, there was a strong correla- The importance of IL-1␤ for type 2 diabetes tion between the aggregation state of SOD1 and has long been established, and clinical trials with the amount of mature IL-1 ␤ released from cells, IL-1RA, a reagent that blocks IL-1-induced sig- suggesting that fibrillar rather than oligomeric naling, have shown a positive outcome [38]. Fur- SOD1 species are potent caspase 1 and IL-1 ␤ thermore, IL-1 ␤ can induce apoptosis in insulin- activators [40]. Importantly, SOD1 transgenic producing pancreatic ␤ -cells, a typical patholog- mice in which IL-1␤ or caspase 1 had been ical hallmark of type 2 diabetes [39]. A possible deleted and SOD1 transgenic mice treated with

88 Beckert Halle

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 86–90 (DOI: 10.1159/000346509) IL-1RA showed increased survival and delayed duce proinflammatory cytokines, including IL- onset of symptoms, highlighting the functional 1 ␤ . Furthermore, activation of microglia and IL- importance of IL-1 ␤ and caspase 1 in ALS 1 ␤ release was dependent on NLRP3 activation, as progression in vivo. indicated by knock-down experiments [37]. F i n a l l y , P r P Sc, involved in prion disease, has also been shown to activate the inflammasome. Prion diseases are caused by conformational Conclusions changes of the prion protein towards an aggrega- tion-prone ␤-sheet-rich form (PrP Sc). Upon gen- Innate immune processes and in particular IL-1 ␤ eration of this abnormal protein conformation, and inflammasome activation are important in the disease-associated prion protein induces con- AD and other diseases with protein misfolding. version of non-pathological ␣-helix-rich prion Thus, it will be important to further investigate proteins (PrPC ) into the pathological conforma- how amyloidogenic or misfolded proteins acti- tion [37]. This causes an exponential increase in vate the NLRP3 inflammasome. This knowledge PrPSc that is associated with progressive neuronal will be crucial for the development of new treat- cell death. Recent work has shown that microglia ment options for AD and other important dis- are activated during propagation of PrP Sc and pro- eases.

References

1 Brookmeyer R, Johnson E, Ziegler-Gra- 8 Kosik KS, Joachim CL, Selkoe DJ: Mi- 15 Dahlgren KN, Manelli AM, Stine WB ham K, Arrighi HM: Forecasting the crotubule-associated protein tau (tau) Jr, Baker LK, Krafft GA, LaDu MJ: global burden of Alzheimer’s disease. is a major antigenic component of Oligomeric and fibrillar species of am-

Alzheimers Dement 2007; 3: 186–191. paired helical filaments in Alzheimer yloid-beta peptides differentially affect

2 Alzheimer A: Über eine eigenartige disease. Proc Natl Acad Sci USA 1986; neuronal viability. J Biol Chem 2002;

Erkrankung der Hirnrinde. Allg Z 83: 4044–4048. 277: 32046–32053.

Psychiat Psych-Gericht Med 1907; 64: 9 Karran E, Mercken M, De Strooper B: 16 Tomic JL, Pensalfini A, Head E, Glabe 146–148. The amyloid cascade hypothesis for CG: Soluble fibrillar oligomer levels are 3 Jellinger KA: Recent advances in our Alzheimer’s disease: an appraisal for elevated in Alzheimer’s disease brain understanding of neurodegeneration. J the development of therapeutics. Nat and correlate with cognitive dysfunc-

Neural Transm 2009; 116: 1111–1162. Rev Drug Discov 2011; 10: 698–712. tion. Neurobiol Dis 2009; 35: 352–358. 4 Czirr E, Wyss-Coray T: The immunol- 10 Hardy J, Allsop D: Amyloid deposition 17 Itagaki S, McGeer PL, Akiyama H, Zhu ogy of neurodegeneration. J Clin Invest as the central event in the aetiology of S, Selkoe D: Relationship of microglia

2012; 122: 1156–1163. Alzheimer’s disease. Trends Pharmacol and astrocytes to amyloid deposits of

5 Glenner GG, Wong CW, Quaranta V, Sci 1991; 12: 383–388. Alzheimer disease. J Neuroimmunol

Eanes ED: The amyloid deposits in 11 LaFerla FM, Green KN, Oddo S: Intra- 1989; 24: 173–182. Alzheimer’s disease: their nature and cellular amyloid-beta in Alzheimer’s 18 Frautschy SA, Yang F, Irrizarry M, Hy-

pathogenesis. Appl Pathol 1984; 2: 357– disease. Nat Rev Neurosci 2007; 8: 499– man B, Saido TC, Hsiao K, Cole GM: 369. 509. Microglial response to amyloid plaques 6 Masters CL, Simms G, Weinman NA, 12 St George-Hyslop PH, Petit A: Molecu- in APPsw transgenic mice. Am J Pathol

Multhaup G, McDonald BL, Beyreuther lar biology and genetics of Alzheimer’s 1998; 152: 307–317.

K: Amyloid plaque core protein in disease. C R Biol 2005; 328: 119–130. 19 Koenigsknecht J, Landreth G: Microg- Alzheimer disease and Down syn- 13 Zhang H, Ma Q, Zhang YW, Xu H: Pro- lial phagocytosis of fibrillar beta-amy- ␤ drome. Proc Natl Acad Sci USA 1985; teolytic processing of Alzheimer’s - loid through a beta1 integrin-depen-

82: 4245–4249. amyloid precursor protein. J Neuro- dent mechanism. J Neurosci 2004; 24:

7 Ihara Y, Nukina N, Miura R, Ogawara chem 2012; 120(suppl 1):9–21. 9838–9846. M: Phosphorylated Tau protein is inte- 14 Broersen K, Rousseau F, Schymkowitz 20 Mandrekar S, Jiang Q, Lee CY, Koenigs- grated into paired helical filaments in J: The culprit behind amyloid beta pep- knecht-Talboo J, Holtzman DM, Lan-

Alzheimer’s disease. J Biochem 1986; tide related neurotoxicity in Alzhei- dreth GE: Microglia mediate the clear-

99: 1807–1810. mer’s disease: oligomer size or confor- ance of soluble Abeta through fluid

mation? Alzheimers Res Ther 2010; 2: phase macropinocytosis. J Neurosci

12. 2009; 29: 4252–4262.

The Innate Immune System in Alzheimer’s Disease 89

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 86–90 (DOI: 10.1159/000346509) 21 Heneka MT, Nadrigny F, Regen T, Mar- 27 Shaftel SS, Griffin WS, O’Banion MK: 35 Hornung V, Bauernfeind F, Halle A, tinez-Hernandez A, Dumitrescu- The role of interleukin-1 in neuroin- Samstad EO, Kono H, Rock KL, Fitzger- Ozimek L, Terwel D, Jardanhazi-Kurutz flammation and Alzheimer disease: an ald KA, Latz E: Silica crystals and alu- D, Walter J, Kirchhoff F, Hanisch UK, evolving perspective. J Neuroinflam- minum salts activate the NALP3 in-

Kummer MP: Locus ceruleus controls mation 2008; 5: 7. flammasome through phagosomal

Alzheimer’s disease pathology by mod- 28 Griffin WS, Stanley LC, Ling C, White destabilization. Nat Immunol 2008; 9: ulating microglial functions through L, MacLeod V, Perrot LJ, White CL 3rd, 847–856. norepinephrine. Proc Natl Acad Sci Araoz C: Brain interleukin 1 and S-100 36 Masters SL, O’Neill LAJ: Disease-asso-

USA 2010; 107: 6058–6063. immunoreactivity are elevated in ciated amyloid and misfolded protein 22 Bolmont T, Haiss F, Eicke D, Radde R, Down syndrome and Alzheimer dis- aggregates activate the inflammasome.

Mathis CA, Klunk WE, Kohsaka S, ease. Proc Natl Acad Sci USA 1989; 86: Trends Mol Med 2011; 17: 276–282. Jucker M, Calhoun ME: Dynamics of 7611–7615. 37 Shi F, Yang L, Kouadir M, Yang Y, the microglial/amyloid interaction in- 29 Simard AR, Soulet D, Gowing G, Julien Wang J, Zhou X, Yin X, Zhao D: The dicate a role in plaque maintenance. J JP, Rivest S: Bone marrow-derived mi- NALP3 inflammasome is involved in

Neurosci 2008; 28: 4283–4292. croglia play a critical role in restricting neurotoxic prion peptide-induced mi- 23 Nicoll JA, Barton E, Boche D, Neal JW, senile plaque formation in Alzheimer’s croglial activation. J Neuroinflamm

Ferrer I, Thompson P, Vlachouli C, disease. Neuron 2006; 49: 489–502. 2012; 9: 73. Wilkinson D, Bayer A, Games D, Seu- 30 Rathinam VA, Vanaja SK, Fitzgerald 38 Larsen CM, Faulenbach M, Vaag A, bert P, Schenk D, Holmes C: Abeta spe- KA: Regulation of inflammasome sig- Vølund A, Ehses JA, Seifert B, Man-

cies removal after abeta42 immuniza- naling. Nat Immunol 2012; 13: 333–332. drup-Poulsen T, Donath MY: Interleu-

tion. J Neuropathol Exp Neurol 2006; 31 Martinon F, Petrilli V, Mayor A, Tardi- kin-1-receptor antagonist in type 2 dia-

65: 1040–1048. vel A, Tschopp J: Gout-associated uric betes mellitus. N Engl J Med 2007; 356: 24 Schenk D, Barbour R, Dunn W, Gordon acid crystals activate the NALP3 in- 1517–1526.

G, et al: Immunization with amyloid- flammasome. Nature 2006; 440: 237– 39 Masters SL, Dunne A, Subramanian SL, beta attenuates Alzheimer-disease-like 241. Hull RL, et al: Activation of the NLRP3 pathology in the PDAPP mouse. Nature 32 Pope RM, Tschopp J: The role of inter- inflammasome by islet amyloid poly-

1999; 400: 173–177. leukin-1 and the inflammasome in peptide provides a mechanism for en- 25 Weggen S, Eriksen JL, Das P, Sagi SA, et gout: implications for therapy. Arthri- hanced IL-1beta in type 2 diabetes. Nat

al: A subset of NSAIDs lower amyloido- tis Rheum 2007; 56: 3183–3188. Immunol 2010; 11: 897–904. genic Abeta42 independently of cyclo- 33 So A, De Smedt T, Revaz S, Tschopp J: A 40 Meissner F, Molawi K, Zychlinsky A:

oxygenase activity. Nature 2001; 414: pilot study of IL-1 inhibition by Mutant superoxide dismutase 1-in- 212–216. anakinra in acute gout. Arthritis Res duced IL-1beta accelerates ALS patho-

26 Weiner HL, Frenkel D: Immunology Ther 2007; 9:R28. genesis. Proc Natl Acad Sci USA 2010;

and immunotherapy of Alzheimer’s 34 Halle A, Hornung V, Petzold GC, Stew- 107: 13046–13050.

disease. Nat Rev Immunol 2006; 6: 404– art CR, Monks BG, Reinheckel T, 416. Fitzgerald KA, Latz E, Moore KJ, Go- lenbock DT: The NALP3 inflamma- some is involved in the innate immune response to amyloid-beta. Nat Immu-

nol 2008; 9: 857–865.

Annett Halle Center of Advanced European Studies and Research (caesar) Ludwig-Erhard-Allee 2 DE–53175 Bonn (Germany)

E-Mail annett.halle @ caesar.de

90 Beckert Halle

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 86–90 (DOI: 10.1159/000346509) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 91–95 (DOI: 10.1159/000346505)

Role of Inflammasomes in Obesity

Vishwa Deep Dixit

Immunobiology Laboratory, Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, La., USA

A b s t r a c t It is recognized that activation of adipose tissue The adaptive expansion of adipose tissue during energy macrophages (ATMs) and T cells within adipose excess involves significant tissue remodeling and angio- depots is coupled with the development of obesi- genesis that requires controlled local proinflammatory ty-induced proinflammatory state and insulin responses. However, sustained adipose tissue inflamma- resistance [1–3] . The activation of classically acti- tion associated with chronic obesity adversely impacts vated M1 macrophages at the expense of anti- adipose tissue function and is a major instigator of meta- inflammatory alternative activated M2 macro- bolic diseases such as type 2 diabetes. Despite the ab- phages has been causally linked to the devel- sence of overt infection in obesity, the adipose tissue of opment of adipose tissue inflammation and met- overweight and obese patients is infiltrated with a sub- abolic syndrome [4, 5]. The capacity of adipo- stantially high frequency of cells of hematopoietic lin- cytes to expand and store excess energy as lipids eage. A large body of evidence suggests that activated is a critical adaptation during overnutrition [6] . innate immune cell subsets chiefly contribute to the per- Apart from adipocytes, adipose tissue is com- sistent state of inflammation and maybe a common posed of stromal-vascular fraction (SVF) cells mechanism of several obesity-associated chronic diseas- and leukocytes [2] . Adipose tissue expansion in es. It has been suggested that lipid fluxes in the adipose obesity is associated with an increase in macro- tissue and the presence of damage-associated molecular phages, neutrophils, T cells, B cells, and mast patterns serve as a potential trigger for macrophage in- cells in adipose tissue [7–10] . Thus ‘adipose leu- filtration and activation. Emerging evidence shows that kocytosis’ or increased infiltration of immune inflammasomes, the multiprotein platforms that control cell subsets into the adipose tissue of obese pa- the activation and secretion of IL-1␤ and IL-18 in adipose tients represents an important link between tissue, impact the adipocyte function and T cell activa- adaptive adipose tissue remodeling in response tion. This chapter discusses the mechanisms that regu- to energy excess and the emergence of chronic late inflammasome activation in obesity and dietary and inflammation-associated diseases [2] . pharmacological approaches to control immune-meta- The enzymatic dispersion of adipose tissue bolic interactions that link sustained adiposity to chron- and subsequent processing yields adipocytes and ic diseases. Copyright © 2013 S. Karger AG, Basel SVF which contains substantial immune cells [2] . The adipose depots can contain up to 5 million with the key role of the NLRP3 inflammasome in SVF cells/g fat tissue, and approximately 50–65% caspase 1 activation and IL1␤ release, the gain of of these cells are leukocytes [2] . Importantly, in function mutations within the NACHT domain severe obesity in humans, the total fat mass can of Nlrp3 are associated with autoinflammatory constitute up to 50% of the total body mass. disorders in humans that are characterized by Therefore, the expanded adipose tissue repre- periodic fever and inflammation without any in- sents a large immunological organ with distinct fection [17] . leukocyte populations that mediate immune- Interestingly, multiple groups have recently metabolic crosstalk. Among innate immune cell found important roles for inflammasomes in subsets, ATM activation and secretion of proin- both the induction and sustenance of obesity-as- flammatory cytokines such as IL-1␤ and IL-18 is sociated inflammation [11, 12, 15, 16] . Obesity linked with adipose tissue dysfunction [10] . was shown to trigger NLRP3 inflammasome-de- The ATMs sense damage-associated molecu- pendent caspase 1 activation in multiple meta- lar patterns (DAMPs) via pattern-recognition re- bolic organs like adipose tissue and liver [11, 15] . ceptors, such as Toll-like receptors on the cell Importantly, diet-induced obese mice that are de- surface and Nod-like receptors (NLRs) in cytosol ficient in central inflammasome molecules (cas- [2]. The ATMs are activated by fatty acids like pase 1, ASC, NLRP3, and IL-1␤ ) demonstrate im- palmitate- and sphingosine-linked fatty acids proved glucose homeostasis and reduced insulin- like ceramides (DAMPs) and release proinflam- resistance [11, 12, 15, 16]. Interestingly, reduction matory cytokines such as IL-1 ␤ and IL-18 [11, 12] . in obesity-associated metabolic dysfunction in The release of active IL-1␤ /IL-18 from ATMs is inflammasome-deficient mice was associated dependent on autocatalytic activation of procas- with lower M1 activation and increases in M2 pase 1 zymogen into enzymatically active 10- and numbers in adipose depots together with reduced 20-kDa caspase 1 heterodimers [13, 14]. The pres- Th1 response in fat [11] . Furthermore, treatment ence of a large N-terminal protein-protein inter- of obese mice with caspase 1 inhibitors partially action motif called caspase activation recruit- reversed metabolic dysfunction, which suggests ment domain (CARD) is critical for formation the therapeutic potential of inhibiting inflam- of multiprotein scaffolds, wherein caspase 1 un- masomes to reduce obesity-associated diseases dergoes conformational change required for its [15, 16] . cleavage and full activation. The soluble cytosolic Obesity is associated with an increase in sev- proteins that comprise the NLR family are the eral DAMPs; these include extracellular ATP sensing and scaffolding constituents of this mac- from necrotic cells, urate, free cholesterol, free romolecular caspase 1 activation complex called fatty acids like palmitate, and byproducts of fatty an ‘inflammasome’ [14] . Several studies have acid metabolism including ceramides, reactive shown that NLRP3-mediated inflammasome as- oxygen species, glycation end products, and lipid sembly controls adipose tissue inflammation and peroxidation, etc. [2] . Recent data suggest that obesity-associated comorbidities [11, 12, 15, 16] . saturated fatty acids and spingosine-linked fatty The formation of the NLRP3 inflammasome in- acids like ceramides cause NLRP3 inflamma- volves interaction of pyrin domain (PYD) of ASC some activation and may trigger metabolic in- (for apoptosis-associated speck-like protein con- flammation in obesity [11, 12]. Furthermore, re- taining carboxy-terminal CARD) with PYD of cent evidence suggests that endoplasmic retic- NLRP3, and a functional inflammasome com- ulum stress may also trigger inflammasome plex is formed through CARD-CARD interac- assembly [18]. Given that obesity is associated tion of ASC with procaspase 1 [14] . Consistent with defective autophagy, and an increase in ROS

92 Dixit

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 91–95 (DOI: 10.1159/000346505) and endoplasmic reticulum stress, it is possible Nlrp3 is predominantly expressed in cells of my- that such mechanisms may participate in regu- eloid lineage [26] . Other studies reported that lating inflammasome activation through various high Nlrp3 inflammasome in the ‘floating frac- stages of chronic obesity. The broad effects of in- tion’ of primary adipocytes derived from adipose hibition of the NLRP3 inflammasome in obesity tissue may be due to contamination from buoy- are evident by improved insulin signaling in adi- ant lipid-laden macrophages [11]. Thus, the exact pose depots, liver, and skeletal muscle, as well as physiological relevance of inflammasome-de- increased insulin secretion due to reduced ␤ -cell pendent caspase 1 activation in adipocytes re- death in the pancreatic islets [19, 20] . mains to be fully substantiated. The insulin’s binding to the insulin recep- It has also been reported that Asc-deficient tor triggers a signaling cascade via the PI3K- mice develop dysbiosis when fed high-fat diet AKT and the mitogenic MAP kinase-ERK path- [27] . The resultant steatosis and increased adi- ways. In the adipose tissue of diet-induced obese posity in Asc-deficient mice are thought to in- Nlrp3 –/– mice, activation of AKT is increased, in- duce insulin resistance [27] . On the other hand, dicating greater insulin signaling [21]. The genet- other laboratories reported that Asc- and Nlrp3- ic deletion of IL1R and IL1␤ in obese mice and deficient mice fed high-fat diet are protected the resultant loss of IL-1 signaling also enhances from steatosis and insulin resistance [11, 12, 15, adipose tissue insulin sensitivity [22–24] . These 16] . In addition, it has been shown that Nlrp3-, results demonstrate the strong immune-meta- Asc- and caspase 1-null mice have increased en- bolic consequences of NLRP3 inflammasome ac- ergy expenditure which leads to reduced adipos- tivation and IL-1 signaling during obesity. Nota- ity and improved insulin sensitivity [16]. The di- bly, despite strong proinflammatory and Th1- vergent data on adiposity, dysbiosis, and energy inducing effects of inflammasome-dependent expenditure between various laboratories could IL-18, the IL18-deficient mice fed a high-fat diet be due to differences in genetic mouse lines, di- display a higher degree of adiposity, hyperphagia, ets, and animal husbandry conditions. For exam- and insulin-resistance than the control mice [25] . ples, some groups have reported that some, but Further studies are needed to clarify the mecha- not all, Asc-deficient mice lack DOCK2 protein nism of IL-18 in metabolic stress. [28]. In addition, the current caspase 1-deficient The caspase 1 activation has also been report- mouse generated from 129 embryonic stem cells ed to exert direct effects on adipocyte growth, also lacks non-canonical caspase 11 inflamma- differentiation, and metabolism [16] . Interesting- some and c-IAP [29, 30] . Additional careful stud- ly, both human and murine adipocyte cell lines ies are needed to fully characterize this pathway express the caspase 1 protein, and its expression during metabolic stress and obesity. is increased over the course of adipocyte differ- Initial studies in humans suggest that NLRP3 in- entiation [16] . It has been suggested that the flammasome activation in obesity could be im- Nlrp3 inflammasome controls caspase 1 activa- portant in the development and treatment of in- tion in adipocytes and, hence, impairs metabolic sulin resistance and type 2 diabetes. Inhibition of function [16] . Consistent with this, Nlrp3-, Asc-, IL1 signaling by Anakinra has been shown to and caspase 1-deficient mice display reduced fat elicit a significant improvement in type 2 diabe- mass when challenged with the high-fat diet and tes and glycemic control [31]. Furthermore, de- are resistant to obesity [15, 16] . However, stud- velopment of adiposity and insulin resistance in ies from reporter knock-in mice, in which the human adipose tissue was positively correlated Nlrp3 coding sequence was substituted for the with an increased expression of Nlrp3, caspase 1 enhanced green fluorescent protein, found that and IL-18 [11] . In addition, weight-loss-induced

Role of Inflammasomes in Obesity 93

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 91–95 (DOI: 10.1159/000346505) improvement in insulin-resistance in obese dia- mans are required to determine the efficacy of betic humans was found to associate with signif- IL-1␤ signaling inhibitors as a therapeutic strat- icant reduction in IL-1␤ and Nlrp3 expression egy to control type 2 diabetes. The development [11] . Interestingly, the hypoglycemic sulfonylurea of future inflammasome or caspase 1 inhibitors glyburide has been identified to also inhibit the and preclinical testing of their mechanism of ac- NLRP3 inflammasome [32]. Glyburide, however, tion appears to be a promising avenue with po- is not a preferred choice as diabetes treatment due tential for improved treatment strategies for obe- to an increased risk of hypoglycemic episodes in sity-associated chronic diseases. treated patients. Additional studies in obese hu-

References

1 Weisberg SP, McCann D, Desai M, 8 Yang H, Youm YH, Vandanmagsar B, 15 Stienstra R, van Diepen JA, Tack CJ, Rosenbaum M, Leibel RL, Ferrante AW Ravussin A, Gimble JM, Greenway F, Zaki MH, et al: Inflammasome is a cen- Jr: Obesity is associated with macro- Stephens JM, Mynatt RL, Dixit VD: tral player in the induction of obesity phage accumulation in adipose tissue. J Obesity increases the production of and insulin resistance. Proc Natl Acad

Clin Invest 2003; 112: 1796–1808. proinflammatory mediators from adi- Sci USA 2011; 108: 15324–15329. 2 Kanneganti TD, Dixit VD: Immunolog- pose tissue T cells and compromises 16 Stienstra R, Joosten LA, Koenen T, van ical complications of obesity. Nat Im- TCR repertoire diversity: implications Tits B, et al: The inflammasome-medi-

munol 2012; 13: 707–712. for systemic inflammation and insulin ated caspase-1 activation controls adi-

3 Feuerer M, Herrero L, Cipolletta D, resistance. J Immunol 2010; 185: 1836– pocyte differentiation and insulin sen-

Naaz A, Wong J, Nayer A, Lee J, Gold- 1845. sitivity. Cell Metab 2010; 12: 593–605. fine AB, Benoist C, Shoelson S, Mathis 9 Wu D, Molofsky AB, Liang HE, Ricar- 17 Aksentijevich I, Kastner DL: Genetics D: Lean, but not obese, fat is enriched do-Gonzalez RR, Jouihan HA, Bando of monogenic autoinflammatory dis- for a unique population of regulatory T JK, Chawla A, Locksley RM: Eosino- eases: past successes, future challeng-

cells that affect metabolic parameters. phils sustain adipose alternatively acti- es. Nat Rev Rheumatol 2011; 7: 469–

Nat Med 2009; 15: 930–939. vated macrophages associated with 478.

4 Lumeng CN, Bodzin JL, Saltiel AR: glucose homeostasis. Science 2011; 332: 18 Menu P, Mayor A, Zhou R, Tardivel A, Obesity induces a phenotypic switch in 243–247. Ichijo H, Mori K, Tschopp J: ER stress adipose tissue macrophage polariza- 10 Hotamisligil GS: Inflammation and activates the NLRP3 inflammasome

tion. J Clin Invest 2007; 117: 175–184. metabolic disorders. Nature 2006; 444: via an UPR-independent pathway. Cell

5 Nguyen KD, Qiu Y, Cui X, Goh YP, 860–867. Death Dis 2012; 3:e261. Mwangi J, David T, Mukundan L, 11 Vandanmagsar B, Youm YH, Ravussin 19 Masters SL, Dunne A, Subramanian SL, Brombacher F, Locksley RM, Chawla A: A, Galgani JE, Stadler K, Mynatt RL, Hull RL, et al: Activation of the NLRP3 Alternatively activated macrophages Ravussin E, Stephens JM, Dixit VD: inflammasome by islet amyloid poly- produce catecholamines to sustain The NLRP3 inflammasome instigates peptide provides a mechanism for en-

adaptive thermogenesis. Nature 2011; obesity-induced inflammation and hanced IL-1beta in type 2 diabetes. Nat

480: 104–108. insulin resistance. Nat Med 2011; 17: Immunol 2010; 11: 897–904. 6 Spiegelman BM, Flier JS: Obesity and 179–188. 20 Youm YH, Adijiang A, Vandanmagsar the regulation of energy balance. Cell 12 Wen H, Gris D, Lei Y, Jha S, Zhang L, B, Burk D, Ravussin A, Dixit VD: Elim-

2001; 104: 531–543. Huang MT, Brickey WJ, Ting JP: Fatty ination of the NLRP3-ASC inflamma- 7 Winer DA, Winer S, Shen L, Wadia PP, acid-induced NLRP3-ASC inflamma- some protects against chronic obesity- et al: B cells promote insulin resistance some activation interferes with insulin induced pancreatic damage.

through modulation of T cells and pro- signaling. Nat Immunol 2011; 12: 408– Endocrinology 2011; 152: 4039–4045. duction of pathogenic IgG antibodies. 415. 21 Virkamäki A, Ueki K, Kahn CR: Pro-

Nat Med 2011; 17: 610–617. 13 Dinarello CA: Interleukin-1 in the tein-protein interaction in insulin sig- pathogenesis and treatment of inflam- naling and the molecular mechanisms

matory diseases. Blood 2011; 117: 3720– of insulin resistance. J Clin Invest 1999;

3732. 103: 931–943. 14 Martinon F, Mayor A, Tschopp J: The inflammasomes: guardians of the body.

Annu Rev Immunol 2009; 27: 229–265.

94 Dixit

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 91–95 (DOI: 10.1159/000346505) 22 McGillicuddy FC, Harford KA, Reyn- 25 Netea MG, Joosten LA, Lewis E, Jensen 29 Kayagaki N, Warming S, Lamkanfi M, olds CM, Oliver E, Claessens M, Mills DR: Deficiency of interleukin-18 in Vande Walle L, et al: Non-canonical KH, Roche HM: Lack of interleukin-1 mice leads to hyperphagia, obesity and inflammasome activation targets cas-

receptor I (IL-1RI) protects mice from insulin resistance. Nat Med 2006; 12: pase-11. Nature 2011; 479: 117–121. high-fat diet-induced adipose tissue 650–656. 30 Kenneth CS, Younger JM, Hughes ED, inflammation coincident with im- 26 Guarda G, Zenger M, Yazdi AS, Schro- Marcotte D, Barker PA, Saunders TL, proved glucose homeostasis. Diabetes der K, Ferrero I, Menu P, Tardivel A, Duckett CS: An inactivating caspase 11

2011; 60: 1688–1698. Mattmann C, Tschopp J: Differential passenger mutation originating from 23 Jager J, Grémeaux T, Cormont M, Le expression of NLRP3 among hemato- the 129 murine strain in mice targeted

Marchand-Brustel Y, Tanti JF: Inter- poietic cells. J Immunol 2011; 186: 2529– for c-IAP1. Biochem J 2012; 443: 355– leukin-1beta-induced insulin resis- 2534. 359. tance in adipocytes through down-reg- 27 Henao-Mejia J, Elinav E, Jin C, Hao L, 31 Larsen CM, Faulenbach M, Vaag A, ulation of insulin receptor substrate-1 et al: Inflammasome-mediated dysbio- Vølund A, Ehses JA, Seifert B, Man-

expression. Endocrinology 2007; 148: sis regulates progression of NAFLD and drup-Poulsen T, Donath MY: Interleu-

241–251. obesity. Nature 2012; 482: 179–185. kin-1-receptor antagonist in type 2 dia-

24 Lagathu C, Yvan-Charvet L, Bastard JP, 28 Ippagunta SK, Malireddi RK, Shaw PJ, betes mellitus. N Engl J Med 2007; 356: Maachi M, Quignard-Boulangé A, Ca- Neale GA, Walle LV, Fukui Y, Green 1517–1526. peau J, Caron M: Long-term treatment DR, Lamkanfi M, Kanneganti TD: Ad- 32 Lamkanfi M, Mueller JL, Vitari AC, with interleukin-1beta induces insulin dendum: defective Dock2 expression in Misaghi S, Fedorova A, Deshayes K, Lee resistance in murine and human adipo- a subset of ASC-deficient mouse lines. WP, Hoffman HM, Dixit VM: Gly-

cytes. Diabetologia 2006; 49: 2162–2173. Nat Immunol 2012; 13: 701–702. buride inhibits the Cryopyrin/Nalp3

inflammasome. J Cell Biol 2009; 187: 61–70.

Vishwa Deep Dixit, PhD Immunobiology Laboratory Pennington Biomedical Research Center 6400 Perkins Road Baton Rouge, LA 70808 (USA)

E-Mail Vishwa.Dixit @ pbrc.edu

Role of Inflammasomes in Obesity 95

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 91–95 (DOI: 10.1159/000346505) Chapter 4: Innate Immunity and Disease Promotion

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539)

Gut-Brain Communication in the Regulation of System Metabolism

Timo D. Müller Paul T. Pfluger Matthias H. Tschöp

Institute for Diabetes and Obesity (IDO), Helmholtz Centre Munich, Department of Medicine, Technical University of Munich, Munich , Germany; Obesity Research Center, Department of Internal Medicine, University of Cincinnati, Cincinnati, Ohio, USA

Abstract It was in 1841 when the German Physician Ju- Overweight and obesity are major health threats in mod- lius Robert von Mayer (1814–1878) enunciated the ern societies. The regulation of energy metabolism, and ‘first law of thermodynamics’, saying that ‘energy thus also hunger and satiety, is a complex process that can neither be created nor destroyed’. Today, the depends on a constant cross talk of peripheral and cen- law of thermodynamics is still one of the most fun- tral signal mechanisms. To sustain a stable body weight damental basics of energy metabolism control. over time, this network must not only take into account Translated to the human body, it means that the the fuel signals from digestive organs but also long-term calories consumed must match the body’s caloric signals about the energy stored as fat in the body. The demand to sustain a stable body weight over time. aim of this chapter is to summarize the current knowl- Any deviation from this equilibrium will inevita- edge of how the gut-brain axis regulates system metab- bly result in either weight gain or weight loss. As olism. Copyright © 2013 S. Karger AG, Basel the body’s energy demand constantly changes, it seems obvious that food intake constantly needs to be adjusted in order to maintain a stable body weight over time. Importantly, this adjustment of Obesity and its comorbidities such as diabetes, caloric intake cannot only be achieved through the cardiovascular disease and cancer are increasing (short-term) regulation of food intake as the body health threats in Western societies. In light of the also needs to consider (long-term) information worldwide alarming prevalence rate of obesity, about the energy stored as fat. the understanding of the molecular mechanisms The gastrointestinal (GI) tract, as the place of of how the brain regulates hunger and satiety is nutrient absorption, produces a variety of short- currently one of the greatest challenges of mod- term regulators of food intake, which are secreted ern obesity research. either preprandially in anticipation (such as ghrelin) or postprandially in response to incom- ing nutrients [such as cholecystokinin (CCK), signals the GI fuel status from the periphery to PYY, glucagon-like peptide 1 (GLP-1) and oxyn- the CNS in order to adjust energy balance tomodulin (OXM). Together, these peptides sig- through a stimulation of food intake and a de- nal the GI fuel status to the brain in order to ad- crease in energy expenditure. This classical view just food intake and satiety. In contrast to the of a ‘hunger hormone’ was supported by the ob- short-term regulators of food intake, which are servation that plasma levels of ghrelin increase mainly secreted in response or anticipation of preprandially with a peak directly at meal initia- food, long-term adiposity signals (such as leptin tion followed by a postprandial decrease within or insulin) are constantly secreted from periph- the first hour after a meal [5–7] . These days, the eral organs such as the adipose tissue (leptin) or classical view of ghrelin as a ‘hunger’ hormone is the pancreas (insulin) into the circulation and questioned by more recent studies which suggest give information to the brain in proportion to the that ghrelin rather acts as a nutrient sensor that body’s energy stored as fat. Changes in the prepares the CNS for incoming nutrients. Using amount of body fat are thus reflected by changes non-naturally occurring C-7 fatty acids, it was in circulating adiposity signals, and the brain re- shown in these studies that the acyl side chain sponds to these changes by changing its sensitiv- necessary for ghrelin acylation can directly come ity for short-term satiety signals in order to keep from digested nutrients [8] . the body weight at a specific ‘set point’. No matter whether ghrelin is considered a ‘hunger’ hormone or a nutrient sensor, both pe- ripheral and central administration of ghrelin G h r e l i n stimulate food intake and adiposity through stimulation of hypothalamic orexigenic neuro- The GI peptide hormone ghrelin is so far the peptides, such as agouti-related peptide (AgRP) only peripheral hormone with the ability to pro- and neuropeptide Y (Npy), both anabolic neuro- mote body weight gain and adiposity through a peptides that promote a positive energy balance stimulation of food intake while decreasing en- through stimulating food intake while decreas- ergy expenditure and body fat utilization. Ghrel- ing energy expenditure [9] . Accordingly, ghrelin- in was discovered in 1999 as an endogenous li- mediated activation of Ghsr1a entails an in- gand for the growth hormone secretagogue re- creased release of Npy and AgRP in the arcuate ceptor (Ghsr1a). Ghrelin, which is primarily nucleus, which in turn leads to the activation of synthesized and secreted by X/A-like cells in the anabolic downstream pathways that finally lead oxyntic glands of the stomach mucosa [1] pro- to a stimulation of food intake and to a decrease motes its biological action through binding and in energy expenditure [10, 11]. Inhibition of activation of the growth hormone secretagogue AgRP/Npy neurons blunts ghrelin’s effect on receptor (GHSR1a) [2] . To activate GHSR1a, food intake, thus indicating that the orexigenic ghrelin needs to be acylated on its serine 3 resi- effect of ghrelin is mainly mediated over the hy- due, a reaction that is catalyzed by the ghrelin pothalamic melanocortinergic system. O-acyltransferase (GOAT) [3, 4]. The observa- tion that acyl-ghrelin is absent in mice lacking GOAT indicates that GOAT is the only endoge- I n s u l i n nous enzyme capable of activating ghrelin in vivo [4]. Ghrelin is secreted into the circulation Albeit best known for its role in glucose homeo- in response to fasting. Accordingly, ghrelin was stasis, insulin was the first hormone reported to long believed to serve as a ‘hunger’ hormone that control body weight via CNS-dependent mecha-

Gut-Brain Communication in the Regulation of System Metabolism 97

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539) nisms. Accordingly, in the late 1970s, Woods and a meal-terminating factor, blockade of amylin Porte proposed that insulin acts as an adiposity signaling increases food intake, body weight and signal that informs the brain about the amount of adiposity [17] . Amylin’s effect on food intake fat stored in the body and that the brain in re- seems to be mediated over the area postrema sponse to this signal adjusts body weight through (AP), as amylin increases c-fos immunoreactivity an inhibition of food intake. The hypothesis of in this region and as lesions of the AP abolish am- Woods and Porte was based on the observation ylin’s effect on food intake and adiposity [18] . The that plasma level of insulin are under both basal signaling mechanisms of how amylin decreases and stimulated conditions, directly proportional food intake are not yet fully understood. How- to the amount of body fat and that insulin, when ever, coadministration of leptin and amylin was administered directly into the brain of baboons, recently shown to synergistically decrease body reduced food intake and adiposity [12, 13] . In- weight and adiposity in both obese rodents and deed, insulin receptors are expressed in several humans, thus indicating that amylin is able to re- hypothalamic areas governing energy balance, store leptin sensitivity [19, 20]. However, other such as the ARC and the dorsomedial hypothala- studies indicate that also cotreatment of diet-in- mus. Accordingly, administration of insulin di- duced obese mice with leptin and exendin-4, and rectly in the brain reduces food intake in a variety with fibroblast growth factor 21 improves leptin of species, including rodents and non-human sensitivity, thus indicating the improvement of primates [14] . The proposed mechanism of how leptin sensitivity is not unique to amylin signal- insulin inhibits food intake is that insulin enters ing [21] . the CNS in proportion to its plasma concentra- tions where high levels of insulin stimulate the activity of neurons expressing proopiomelano- Cholecystokinin cortin and cocaine- and amphetamine-related transcript, while those expressing NPY and Secreted from I-cells of the small intestine in AgRP are inhibited [9] . response to nutrient ingestion, CCK was the first gut peptide reported to be implicated in the regu- lation of food intake and system metabolism [22] . Amylin Accordingly, peripheral administration of CCK dose-dependently decreases food intake through Amylin is a 37-amino acid peptide cosecreted a reduction in meal size in a variety of species in- with insulin from the pancreatic ␤ -cells. Similar cluding rodents [22] and humans [23, 24] . The to insulin, plasma concentrations of amylin in- anorexigenic effect of CCK is mediated through crease in response to nutrient stimuli and subse- the CCK1 receptor. Accordingly, CCK does not quently decrease upon fasting. Like leptin and in- affect food intake in CCK1 receptor-deficient ro- sulin, circulating levels of amylin are increased in dents [25] . CCK1 receptors are found in many ar- obese compared to lean individuals and accord- eas implicated in energy metabolism control, ingly decrease upon weight loss [15] . Secreted such as the afferent and efferent vagal neurons, from the pancreas, amylin regulates gastric emp- the nucleus tractus solitaries (NTS), the AP and tying and modulates glucose metabolism through the hypothalamus. The exact mechanism of how inhibition of glucagon secretion. When injected CCK inhibits food intake is not yet fully under- into the brain or the periphery, amylin further stood. However, the most common conceptual- dose-dependently decreases food intake due to a ization is that CCK activates CCK1 receptors on reduction in meal size [16] . In line with its role as sensory fibers of the vagus nerve that remit the

98 Müller Pfluger Tschöp

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539) CCK signal over the NTS to the brainstem from humans. An important site of glucagon’s action is which it is transmitted to the hypothalamus. In thereby the liver, which informs the brain via the hypothalamus, CCK signal then leads to the sensory fibers of the vagus nerve about changes activation of downstream pathways that finally in circulating concentrations of glucagon. The lead to the termination of food intake [9] . brain responds to increased circulating concen- trations of glucagon by inhibiting food intake [29–31] . The identification of the liver as the pri- Cleavage Products of Proglucagon mary site of glucagon’s action on food intake is based on the observation that infusion of gluca- Proglucagon is generated in ␣ -cells of the pan- gon in the hepatic portal vein decreases food in- creas, in L-cells of the distal ileum and colon, and take at concentrations ten times lower as com- in the NTS of the brainstem. In a tissue-specific pared to infusions into the vena cava [29] and that manner, proglucagon is cleaved by the prohor- the anorexigenic effect of glucagon is abrogated mone convertase 1 or 2 into either glucagon, by hepatic vagotomy. In line with its role as a meal GLP-1, GLP-2, OXM, or glicentin. terminating factor, plasma concentrations of glu- cagon increase physiologically during meals and Glucagon preprandial inhibition of glucagon signaling in- Glucagon is exclusively secreted from the pancre- creases the meal size [31, 32] , whereas stimulation atic ␣ -cells in response to decreasing concentra- of glucagon signaling during a meal has the op- tions of blood glucose. The tissue specificity is posite effect. In addition to its ability to decrease achieved through posttranslational cleavage of food intake, glucagon participates in the regula- the proglucagon peptide by the prohormone con- tion of energy metabolism by stimulating lipoly- vertase-2. In its classical role, glucagon counter- sis and fatty acid oxidation while increasing en- acts the glucose-lowering effect of insulin by ergy expenditure most likely through an increase stimulating hepatic gluconeogenesis, thus help- in SNS-mediated activation of brown fat thermo- ing to maintain normal level of blood glucose in genesis. states of rapid glucose utilization. Glucagon promotes its biological action Glucagon-Like Peptide 1 through activation of the glucagon receptor), a GLP-1 is cosecreted with PYY from enteroendo- seven-transmembrane G-protein-coupled recep- crine L-cells of the small intestine in response to tor which is highly expressed in the liver and kid- food ingestion. GLP-1 exists as either a 36- or ney and to a lesser extent in the brain, adipocytes, 37-amino acid peptide and exerts its biological heart, spleen, lymphoblasts, the adrenal gland effects through activation of the GLP-1 receptor and the GI tract [26] . In the pancreas, glucagon (GLP-1R). Upon activation, GLP-1R leads to an receptors are predominantly located in the ␤ - increase in intracellular cAMP production and cells, where high concentrations of glucagon subsequent activation of downstream pathways stimulate the secretion of insulin to prevent hy- via stimulation of the adenylate cyclase. GLP-1R perglycemia. In line with glucagon’s opposing ef- is predominantly expressed in the brain, the fect to insulin, glucagon receptor knockout mice pancreas and the GI tract [33] . In its classical role ( Gcgr –/– ) have lower levels of blood glucose [27] as an incretin (a peptide that stimulates the re- and enhanced insulin sensitivity [28] . In addition lease of insulin), GLP-1 promotes insulin secre- to its role in glucose homeostasis, glucagon de- tion in a glucose-dependent manner while in- creases food intake and promotes body weight hibiting the release of glucagon. In addition to its loss in a variety of species, including rodents and effect on blood glucose, GLP-1 affects energy

Gut-Brain Communication in the Regulation of System Metabolism 99

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539) metabolism through an inhibition of food intake cagon/GLP-1 coagonist synergistically normal- and through a delay in gastric emptying [34] . In ized body weight and glucose tolerance and de- the brainstem and the paraventricular nucleus, creased liver steatosis within 4 weeks without any peripheral administration of GLP-1 leads to in- adverse effects [37] . Together, these data show the creased immunoreactivity of c-fos [35] , thus in- principle that new highly active peptides that si- dicating that the effect of GLP-1 on food intake multaneously activate multiple signaling path- is mediated over both the hypothalamus and the ways can be designed to safely and efficiently brainstem. In line with this observation, intra- normalize body weight and blood glucose, thus cerebroventricular administration of GLP-1 ro- paving the way for a new area in obesity and dia- bustly decreases food intake in rats, whereas ex- betes research. endin 9–39, an inhibitor of GLP-1R signaling, has the opposite effect. The effect of GLP-1 on food Oxyntomodulin intake is, however, strikingly ameliorated upon OXM is a 37-amino acid protein containing the vagotomy, thus indicating that GLP-1 regulates full 29 amino acid sequence of glucagon. OXM is food intake over the vagus-brainstem-hypothal- secreted from enteroendocrine L-cells of the amus pathway [36] . Due to its effect on food in- small intestine in response to a meal, and exerts take and glucose homeostasis, GLP-1 has gained its biological effects through activation of both much scientific attention as a target for the treat- the GLP-1 and the glucagon receptor. When in- ment of obesity and type 2 diabetes. However, jected into either the brain or the periphery, acute the therapeutic potential of GLP-1 is hampered and chronic administration of OXM reduces by its short half-life (in humans 1–2 min), which food intake and promotes weight loss in rodents is due to its rapid degradation by the dipeptidyl [38] . In line with these findings, a 4-week treat- peptidase IV (DPP-IV). In contrast to native ment of overweight and obese people with OXM GLP-1, exendin-4, a GLP-1R agonist originally (at doses of 1,200 nmol/day) decreased body isolated from the venom of the gila monster (He- weight by 2.3 8 0.4 kg compared to 0.5 8 0.5 kg loderma suspectum) , exhibits a greatly enhanced in the control group [39]. In another clinical trial, half-life due to relative resistance to DPP-IV a 4-day treatment of overweight and obese people degradation. Accordingly, DPP-IV-resistant with OXM (at doses of 1,200 nmol/day) decreased GLP-1 analogs, such as Exenatide€ (Bayetta€ , food intake while increasing energy expenditure, Amylin Pharmaceuticals Inc., San Diego, Calif., thus supporting the therapeutic potential of USA) and Liraglutide€ (Novo Nordisk, Den- OXM in the treatment of obesity. Peripheral ad- mark) are approved by the FDA and are current- ministration of OXM increases c-fos immunore- ly successfully used for the treatment of obesity activity in the ARC, whereas selective inhibition and diabetes. of GLP-1Rs in the ARC abolished the effect of New ground in diabetes and obesity research OXM on food intake and weight loss. According- was recently broken by the biochemical engineer- ly, these data indicate that OXM inhibits food in- ing of a single peptide with agonism at both the take over GLP-1R signaling in the hypothalamus glucagon and the GLP-1R, thereby combining the like with other GI peptides, but the therapeutic antihyperglycemic effects of GLP-1 with the lipo- potential of OXM to promote weight loss is ham- lytic and thermogenic properties of glucagon in a pered by its short half-life (in humans 12 8 single peptide of improved pharmacokinetic and 1 min), which is due to its rapid degradation by sustained action as compared to native glucagon DPP-IV. Accordingly, as with other members of and GLP-1. Once-weekly treatment of diet-in- the incretin family, biochemically modified duced obese mice with this newly designed glu- OXM derivates have been generated that show an

100 Müller Pfluger Tschöp

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539) increased half-life due to decreased DPP-IV deg- or response to incoming nutrients. In concert, radation. these peptides signal the GI fuel status to the CNS In summary, the GI tract as the largest endo- in order to adjust food intake and energy expen- crine organ of the body produces a variety of neu- diture. ropeptides that are either secreted in anticipation

References

1 Sakata I, Nakamura K, Yamazaki M, 10 Kamegai J, Tamura H, Shimizu T, Ishii 19 Roth JD, Roland BL, Cole RL, Trevaskis Matsubara M, Hayashi Y, Kangawa K, S, Sugihara H, Wakabayashi I: Central JL, Weyer C, Koda JE, Anderson CM, Sakai T: Ghrelin-producing cells exist effect of ghrelin, an endogenous Parkes DG, Baron AD: Leptin respon- as two types of cells, closed- and growth hormone secretagogue, on hy- siveness restored by amylin agonism in opened-type cells, in the rat gastroin- pothalamic peptide gene expression. diet-induced obesity: evidence from

testinal tract. Peptides 2002; 23: 531– Endocrinology 2000; 141: 4797–4800. nonclinical and clinical studies. Proc

536. 11 Kamegai J, Tamura H, Shimizu T, Ishii Natl Acad Sci USA 2008; 105: 7257– 2 Kojima M, Hosoda H, Date Y, Nakazato S, Sugihara H, Wakabayashi I: Chronic 7262. M, Matsuo H, Kangawa K: Ghrelin is a central infusion of ghrelin increases 20 Trevaskis JL, Coffey T, Cole R, Lei C, growth-hormone-releasing acylated hypothalamic neuropeptide Y and Wittmer C, Walsh B, Weyer C, Koda J,

peptide from stomach. Nature 1999; Agouti-related protein mRNA levels Baron AD, Parkes DG, et al: Amylin-

402: 656–660. and body weight in rats. Diabetes 2001; mediated restoration of leptin respon-

3 Gutierrez JA, Solenberg PJ, Perkins DR, 50: 2438–2443. siveness in diet-induced obesity: Willency JA, Knierman MD, Jin Z, 12 Woods SC, Lotter EC, McKay LD, Porte magnitude and mechanisms. Endocri-

Witcher DR, Luo S, Onyia JE, Hale JE: D Jr: Chronic intracerebroventricular nology 2008; 149: 5679–5687. Ghrelin octanoylation mediated by an infusion of insulin reduces food intake 21 Muller TD, Sullivan LM, Habegger K, orphan lipid transferase. Proc Natl and body weight of baboons. Nature Yi CX, Kabra D, Grant E, Ottaway N,

Acad Sci USA 2008; 105: 6320–6325. 1979; 282: 503–505. Krishna R, Holland J, Hembree J, et al: 4 Yang J, Zhao TJ, Goldstein JL, Brown 13 Woods SC, Porte D Jr: Insulin and the Restoration of leptin responsiveness in MS: Inhibition of ghrelin O-acyltrans- set-point regulation of body weight; in diet-induced obese mice using an opti- ferase (GOAT) by octanoylated penta- Novin D, Bray GA, Wyrwichka W (eds): mized leptin analog in combination

peptides. Proc Natl Acad Sci USA 2008; Hunger: Basic Mechanisms and Clini- with exendin-4 or FGF21. J Pept Sci

105: 10750–10755. cal Implications. New York, Raven 2012; 18: 383–393. 5 Cummings DE, Purnell JQ, Frayo RS, Press, 1976, pp 273–280. 22 Gibbs J, Young RC, Smith GP: Chole- Schmidova K, Wisse BE, Weigle DS: A 14 Sipols AJ, Baskin DG, Schwartz MW: cystokinin decreases food intake in

preprandial rise in plasma ghrelin lev- Effect of intracerebroventricular insu- rats. J Comp Physiol Psychol 1973; 84: els suggests a role in meal initiation in lin infusion on diabetic hyperphagia 488–495.

humans. Diabetes 2001; 50: 1714–1719. and hypothalamic neuropeptide gene 23 Kissileff HR, Pi-Sunyer FX, Thornton J,

6 Tschop M, Wawarta R, Riepl RL, Fried- expression. Diabetes 1995; 44: 147–151. Smith GP: C-terminal octapeptide of rich S, Bidlingmaier M, Landgraf R, 15 Reda TK, Geliebter A, Pi-Sunyer FX: cholecystokinin decreases food intake

Folwaczny C: Post-prandial decrease of Amylin, food intake, and obesity. Obes in man. Am J Clin Nutr 1981; 34: 154–

circulating human ghrelin levels. J En- Res 2002; 10: 1087–1091. 160.

docrinol Invest 2001; 24:RC19–RC21. 16 Lutz TA, Del Prete E, Scharrer E: Re- 24 Muurahainen N, Kissileff HR, Deroga- 7 Williams DL, Cummings DE: Regula- duction of food intake in rats by intra- tis AJ, Pi-Sunyer FX: Effects of chole- tion of ghrelin in physiologic and peritoneal injection of low doses of am- cystokinin-octapeptide (CCK-8) on

pathophysiologic states. J Nutr 2005; ylin. Physiol Behav 1994; 55: 891–895. food intake and gastric emptying in

135: 1320–1325. 17 Rushing PA, Hagan MM, Seeley RJ, man. Physiol Behav 1988; 44: 645–649. 8 Kirchner H, Gutierrez JA, Solenberg PJ, Lutz TA, D’Alessio DA, Air EL, Woods 25 Kopin AS, Mathes WF, McBride EW, Pfluger PT, Czyzyk TA, Willency JA, SC: Inhibition of central amylin signal- Nguyen M, Al-Haider W, Schmitz F, Schurmann A, Joost HG, Jandacek RJ, ing increases food intake and body adi- Bonner-Weir S, Kanarek R, Beinborn

Hale JE, et al: GOAT links dietary lipids posity in rats. Endocrinology 2001; 142: M: The cholecystokinin-A receptor me- with the endocrine control of energy 5035. diates inhibition of food intake yet is

balance. Nat Med 2009; 15: 741–745. 18 Becskei C, Grabler V, Edwards GL, not essential for the maintenance of

9 Barsh GS, Schwartz MW: Genetic ap- Riediger T, Lutz TA: Lesion of the lat- body weight. J Clin Invest 1999; 103: proaches to studying energy balance: eral parabrachial nucleus attenuates the 383–391. perception and integration. Nat Rev anorectic effect of peripheral amylin

Genet 2002; 3: 589–600. and CCK. Brain Res 2007; 1162: 76–84.

Gut-Brain Communication in the Regulation of System Metabolism 101

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539) 26 Svoboda M, Tastenoy M, Vertongen P, 31 Le Sauter J, Noh U, Geary N: Hepatic 36 Abbott CR, Monteiro M, Small CJ, Sa- Robberecht P: Relative quantitative portal infusion of glucagon antibodies jedi A, Smith KL, Parkinson JR, Ghatei analysis of glucagon receptor mRNA in increases spontaneous meal size in MA, Bloom SR: The inhibitory effects

rat tissues. Mol Cell Endocrinol 1994; rats. Am J Physiol 1991; 261:R162–R165. of peripheral administration of peptide

105: 131–137. 32 Langhans W, Zeiger U, Scharrer E, YY(3–36) and glucagon-like peptide-1 27 Gelling RW, Du XQ, Dichmann DS, Geary N: Stimulation of feeding in rats on food intake are attenuated by abla- Romer J, Huang H, Cui L, Obici S, Tang by intraperitoneal injection of antibod- tion of the vagal-brainstem-hypotha-

B, Holst JJ, Fledelius C, et al: Lower ies to glucagon. Science 1982; 218: 894– lamic pathway. Brain Res 2005; 1044: blood glucose, hyperglucagonemia, and 896. 127–131. pancreatic alpha cell hyperplasia in 33 Yamato E, Ikegami H, Takekawa K, 37 Day JW, Ottaway N, Patterson JT, Gel- glucagon receptor knockout mice. Proc Fujisawa T, Nakagawa Y, Hamada Y, fanov V, Smiley D, Gidda J, Findeisen

Natl Acad Sci USA 2003; 100: 1438– Ueda H, Ogihara T: Tissue-specific and H, Bruemmer D, Drucker DJ, Chaud- 1443. glucose-dependent expression of recep- hary N, et al: A new glucagon and GLP- 28 Sorensen H, Winzell MS, Brand CL, tor genes for glucagon and glucagon- 1 co-agonist eliminates obesity in ro-

Fosgerau K, Gelling RW, Nishimura E, like peptide-1 (GLP-1). Horm Metab dents. Nat Chem Biol 2009; 5: 749–757.

Ahren B: Glucagon receptor knockout Res 1997; 29: 56–59. 38 Dakin CL, Gunn I, Small CJ, Edwards mice display increased insulin sensitiv- 34 Cummings DE, Overduin J: Gastroin- CM, Hay DL, Smith DM, Ghatei MA, ity and impaired beta-cell function. testinal regulation of food intake. J Clin Bloom SR: Oxyntomodulin inhibits

Diabetes 2006; 55: 3463–3469. Invest 2007; 117: 13–23. food intake in the rat. Endocrinology

29 Geary N, Le Sauter J, Noh U: Glucagon 35 Turton MD, O’Shea D, Gunn I, Beak SA, 2001; 142: 4244–4250. acts in the liver to control spontaneous Edwards CM, Meeran K, Choi SJ, Taylor 39 Wynne K, Park AJ, Small CJ, Patterson

meal size in rats. Am J Physiol 1993; GM, Heath MM, Lambert PD, et al: A M, Ellis SM, Murphy KG, Wren AM, 264:R116–R122. role for glucagon-like peptide-1 in the Frost GS, Meeran K, Ghatei MA, et al: 30 Le Sauter J, Geary N: Pancreatic gluca- central regulation of feeding. Nature Subcutaneous oxyntomodulin reduces

gon: physiological signal of postpran- 1996; 379: 69–72. body weight in overweight and obese dial satiety (in French). Ann Endocri- subjects: a double-blind, randomized,

nol (Paris) 1993; 54: 149–161. controlled trial. Diabetes 2005; 54: 2390–2395.

Prof. Matthias H. Tschöp Institute for Diabetes and Obesity (IDO) Business Campus Garching-Hochbrück Parkring 13 DE–85748 Garching (Germany)

E-Mail matthias.tschoep @ helmholtz-muenchen.de

102 Müller Pfluger Tschöp

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 96–102 (DOI: 10.1159/000346539)

Chapter 5: Drosophila and Immunity

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516)

Metabolism and Innate Immunity: FOXO Regulation of Antimicrobial Peptides in Drosophila

a a a b Gerrit Loch Eva Jentgens Margret Bülow Ingo Zinke c c c a Tetsushi Mori Sayaka Suzuki Haruko Takeyama Michael Hoch a b Molecular Developmental Biology and Molecular Brain Physiology and Behavior, LIMES Institute, University of Bonn, Bonn , c Germany; Center for Advanced Biomedical Science, Faculty of Science and Engineering, Waseda University, Tokyo , Japan

A b s t r a c t The Innate Immune System Metabolic homeostasis is fundamental for the develop- ment and the survival of animals. It requires the proper It is well known that all animals constantly com- functioning of pathways that control the sensing and bat infections. The evolution of various types of processing of nutrients, the storage and mobilization of immune defense mechanisms appears to be na- energy. Recent data demonstrate that energy homeosta- ture’s method of choice to fight this battle. sis and immune responses are tightly connected and that A distinction is commonly made between inaccurate metabolic regulation can adversely influence the adaptive immune system, which is exclu- immune functions. Dysfunctions of the immune system sively found in vertebrate species, and the have been demonstrated to underlie many chronic met- phyloge netically more ancient innate immune abolic diseases, including diabetes, metabolic syndrome system which uses germline-encoded receptors and atherosclerosis. The molecular mechanisms behind for the recognition of infectious microorgan- the cross-regulation of metabolism and immunity in isms. While the latter might appear to be the health and disease are only beginning to emerge. We use less effective immune response at first sight, up the model organism Drosophila melanogaster to gain to ten million species successfully deal with in- insights into evolutionary conserved mechanisms by nate immunity alone, compared to about 45,000 which energy homeostasis and innate immunity inter- vertebrate species which make use of both im- act. Copyright © 2013 S. Karger AG, Basel mune responses [1] . The fruit fly Drosophila serves as an excellent model to study innate im- munity because it offers a broad repertoire of well-established genetic and molecular meth- AMP Genes Primary target ods and, as an invertebrate, lacks an adaptive Drosocin 1 Gram-negative bacteria immune system. Diptericin 2 Gram-negative bacteria Attacin 4 Gram-negative bacteria Cecropin 4 Gram-negative bacteria Antimicrobial Peptides Defensin 1 Gram-positive bacteria Metchnikowin 1 Fungi Antimicrobial peptides (AMPs), small cationic Drosomycin 7 Fungi molecules, are key factors of the innate immune response. The first AMPs have been purified Fig. 1. Seven classes of Drosophila AMPs have been de- about 30 years ago from the giant silk moth Hy- scribed. They can be classified into three groups de- pending on their main target. alophora cecropia and named Cecropins by a group around Hans G. Boman [2, 3] . Since then, many different AMPs which counter a vast range of bacteria, fungi and even viruses have been discovered throughout the plant and ani- tivated upon infection with Gram-positive bacte- mal kingdom [4] . Concerning bacteria, AMPs ria and fungi, the Imd signaling pathway is main- have been proposed to function by interacting ly activated upon infection with Gram-negative with the negatively charged outer layers of bac- bacteria [7–9] . Recognition of microorganisms is terial membranes, although the exact mecha- supposed to be regulated by specific pattern rec- nism by which they destroy their targets has yet ognition receptors, peptidoglycan recognition to be revealed. Due to their antibacterial activi- proteins and Gram-negative binding proteins ty, AMPs may provide a possible future alterna- [10–14] . tive to classical antibiotics [5] . In humans, three T h e Drosophila Toll gene had been identified main AMP classes, the defensins, cathelicidins in the 1980s, and Toll’s role in the development and histatins, are known [4] , whereas seven of the dorsoventral body axis was subsequent- classes of Drosophila AMPs have been discov- ly described [15] . Fifteen years after this initial ered so far. They can be classified into three finding, it was discovered in 1996 that the Toll groups depending on their primary targets: dro- pathway plays also an essential role in Drosoph- socin, diptericin, attacin and cecropin respond ila innate immunity [7] . In mammals, Toll-like mainly to infections with Gram-negative bacte- receptors (TLRs) were subsequently character- ria, defensin responds mostly to infections with ized as binding proteins of the bacterial product Gram-positive bacteria, and metchnikowin as lipopolysaccharide [16]. While the mammalian well as drosomycin are mainly active against TLRs function directly as pattern recognition fungi (fig. 1 ) [6] . receptors, the Drosophila Toll is activated upon binding by a cleaved form of the cytokine Spätz- le [17–19] . Pathway activation leads to the nucle- T h e Drosophila Immune Pathways ar translocation of dorsal and the dorsal-related immunity factor, two transcription factors of the Two distinct immune pathways, the Toll and the Rel family to which also NF-␬ B belongs. In con- immune deficiency (Imd) pathway, regulate the trast to this, the Imd pathway signals through infection-dependent expression of AMPs in the the Rel family member relish. Activation of the fly [7, 8]. While the Toll pathway is primarily ac- transcription factors directly results in AMP ex-

104 Loch Jentgens Bülow Zinke Mori Suzuki Takeyama Hoch

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) pression. Consequently, the specific transcrip- tion factor binding sites in their promoters de- termine if the AMP genes are more sensitive to the Toll or the Imd pathway or are coregulated [6] . Several other signaling pathways additional- ly contribute to the immune defense in Drosoph- ila. Among them are the c-Jun N-terminal ki- nase pathway that is involved in wound healing processes and the JAK/STAT pathway that has Fig. 2. Body sizes of female adult Drosophila flies. Com- been proposed to respond to tissue damage [20, parison of a wild-type (left) and a step mutant (right) fly. Decreased insulin signaling leads to a reduced organis- 21] . mal growth.

Local and Systemic Immune Defense

The immune system of Drosophila combines lo- Insulin Signaling and the FOXO Transcription cal immune reactions with cellular and system- Factor ic responses. Several surface epithelia serve as a point of attack for microorganisms, among The insulin/insulin-like growth factor signaling them the epithelial surfaces covering the repro- (IlS) cascade is one of the most prominent meta- ductive, the respiratory and the digestive tract. bolic pathways and known to be a major regula- These barrier tissues therefore form a first line tor of energy homeostasis, longevity, organismal of defense against invading pathogens that ex- and cellular growth [22, 23] . The pathway is presses AMPs both constitutively and in an in- highly conserved throughout the metazoans. fection-dependent manner. As flies often feed While screening for genes controlling larval on decaying matter, the gut system can be de- growth, we identified a new component of the scribed as one major site of infection. In addi- IlS pathway which we named Steppke (Step). tion to the local AMP expression, reactive oxy- Mutation of the step gene locus leads to a devel- gen species can be produced in the intestine opmental delay and to a reduced body size in upon infection. It remains to be fully unrevealed larval and adult Drosophila stages ( fig. 2 ) [24] . though, how commensal bacteria can exist in The Step protein is a member of the cytohesin the gut without being wiped out by the host’s GEF (guanine nucleotide exchange factor) fam- immune system. ily and has been shown to be conserved in mam- Pathogens that have succeeded in overcoming mals [25] . Blocking the function of the Step ho- the barrier tissues are countered by free-float- mologues in mice causes elevated transcription ing or sessile hemocytes, which are involved in levels of insulin-repressed gluconeogenic genes phagocytosis and encapsulation as well as mela- and inhibition of glycogen and fatty acid syn- nization processes. During the systemic immune thesis [26] . In the signaling cascade, Step could response, AMPs are expressed in the fat body, the be located upstream of the phosphatidylinositol- main immune organ of Drosophila , and are se- 3-OH kinase (PI3K) [24] . In Dr osophila , the ac- creted into the hemolymph [6] . tivation of the insulin receptor leads to a phos- phorylation cascade including the PI3K and the Akt serine-threonine kinase. The Akt kinase

FOXO Regulation of AMPs in Drosophila 105

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) InR InR

Step Step PI3K PI3K

Akt Akt P FOXO

FOXO

Energy saving Growth inhibition Longevity

Fig. 3. Regulation of FOXO-dependent gene expression by IlS. When the insulin receptor (InR) is activated, a phosphor- ylation cascade leads to the inhibition of the transcription factor FOXO. In the phosphorylated form, FOXO is retained in the cytoplasm of the cell. Without phosphorylation, FOXO is translocated to the nucleus and binds to the regulatory regions of target genes leading to processes like energy saving and storage, growth inhibition and longevity.

links the IlS to the target of rapamycin pathway pression in Drosophila larvae inhibits cell via the TSC (tuberous sclerosis complex) but growth and proliferation and copies a starvation also directly regulates the activation of the phenotype [28] . FOXO has also been linked to FOXO (forkhead box O) transcription factor by stress resistance and in the adult stage of Dro- affecting its localization in the cell (fig. 3 ) [27] . sophila to aging and longevity [29] . The phosphorylated inactive FOXO protein is located in the cytoplasm. Without IlS signaling, the active form is translocated into the nucleus FOXO-Dependent Regulation of where the forkhead box domain of FOXO allows Antimicrobial Peptides direct binding of the protein to regulatory DNA regions. FOXO and its conserved homologues in We recently discovered a new, fundamental mammals are key regulators of metabolic pro- mechanism of cross-regulation between metab- cesses [28] . A large number of genes is regulated olism and the innate immune response, which by FOXO. One of the most prominent ones is a operates under normal physiological conditions translational repressor and effector of cell of oscillating energy levels. Low energy levels growth, the 4E-binding protein. FOXO overex- induced by starvation or absence of insulin sig-

106 Loch Jentgens Bülow Zinke Mori Suzuki Takeyama Hoch

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) TOLL PGRP-LC InR

Step PI3K

DIF Rel Akt

FOXO

AMP expression AMP expression

Fig. 4. Regulation of AMP expression. The Toll and the Imd pathways (left drawing) are activated during an infection depending on the type of pathogen. The Toll pathway is mainly activated by fungi and Gram-positive bacteria, whereas the Imd pathway is mainly activated by Gram-negative bacteria. Dorsal-related immunity factor (DIF) and Rel are NF- ␬B homologs which activate the transcription of AMPs by binding to NF-␬ B binding sites of the regula- tory regions of AMP genes in response to an infection. In contrast, under normal physiological conditions without an infection, the expression of AMPs can be regulated independently of the classical immune pathways, via the IlS pathway: if the energy status of the cell is low, e.g. upon fasting or starvation, IlS is downregulated, and the FOXO transcription factor is translocated to the nucleus where it binds to the regulatory region of AMP genes, thereby in- ducing their expression. This FOXO-dependent mechanism of regulating innate immune effectors is used under normal physiological (non-infection) conditions for regulating microbial communities at barrier tissues. PGRP-LC = Peptidoglycan recognition protein LC.

naling in Drosophila larvae or in human tissue Analyzing the phenotype of step mutants, we culture cells result in the downregulation of IlS observed elevated AMP transcription levels in and the activation of the transcription factor non-infected larvae. This induction was also no- FOXO, which directly initiates AMP gene ex- ticed in other IlS mutants and after blocking the pression. These data show an evolutionary con- IlS cascade with the Step inhibitor SecinH3 in served, direct connection between IlS and im- adult flies. Whereas AMP induction could not be mune effector function that operates indepen- triggered any more in FOXO mutants, overex- dently of the classic immunity pathways ( fig. 4 ) pression of FOXO led to elevated AMP transcript [30] . levels indicating a key role of the transcription factor in regulating AMPs. We subsequently

FOXO Regulation of AMPs in Drosophila 107

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) Basement membrane

Fig. 5. Illustration of the interactions occurring between the microflora and the Drosophila gut. Commensal microorganisms impact on nutrition GUT LUMEN and metabolism of the host and Potential pathogens Commensals play an important role in regulating gut development. Upon pathogen Pathogen-commensal attack, the epithelial enterocytes interactions mount an innate immune response. The FOXO-dependent activation of AMPs occurs when energy levels be- come low and may allow to regulate Pathogen-host Commensal-host the composition of microorganisms interactions interactions in the gut under normal physiologi- Enterocytes cal conditions when the organism is not infected. Pathogen-commensal interactions may also play an impor- HOST INNATE IMMUNITY tant role in the homeostasis of the host microbiota.

demonstrated that FOXO directly regulates the Low energy levels and stress seem to render expression of AMPs such as drosomycin, which animals vulnerable to pathogens. Inucing a was most strongly regulated by IlS and FOXO in FOXO-dependent AMP response under these larvae, by binding to their regulatory regions. conditions might be crucial to fight in particu- The FOXO-dependent regulation of AMPs is in- lar the early steps of an upcoming infection at dependent and operates in parallel to the classical the epithelial barriers. The gut is the major or- immune pathways as we could show by triggering gan for the breakdown of macromolecules in the a FOXO-dependent AMP expression in immune- food and the absorption of nutrients and miner- deficient double mutants for the Toll and the Imd als. The barrier epithelium of the intestine is in cascades [30] . permanent contact with the commensal micro- The FOXO-dependent mechanism of AMP flora, but also with pathogens, and is therefore a regulation was found to act in the fat body, the prominent place for exploring a metabolism-de- main Drosophila organ responsible for systemic pendent regulation of innate immunity. immune responses, and also in all analyzed epi- thelial barrier tissues, such as the tracheal air- way system and the gut. As described before, Drosophila Innate Immunity and the Gut these tissues constitute a first line of defense Microflora Community against pathogens. The regulation of AMP gene expression by FOXO in several human cell lines Gut commensals, non-pathogenic microbes sur- provides evidence for the conservation of this viving within the gut of organisms, are key play- mechanism in mammals and points out its gen- ers in relation to the host’s innate immunity. In eral importance [30] . mammals, they do not only contribute to the

108 Loch Jentgens Bülow Zinke Mori Suzuki Takeyama Hoch

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) host’s energy metabolism but also provide an en- Receptors for Antimicrobial Peptides in the vironment that protects against pathogens [31, Brain of Drosophila? 32]. This beneficial mutual relationship between the host and its commensals was also described Drosophila is prone to infections since it feeds on in the Drosophila gut. It could be shown that the food sources like rotten fruits. The first barrier inhibition of the intestinal homeobox gene Cau- against infections is the gut, which provides a lo- dal led to an overexpression of infection-depen- cal immune response by AMP production. A sys- dent AMPs, which in turn altered the commen- temic response against pathogens is achieved by sal population within the intestine. In particular, AMP production in the fat body and secretion this led to the dominance of Gluconobacter sp. into the hemolymph. As described above, a con- strain EW707, eventually causing gut cell apop- nection between immunity and metabolism ex- tosis and host mortality. When AMPs were ists via the insulin pathway: upon starvation, downregulated by reintroducing Caudal , a FOXO, the key transcription factor downstream healthy microbiota community could be restored of insulin signaling, activates the transcription of [33] . several AMPs [30] and prepares the fly for an up- Further key roles of commensals in Drosophi- coming infection. It is important for the fly to la were recently observed when Lactobacillus react appropriately to an infection by e.g. chang- plantarum , a dominant commensal of the Dro- ing the food source. This could be mediated by sophila gut, was shown to promote larval growth receptors for AMPs. Receptors for AMPs which under nutrient scarcity conditions against the lead to alterations of the metabolic function or target of rapamycin-dependent host nutrient changes in the behavior are not yet known. How- sensing system [34] . Subsequently Ac etobacter ever, it is likely that such receptors exist in the pomorum , another dominant commensal, recog- body and in the brain. Binding of AMPs to their nized via its pyrroloquinoline quinone-depen- receptors could for example lead to reduced food dent alcohol dehydrogenase activity, modulates intake or search for new, uninfected food sourc- IlS in Drosophila to regulate host homeostatic es. In mammals, a mutated form of ␤ -defensin programs controlling developmental rate, body binds to the melanocortin receptor 1 [36] . The size, energy metabolism, and intestinal stem cell melanocortin circuit regulates energy and glu- activity [35] . cose homeostasis from the central nervous sys- Thus far, wild or lab-raised Drosophila are tem and responds to leptin, estrogens and sero- known to harbor gut microflora communities tonin [37]. A melanocortin circuit is missing in comprised of ϳ20 phylotypes, with those from Drosophila , but the investigation of similar the Lactobacillus and Acetobacter genera being mechanisms could lead to the identification of a predominant. With the recent studies related to receptor which can bind to both metabolic neu- innate immunity and pathogenicity, it has be- ropeptides and AMPs. Possible targets of AMPs come evident that the interrelations between in the Drosophila brain are neurons which regu- host, commensals and pathogens are manifold late feeding behavior and metabolism, such as and still require intensive research ( fig. 5 ). dopamine-, serotonin-, neuropeptide F (NPF)- or insulin-like-peptide-expressing neurons. Fu- ture studies will reveal whether it might be pos- sible to pull down and analyze the protein com- plex of the secreted AMP by mass spectrometry under conditions of stable substrate-receptor binding.

FOXO Regulation of AMPs in Drosophila 109

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) Acknowledgements grants to M.H.), the Bonn Excellence Cluster Immuno- Sensation (support to M.H.) and the Young Researcher We would like to thank our colleagues from the Hoch Overseas Visits Program for Vitalizing Brain Circulation and Takeyama laboratories for discussions and com- by Overseas Training Program Division, International ments on the manuscript. This work was supported by Program Department, Japan Society for the Promotion the SFBs 645 and 704 (German Research Foundation, of Science (support to T.M. and S.S.).

References

1 Hoffmann JA, Reichhart JM: Drosoph- 10 Yoshida H, Kinoshita K, Ashida M: Pu- 19 Hu X, Yagi Y, Tanji T, Zhou S, Ip YT: ila innate immunity: an evolutionary rification of a peptidoglycan recogni- Multimerization and interaction of Toll

perspective. Nat Immunol 2002; 3: 121– tion protein from hemolymph of the Spatzle in Drosophila . Proc Natl Acad

126. silkworm, Bombyx mori . J Biol Chem Sci USA 2004; 101: 9369–9374.

2 Hultmark D, Steiner H, Rasmuson T, 1996; 271: 13854–13860. 20 Ramet M, Lanot R, Zachary D, Manfru- Boman HG: Insect immunity. Purifica- 11 Lee WJ, Lee JD, Kravchenko VV, Ule- elli P: JNK signaling pathway is re- tion and properties of three inducible vitch RJ, Brey PT: Purification and mo- quired for efficient wound healing in

bactericidal proteins from hemolymph lecular cloning of an inducible gram- Drosophila . Dev Biol 2002; 241: 145– of immunized pupae of Hyalophora negative bacteria-binding protein from 156.

cecropia . Eur J Biochem 1980; 106: 7–16. the silkworm, Bombyx mori . Proc Natl 21 Agaisse H, Petersen UM, Boutros M,

3 Steiner H, Hultmark D, Engstrom A, Acad Sci USA 1996; 93: 7888–7893. Mathey-Prevot B, Perrimon N: Signal- Bennich H, Boman HG: Sequence and 12 Kang D, Liu G, Lundstrom A, Gelius E, ing role of hemocytes in Drosophila specificity of two antibacterial proteins Steiner H: A peptidoglycan recognition JAK/STAT-dependent response to sep-

involved in insect immunity. J protein in innate immunity conserved tic injury. Dev Cell 2003; 5: 441–450.

Immunol 2009; 182: 6635–6637. from insects to humans. Proc Natl 22 Fontana L, Partridge L, Longo VD: Ex-

4 De Smet K, Contreras R: Human anti- Acad Sci USA 1998; 95: 10078–10082. tending healthy life span – from yeast

microbial peptides: defensins, catheli- 13 Chung YS, Kocks C: Recognition of to humans. Science 2010; 328: 321–326. cidins and histatins. Biotechnol Lett pathogenic microbes by the Drosophi- 23 Teleman AA: Molecular mechanisms of

2005; 27: 1337–1347. la phagocytic pattern recognition re- metabolic regulation by insulin in Dro-

5 Zasloff M: Antimicrobial peptides of ceptor Eater. J Biol Chem 2011; 286: sophila . Biochem J 2009; 425: 13–26.

multicellular organisms. Nature 2002; 26524–26532. 24 Fuss B, Becker T, Zinke I, Hoch M: The

415: 389–395. 14 Ferrandon D, Imler JL, Hetru C, Hoff- cytohesin Steppke is essential for insu- 6 Lemaitre B, Hoffmann J: The host de- mann JA: The Drosophila systemic lin signalling in Drosophila . Nature

fense of Drosophila melanogaster . immune response: sensing and signal- 2006; 444: 945–948.

Annu Rev Immunol 2007; 25: 697–743. ling during bacterial and fungal infec- 25 Kolanus W, Nagel W, Schiller B, Zeitl-

7 Lemaitre B, Nicolas E, Michaut L, tions. Nat Rev Immunol 2007; 7: 862– mann L, Godar S, Stockinger H, Seed B: Reichhart JM, Hoffmann JA: The dor- 874. Alpha L beta 2 integrin/LFA-1 binding soventral regulatory gene cassette spat- 15 Gay NJ, Gangloff M: Structure and to ICAM-1 induced by cytohesin-1, a zle/Toll/cactus controls the potent anti- function of Toll receptors and their cytoplasmic regulatory molecule. Cell

fungal response in Drosophila adults. ligands. Annu Rev Biochem 2007; 76: 1996; 86: 233–242.

Cell 1996; 86: 973–983. 141–165. 26 Hafner M, Schmitz A, Grune I, Srivat- 8 Lemaitre B, Kromer-Metzger E, Mi- 16 Poltorak A, He X, Smirnova I, Liu MY, san SG, Paul B, Kolanus W, Quast T, chaut L, Nicolas E, Meister M, Georgel et al: Defective LPS signaling in C3H/ Kremmer E, Bauer I, Famulok M: Inhi- P, Reichhart JM, Hoffmann JA: A reces- HeJ and C57BL/10ScCr mice: mutations bition of cytohesins by SecinH3 leads

sive mutation, immune deficiency in Tlr4 gene. Science 1998; 282: 2085– to hepatic insulin resistance. Nature

(Imd), defines two distinct control 2088. 2006; 444: 941–944. pathways in the Drosophila host de- 17 Armant MA, Fenton MJ: Toll-like re- 27 Teleman AA: Molecular mechanisms of

fense. Proc Natl Acad Sci USA 1995; 92: ceptors: a family of pattern-recognition metabolic regulation by insulin in Dro-

9465–9469. receptors in mammals. Genome Biol sophila . Biochem J 2010; 425: 13–26.

9 Rutschmann S, Kilinc A, Ferrandon D: 2002; 3:reviews3011.1–reviews3011.6. 28 Puig O, Marr MT, Ruhf ML, Tjian R: Cutting edge: the Toll pathway is re- 18 Weber AN, Tauszig-Delamasure S, Control of cell number by Drosophila quired for resistance to gram-positive Hoffmann JA, Lelievre E, Gascan H, FOXO: downstream and feedback regu- bacterial infections in Drosophila . J Ray KP, Morse MA, Imler JL, Gay NJ, lation of the insulin receptor pathway.

Immunol 2002; 168: 1542–1546. Binding of the Drosophila cytokine Genes Dev 2003; 17: 2006–2020. Spatzle to Toll is direct and establishes

signaling. Nat Immunol 2003; 4: 794– 800.

110 Loch Jentgens Bülow Zinke Mori Suzuki Takeyama Hoch

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) 29 Slack C, Giannakou ME, Foley A, Goss 33 Ryu JH, Kim SH, Lee HY, Bai JY, Nam 35 Shin SC, Kim SH, You H, Kim B, Kim M, Partridge L: dFOXO-independent YD, Bae JW, Lee DG, Shin SC, Ha EM, AC, Lee KA, Yoon JH, Ryu JH, Lee WJ: effects of reduced insulin-like signaling Lee WJ: Innate immune homeostasis by Drosophila microbiome modulates

in Drosophila . Aging Cell 2011; 10: 735– the homeobox gene caudal and com- host developmental and metabolic ho- 748. mensal-gut mutualism in Drosophila . meostasis via insulin signaling. Science

30 Becker T, Loch G, Beyer M, Zinke I, Science 2008; 319: 777–782. 2011; 334: 670–674. Aschenbrenner AC, Carrera P, Inhester 34 Storelli G, Defaye A, Erkosar B, Hols P, 36 Candille SI, Kaelin CB, Cattanach BM, T, Schultze JL, Hoch M: FOXO-depen- Royet J, Leulier F: Lactobacillus plan- Yu B, Thompson DA, Nix MA, Kerns dent regulation of innate immune ho- tarum promotes Drosophila systemic JA, Schmutz SM, Millhauser GL, Barsh

meostasis. Nature 2010; 463: 369–373. growth by modulating hormonal sig- GS: A-defensin mutation causes black 31 Backhed F, Ley RE, Sonnenburg JL, nals through TOR-dependent nutrient coat color in domestic dogs. Science

Peterson DA, Gordon JI: Host-bacterial sensing. Cell Metab 2011; 14: 403–414. 2007; 318: 1418–1423. mutualism in the human intestine. 37 Xu Y, Elmquist JK, Fukuda M: Central

Science 2005; 307: 1915–1920. nervous control of energy and glucose 32 Tanoue T, Umesaki Y, Honda K: Im- balance: focus on the central melano-

mune responses to gut microbiota- cortin system. Ann NY Acad Sci 2011;

commensals and pathogens. Gut 1243: 1–14.

Microbes 2010; 1: 224–233.

Prof. Dr. Michael Hoch Rheinische Friedrich-Wilhelms-Universität Bonn Life and Medical Sciences Institute (LIMES) Carl-Troll-Strasse 31 DE–53115 Bonn (Germany)

E-Mail m.hoch@ uni-bonn.de

FOXO Regulation of AMPs in Drosophila 111

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 103–111 (DOI: 10.1159/000346516) Chapter 6: Innate Signaling and Adaptive Immunity

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530)

Dendritic Cells Orchestrate Innate Immunity against Bacterial Kidney Infection

André P. Tittel Christoph Heuser Natalio Garbi Christian Kurts

Institutes of Molecular Medicine and Experimental Immunology, Friedrich-Wilhelms-Universität Bonn, Bonn , Germany

A b s t r a c t early neutrophilia. However, both CD11c.LuciDTR mice Dendritic cells (DCs) form an abundant network in the and CD11c.DTR mice, but not CD11c.DOG mice, showed kidney tubulointerstitium. Their role in bacterial pyelo- neutrophilia after 72 h, which probably resulted from in- nephritis (PN) is unknown. Here, we studied that role in creased granulopoiesis. All three lines feature time win- a murine PN model induced by transurethral instillation dows during which neutrophilia is negligible. Studies in of uropathogenic Escherichia coli twice at a 3-hour in- these time windows demonstrated that the tubulointer- terval, which increased infection rates from 16% after a stitial DC network serves an innate immune sentinel single instillation to 84%. Already 3 h after the second function against bacterial PN by rapidly recruiting neu- instillation, resident kidney DCs produced most of the trophils into the infected kidney. intrarenal CXCL2 and TNF-␣ , and recruited and activated Copyright © 2013 S. Karger AG, Basel neutrophilic granulocytes, which are critical in PN. When we depleted DCs using CD11c.DTR mice during the first bacterial instillation, neutrophil recruitment as well as Bacterial pyelonephritis (PN) is the most relevant bacterial clearance were markedly delayed. However, kidney infection and one of the most prevalent DC depletion also caused infection-independent gran- infections worldwide. Chronic PN may cause ter- ulocyte release from the bone marrow commencing minal kidney failure, resulting in the need for after 24 h. The resulting neutrophilia paradoxically im- regular dialysis therapy or kidney transplanta- proved bacterial clearance when DCs were depleted 1 tion [1] . Most infections are caused by uropatho- day before infection. This side effect was also seen in genic Escherichia coli (UPEC) bearing distinct CD11c.DOG mice, another transgenic line allowing con- virulence factors, such as fimbriae that facilitate ditional DC depletion. We present here a new transgenic attachment to uroepithelial cells, bacterial ascen- line, CD11c.LuciDTR mice, which is unaffected by such sion from the bladder to the kidney and organ infiltration [2, 3] . The presence of UPEC is sensed previously undescribed side effect that interferes by Toll-like receptors expressed both by local im- with the interpretation of experimental results, mune and parenchymal cells [4] . The innate im- and may even lead to diametrical results if ne- mune defense relies on polymorphonuclear cells glected. We describe a newly generated trans- (PMN) [5], which are recruited to the urogenital genic line that offers a time window where this tract by CXCL2 (IL-8, MIP-2) [6] . The identity of side effect is negligible. the cells that produce CXCL2 and recruit PMN is unclear. The most abundant immune cells in the kid- R e s u l t s ney are dendritic cells (DCs), which form an in- tricate network in the tubulointerstitium [7, 8] . Establishing and Characterizing a Murine Although these cells express F4/80, they func- Infection Model of Pyelonephritis tionally resemble immature DCs, and not mac- We first established an experimental model of rophages (M ⌽ s) [7] . Since UPEC ascend from the bacterial PN by transurethral installation of E. bladder through the urinary tract and enter the coli from the uropathogenic strain 536 (UPEC) kidney tissue from the tubular system, DCs are into the bladder of C57BL/6 mice. In previous theoretically well positioned for sensing these studies, infection rates of 20–30% were described microbes. The role of DCs in PN is unknown. [2, 6] . Confirming these reports, we achieved PN DCs are best known as the most effective induc- at 24 h after a single transurethral instillation of ers of adaptive immunity. They survey non-lym- UPEC in 16% of the mice, as detected by deter- phoid tissues, in which they capture antigens and mining bacterial colony forming units (CFU) af- transport them to draining LN for T cell acti- ter culturing homogenates of infected kidneys on vation. Tissue-resident DCs regulate infiltrating LB agar plates. Such low infection rates preclude effector or memory T cells [9] . Also DCs in the comparing groups of experimental mice. How- kidney have been shown to produce mediators ever, we discovered that a second transurethral that attract and activate T cells [10, 11] . Studying instillation of UPEC 3 h after the first one raised the functional role of DCs has been greatly fa- the infection rate to almost 84%. UPEC possess cilitated by transgenic mice expressing the diph- virulence factors that support their ascension theria toxin receptor (DTR) under the CD11c through the ureter into the kidney [3], and these promoter, which allows conditional DC deple- may explain why the first instillation was so per- tion. CD11c.DTR mice are the first and most missive for contracting infection after the second widely employed of these mouse lines, and were one. Kinetic analysis showed that kidney CFU critical, amongst other, for clarifying the roles of at 3 or 21 h after infection were similarly high, DCs in numerous immune processes [11–20] . whereas they had mostly declined after 3 or 5 However, CD11c.DTR mice allow depletion only days. for 2–3 days, and do not survive a second diph- theria toxin (DT) injection within less than 7 The Innate Immune Response against days [19, 21] unless bone marrow chimeras are Pyelonephritis created [22] . CD11c.DOG mice, a different BAC To study the immune response in PN, we deter- transgenic line allowing conditional DC deple- mined phagocyte infiltration by quantifying the hi lo tion, can be depleted for 12 days until antibodies numbers of the CX 3CR1 Gr-1 F4/80+ CD11c+ int int against DT prevent DC depletion [23]. We have DCs [7] , the CX 3CR1 Gr-1 F4/80+ CD11c– hi employed these two murine lines to study the M ⌽s and the CX 3CR1– Gr-1 F4/80– CD11c– course of PN in the absence of DCs. We found a PMN. Already 3 h after the second UPEC instil-

DCs Orchestrate Innate Immunity against Bacterial Kidney Infection 113

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) lation, a strong increase in the number of PMN DCs in the kidney after 24 h. When we infected and a small and non-significant increase of M ⌽ s CD11c.DTR mice 1 day after DC depletion, their was noted. DC numbers in the kidney progres- kidneys surprisingly contained 5-fold less CFU sively decreased, but increased in the renal LN, of UPEC and higher numbers of PMN compared consistent with migration. At 21 h, PMN num- to non-depleted controls after one further day, bers had increased 14-fold, while M ⌽s numbers indicating that mice lacking DCs developed a remained comparable to those after 6 h. stronger anti-infectious response and controlled To determine the role of these phagocytes in PN better. At first glance, this unexpected find- PN, we infected mice with UPECs expressing re- ing might be interpreted as indicative of an im- combinant green fluorescent protein (GFP). We munosuppressive activity of kidney DCs and found that PMN had taken up 77% of the GFP at with their dispensability for PMN recruitment, 3 h, while DCs and M⌽ s accounted for 19 and although this would be incompatible with their 4%, respectively. After 21 h, PMN were respon- CXCL2 and TNF-␣ production. DT injection as sible for almost all phagocytosis, and DCs and such did not affect PN. M ⌽ s accounted for 3 and 7%, respectively. These To reconcile this discrepancy, we examined findings indicated that PMN were the most active PMN recruitment in infected mice more thor- phagocytes in PN, consistent with their essential oughly. We noted that PMN numbers 24 h after role in this infection [5] . DC depletion were elevated not only in the kid- We next asked how PMN are recruited and ac- ney of infected mice, but also in their spleens, and tivated in PN. We first focused on TNF- ␣ , a me- even in non-infected control mice. At 6 h after diator generally important for PMN activation in DT injection into non-infected CD11c.DTR mice, bacterial infections, which can be rapidly pro- PMN in the blood were very slightly and non- duced by kidney DCs [10]. Intracellular cytokine significantly increased; after 24 h, a 2-fold in- staining showed that DCs produced 75% of the crease was found; after 72 h, PMN levels peaked TNF- ␣ at 3 h after second instillation, M ⌽ s 17%, at 5-fold increased numbers, and after 7 days they and the rest was due to other immune cells, ex- were still increased about 2-fold. Neutrophilia cept for PMN, which produced none. We next ex- was mirrored by intrarenal PMN numbers, which amined production of CXCL2, which is critical were significantly increased after 24 h and peaked for PMN recruitment in PN [6]. DCs accounted after 72 h at 5-fold increased values. for 77%, M ⌽ s for 13%, PMN for 10% and other We speculated that DC depletion-induced immune cells for none of the total CXCL2 at 3 h. neutrophilia might have improved clearance of These findings identified DCs as the main early UPEC from the kidney. To establish causality producers of chemokines that recruit PMN and between these two observations, we depleted of cytokines that activate PMN. DCs from CD11c.DTR mice and 24 h later re- moved the resulting neutrophilia with the Neutrophilia in CD11c.DTR Mice Paradoxically PMN-depleting antibody 1A8, which in our Improves Bacterial Clearance hands eliminated 81% of the circulating PMN. The findings above suggested that DCs might Considering the necessity of PMN for defense recruit PMN into the kidney. To address this hy- against PN and the danger of urosepsis, we ana- pothesis, we employed the widely used CD11c. lyzed the mice already 3 h after the second in- DTR mice expressing the DT receptor (DTR) and stillation to avoid urosepsis. If the improved GFP under the CD11c promoter, which allows UPEC clearance after DC depletion was due to conditional DC ablation [13] . Injection of DT into neutrophilia, then bacterial clearance in PMN- non-infected mice caused the loss of nearly all depleted mice should be unaltered. Indeed, this

114 Tittel Heuser Garbi Kurts

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) was the case: intrarenal CFU numbers in mice tious control, the infectious load at 21 h was depleted of PMN and in mice depleted of both lower than at 3 h. Taken together, these findings PMN and DCs were similarly increased over demonstrated that DCs indeed improved the those in mice depleted of DCs alone, and did not early PMN recruitment and the early innate an- significantly differ. This indicated that indeed tibacterial defense in PN. neutrophilia, rather than other hypothetic side effects of DC depletion had improved the de- DTR-Dependent Neutrophilia Is Caused by fense against PN. Granulocyte Release from the Bone Marrow We next investigated the cause of neutrophilia. Resident Renal DCs Recruit Polymorphonuclear Normally, a quarter of all PMN circulate in the Cells for Early Defense against Bacterial blood, half of them reside in the bone marrow Pyelonephritis and the remaining quarter is located in the so- In CD11c.DTR mice, our kinetic analysis re- called marginated pool, which designates intra- vealed a time window until at least 6 h after DT vascular niches within organs like the lungs or injection, during which PMN numbers were not the liver, through which PMN pass at low speed yet significantly increased. If the interpretation [24]. If mobilization of the marginated pool was above is correct, then this time window should the cause of neutrophilia, then PMN numbers in permit studying the role of DCs without con- the liver and lung should be decreased after DT founding neutrophilia. To test this conjecture, we injection. However, we noted an increase in these depleted CD11c.DTR mice of DCs and simulta- organs, similar to that in spleen and blood. By neously instilled UPEC, followed by a second in- contrast, PMN numbers were decreased in the stillation 3 h later, and analyzed the mice at 6 h bone marrow 6 h after DT injection, revealing after depletion. Intrarenal DC numbers were bone marrow release as the origin of neutrophil- only partially reduced at that time point. Never- ia. This release was associated with increased theless, the DCs were already functionally inca- blood levels of the chemokines CXCL1 and pacitated, because they did not produce TNF-␣ CXCL2, which attract PMN. Blocking these che- and CXCL2 anymore. Hence, kidney DCs had mokines with antibodies prevented neutrophilia, ceased to function before their death became ap- indicating that neutrophilia resulted from che- parent. mokine-mediated PMN relocation from the bone In this setting, the bacterial load in the kidney marrow to the bloodstream. was more than 7-fold higher in DC-depleted A previous study by Jung’s group reported in- CD11c.DTR mice at 3 h after the second instilla- creased monocytes and PMN production in mice tion than in non-depleted controls, indicating with constitutive depletion of DCs due to trans- impaired bacterial clearance. Intrarenal PMN genic expression of DT, and this was interpreted numbers were 70% decreased at that time point, as a DC-mediated feedback suppression of my- indicating DC-dependent PMN recruitment. At elopoiesis [25] . However, bone marrow PMN did 21 h after infection, PMN numbers in DC-de- not proliferate differentially at 1 day after DC pleted mice were still reduced. This reduction depletion, excluding increased granulopoiesis at was smaller than after 3 h, which may be ex- that early time point. By contrast, at later time plained by the stronger infection in the absence points, e.g. on day 3, we confirmed myeloprolif- of DCs, which presumably had induced other re- eration in the absence of DCs. cruitment mechanisms that partially compen- sated for the loss of CXCL2-producing DCs, al- beit with a delay. Consistent with delayed infec-

DCs Orchestrate Innate Immunity against Bacterial Kidney Infection 115

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) DTR-Dependent Neutrophilia in Other GFP reporter in splenic CD11c+ cells, including CD11c.DTR Transgenic Mouse Lines CD11b+ and CD8+ conventional DCs, low ex- We finally wished to confirm our interpreta- pression in plasmacytoid DCs and no expression tions in a different transgenic line allowing in CD11c– CD11b+ M ⌽ s and in Ly6G+ granulo- conditional DC ablation, the recently described cytes. Also B220+, CD3+ and NK1.1+ lympho- CD11c.DOG mice expressing DTR, OVA and cytes showed no transgene expression. The de- GFP in DCs [23] . Also these mice showed neutro- pletion of conventional DCs in these mice was ef- philia after DT injection. This finding demon- ficient ( 1 90%), specific and comparable to that in strated that neutrophilia was not unique to CD11c. CD11c.DTR and CD11c.DOG mice, including DTR mice. However, its time course and extent less effective depletion of plasmacytoid DCs differed between these two lines. Neutrophilia ( ϳ 50%). Lymphocyte numbers were not reduced peaked in CD11c.DTR mice after 72 h, in CD11c. after DT application. DOG mice already 24 h after DT. Importantly, Also depletion of kidney DCs in CD11c. when injected with a high dose of DT (40 ng/g LuciDTR mice 24 h after injecting DT was effi- body weight), PMN numbers in CD11c.DOG cient. Importantly, at this time point, there was mice were elevated already 6 h after DC deple- no neutrophilia in these mice. And indeed, when tion and reached levels similar to those detected we depleted DCs in CD11c.LuciDTR mice, intra- in CD11c.DTR mice after 24 h. Since the latter renal PMN were less abundant and UPEC CFU were protected by that number of PMN, we pre- numbers were higher after 1 day, indicating de- dicted that CD11c.DOG mice will be so at 6 h layed bacterial clearance. This corroborated our after DC depletion. Indeed, CFU numbers in conclusion that improved bacterial clearance in DC-depleted CD11.DOG mice were lower than CD11c.DTR and CD11c.DOG mice was due to in non-depleted controls at that time point. neutrophilia induced by DC depletion. Further- When we depleted DCs 24 h before infection, more, it confirmed in a neutrophilia-free setting neutrophilia was at its peak and consistently bac- that renal DCs indeed recruit PMN for rapid de- terial defense was even better. In conclusion, fense against PN. both CD11c.DTR and CD11c.DOG mice develop Neutrophilia also occurred in CD11c. neutrophilia after DC depletion, which can im- LuciDTR mice, albeit 2 days later, 72 h after DC prove bacterial clearance if it coincides with the depletion. However, this was not due to bone infection. The greater supply of PMN in the cir- marrow release, but instead correlated with in- culation presumably allowed their faster recruit- creased granulopoiesis, which unsurprisingly re- ment into the kidney despite the absence of quired longer for causing neutrophilia. This was CXCL2-producing DCs. also seen in CD11c.DTR mice. Increased granu- lopoiesis had been previously observed by Jung’s CD11c.LuciDTR Mice Are Not Affected by Early group in mice constitutively expressing DT in Polymorphonuclear Cell Release from the Bone DCs [25] . Interestingly, granulopoiesis was not Marrow increased in CD11c.DOG mice, and consequent- We generated a further line of BAC transgenic ly these mice did not show neutrophilia 3 days mice termed CD11c.LuciDTR mice, which ex- after DC depletion. Thus, the early PMN release pressed eGFP, CRE recombinase, DTR and lucif- from the bone marrow described in the present erase as individual proteins, as a unique tool for study is a phenomenon distinct from DC deple- work with DCs. The transgenic expression pro- tion-induced myeloproliferation. Future studies file of these mice resembled that of CD11c.DTR may address the underlying molecular mecha- mice, as evidenced by exclusive expression of the nisms in the three transgenic lines.

116 Tittel Heuser Garbi Kurts

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) Conclusions functions, because their reduction after DC de- pletion may have been partially compensated by CD11c.DTR mice arguably are the most impor- neutrophilia. On the other hand, some studies tant tool for exploring the in vivo role of DCs employing CD11c.DTR mice reported anti-in- [12, 13]. However, we found that a slight change flammatory functions. It is speculative whether in the experimental protocol, which at first and to what extent these changes observed after glance seemed more appropriate for the ques- DC depletion in fact had been due to neutrophil- tion asked, led to diametrically different exper- ia. It remains to be studied whether also DTR imental results: When DCs were depleted 1 day transgenic mouse lines allowing conditional de- before infection rather than at the same time, pletion of cells other than DCs may show this side infection was cleared faster than in non-deplet- effect. If so, then especially experiments in mice ed controls. At first glance, this might have been allowing conditional ablation of immunosup- interpreted as indicative of anti-inflammatory pressive cells might be prone to misinterpreta- properties of DCs. However, this discrepancy tions. was in fact due to a previously unnoticed side In summary, we report that 3 DTR transgenic effect in CD11c.DTR mice: early PMN release lines develop neutrophilia after DC depletion due from the bone marrow storage compartment re- to two different mechanisms: (1) PMN release sulting in neutrophilia at 24 h after DC deple- from the bone marrow, causing neutrophilia af- tion. This side effect is not unique to CD11c. ter 24 h in CD11c.DTR and CD11c.DOG mice, DTR mice, but also occurs in CD11c.DOG mice, and (2) granulopoiesis, causing neutrophilia in consistent with the previously observed in- CD11c.DTR and CD11c.LuciDTR mice after 72 h. crease in total splenic CD11b+ Gr1+ cells [23] . Thus, the time point of DC depletion and the We provided three separate lines of evidence for transgenic line need to be considered when plan- a causal connection between neutrophilia and ning experiments under conditions that may be improved bacterial clearance: first, in the ab- affected by PMN. For analysis at 72 h after DC sence of PMN, DC depletion did not improve depletion, CD11c.DOG mice are preferable. How- bacterial clearance in CD11c.DTR mice, ex- ever, these mice preclude the use of the OVA as a cluding a role of other hypothetic side effects model antigen. Furthermore, after 72 h DCs be- after DC depletion; second, CD11c.DOG mice gin to reappear in lymphatic organs. This is not showed early neutrophilia and faster bacterial the case for non-lymphatic organs like the kid- clearance after DC depletion; third, CD11c. ney, where DC repletion takes 7 days or longer LuciDTR mice were free from early neutrophil- [19]. CD11c.DOG mice permit continued DC de- ia and did not clear bacteria faster. In support of pletion, in contrast to CD11c.DTR mice. Howev- our interpretations, Autenrieth et al. [26] re- er, continual DC depletion causes myeloprolifer- cently observed neutrophilia and improved ation [25] , precluding the use also of these mice clearance of Yersinia after DC depletion in after more than 72 h in situations where neutro- CD11c.DOG mice. philia is an issue. CD11c.LuciDTR mice are the The majority of the previous studies employ- line of choice for analysis at 24 h after DC deple- ing CD11c.DTR mice reported immunogenic or tion. proinflammatory roles of DCs, such as the pro- Finally, our study revealed that kidney DCs duction of cytokines or the priming or restimula- are strategically positioned for sensing ascend- tion of T cells. One implication of our present ing bacteria. The sentinel function discovered findings is that these studies might have under- here may be an important purpose of the abun- estimated the extent of pro-inflammatory DC dant tubulointerstitial DC network. In a broader

DCs Orchestrate Innate Immunity against Bacterial Kidney Infection 117

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) sense, our findings illustrate that organ-resident Acknowledgments DCs not only regulate infiltrating effector cells of the adaptive, but also of the innate immune We thank S. Jung for transgenic mice and helpful discus- sions. We acknowledge support by the Central Animal system. Facilities of the Medical Faculty Bonn and the Flow Cy- tometry Core Facility at the IMMEI. This work was sup- ported by the Deutsche Forschungsgemeinschaft (grant Ku1038/5, SFB TR57, KFO228).

References

1 Hooton TM, Stamm WE: Diagnosis and 8 Soos TJ, Sims TN, Barisoni L, Lin K, 15 van Rijt LS, Jung S, Kleinjan A, Vos N, treatment of uncomplicated urinary Littman DR, Dustin ML, Nelson PJ: Willart M, Duez C, Hoogsteden HC, tract infection. Infect Dis Clin North CX3CR1+ interstitial dendritic cells Lambrecht BN: In vivo depletion of

Am 1997; 11: 551–581. form a contiguous network throughout lung CD11c+ dendritic cells during al-

2 Mulvey MA, Lopez-Boado YS, Wilson the entire kidney. Kidney Int 2006; 70: lergen challenge abrogates the charac- CL, Roth R, Parks WC, Heuser J, Hult- 591–596. teristic features of asthma. J Exp Med

gren SJ: Induction and evasion of host 9 Wakim LM, Waithman J, van Rooijen 2005; 201: 981–991. defenses by type 1-piliated uropatho- N, Heath WR, Carbone FR: Dendritic 16 Plaks V, Birnberg T, Berkutzki T, Sela

genic Escherichia coli . Science 1998; cell-induced memory T cell activation S, Benyashar A, Kalchenko V, Mor G,

282: 1494–1497. in nonlymphoid tissues. Science 2008; Keshet E, Dekel N, Neeman M, Jung S:

3 Svanborg C, Bergsten G, Fischer H, 319: 198–202. Uterine DCs are crucial for decidua Godaly G, Gustafsson M, Karpman D, 10 Dong X, Swaminathan S, Bachman LA, formation during embryo implantation

Lundstedt AC, Ragnarsdottir B, Svens- Croatt AJ, Nath KA, Griffin MD: Resi- in mice. J Clin Invest 2008; 118: 3954– son M, Wullt B: Uropathogenic Esch- dent dendritic cells are the predomi- 3965. erichia coli as a model of host-parasite nant TNF-secreting cell in early renal 17 Bamboat ZM, Ocuin LM, Balachandran

interaction. Curr Opin Microbiol 2006; ischemia-reperfusion injury. Kidney VP, Obaid H, Plitas G, DeMatteo RP:

9: 33–39. Int 2007; 71: 619–628. Conventional DCs reduce liver isch- 4 Schilling JD, Martin SM, Hung CS, Lo- 11 Heymann F, Meyer-Schwesinger C, emia/reperfusion injury in mice via

renz RG, Hultgren SJ: Toll-like receptor Hamilton-Williams EE, Hammerich L, IL-10 secretion. J Clin Invest 2010; 120: 4 on stromal and hematopoietic cells Panzer U, Kaden S, Quaggin SE, Floege 559–569. mediates innate resistance to uropatho- J, Grone HJ, Kurts C: Kidney dendritic 18 Connolly MK, Bedrosian AS, Mallen-St genic Escherichia coli . Proc Natl Acad cell activation is required for progres- Clair J, Mitchell AP, Ibrahim J, Stroud

Sci USA 2003; 100: 4203–4208. sion of renal disease in a mouse model A, Pachter HL, Bar-Sagi D, Frey AB, 5 Haraoka M, Hang L, Frendeus B, Go- of glomerular injury. J Clin Invest Miller G: In liver fibrosis, dendritic

daly G, Burdick M, Strieter R, Svanborg 2009; 119: 1286–1297. cells govern hepatic inflammation in

C: Neutrophil recruitment and resis- 12 Sapoznikov A, Jung S: Probing in vivo mice via TNF-alpha. J Clin Invest 2009;

tance to urinary tract infection. J Infect dendritic cell functions by conditional 119: 3213–3225.

Dis 1999; 180: 1220–1229. cell ablation. Immunol Cell Biol 2008; 19 Scholz J, Lukacs-Kornek V, Engel DR,

6 Godaly G, Proudfoot AE, Offord RE, 86: 409–415. Specht S, Kiss E, Eitner F, Floege J, Svanborg C, Agace WW: Role of epithe- 13 Jung S, Unutmaz D, Wong P, Sano G, Groene HJ, Kurts C: Renal dendritic lial interleukin-8 (IL-8) and neutrophil De los Santos K, Sparwasser T, Wu S, cells stimulate IL-10 production and IL-8 receptor A in Escherichia coli - Vuthoori S, Ko K, Zavala F, Pamer EG, attenuate nephrotoxic nephritis. J Am

induced transuroepithelial neutrophil Littman DR, Lang RA: In vivo deple- Soc Nephrol 2008; 19: 527–537.

migration. Infect Immun 1997; 65: tion of CD11c(+) dendritic cells abro- 20 Tadagavadi RK, Reeves WB: Renal den- 3451–3456. gates priming of CD8(+) T cells by ex- dritic cells ameliorate nephrotoxic 7 Kruger T, Benke D, Eitner F, Lang A, ogenous cell-associated antigens. acute kidney injury. J Am Soc Nephrol

Wirtz M, Hamilton-Williams EE, Engel Immunity 2002; 17: 211–220. 2010; 21: 53–63. D, Giese B, Muller-Newen G, Floege J, 14 Osterholzer JJ, Milam JE, Chen GH, 21 Probst HC, Tschannen K, Odermatt B, Kurts C: Identification and functional Toews GB, Huffnagle GB, Olszewski Schwendener R, Zinkernagel RM, Van characterization of dendritic cells in MA: Role of dendritic cells and alveolar Den BM: Histological analysis of the healthy murine kidney and in ex- macrophages in regulating early host CD11c-DTR/GFP mice after in vivo de- perimental glomerulonephritis. J Am defense against pulmonary infection pletion of dendritic cells. Clin Exp

Soc Nephrol 2004; 15: 613–621. with Cryptococcus neoformans. Infect Immunol 2005; 141: 398–404.

Immun 2009; 77: 3749–3758.

118 Tittel Heuser Garbi Kurts

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) 22 Zammit DJ, Cauley LS, Pham QM, Le- 24 Summers C, Rankin SM, Condliffe AM, 26 Autenrieth SE, Warnke P, Wabnitz GH, francois L: Dendritic cells maximize Singh N, Peters AM, Chilvers ER: Neu- Lucero Estrada C, Pasquevich KA, the memory CD8 T cell response to trophil kinetics in health and disease. Drechsler D, Gunter M, Hochweller K,

infection. Immunity 2005; 22: 561–570. Trends Immunol 2010; 31: 318–324. Novakovic A, Beer-Hammer S, Samstag 23 Hochweller K, Striegler J, Hammerling 25 Birnberg T, Bar-On L, Sapoznikov A, Y, Hammerling GJ, Garbi N, Autenrieth GJ, Garbi N: A novel CD11c.DTR trans- Caton ML, Cervantes-Barragan L, IB: Depletion of dendritic cells enhanc- genic mouse for depletion of dendritic Makia D, Krauthgamer R, Brenner O, es innate anti-bacterial host defense cells reveals their requirement for ho- Ludewig B, Brockschnieder D, Rieth- through modulation of phagocyte ho-

meostatic proliferation of natural killer macher D, Reizis B, Jung S: Lack of con- meostasis. PLoS Pathog 2012; 8:

cells. Eur J Immunol 2008; 38: 2776– ventional dendritic cells is compatible e1002552. 2783. with normal development and T cell homeostasis, but causes myeloid prolif-

erative syndrome. Immunity 2008; 29: 986–997.

Prof. Christian Kurts Friedrich-Wilhelms-Universität Bonn Institutes of Molecular Medicine and Experimental Immunology Sigmund-Freud-Strasse 25 DE–53127 Bonn (Germany)

E-Mail ckurts@ uni-bonn.de

DCs Orchestrate Innate Immunity against Bacterial Kidney Infection 119

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 112–119 (DOI: 10.1159/000346530) Chapter 7: Speakers at the Symposium

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503)

Speakers at the Symposium

David Artis, PhD where he isolated TNF. At the UT Southwestern Medical Center in Dallas (1993–1998), he used a D.A. focused in his PhD thesis work at the Uni- classical genetic approach to identify TLR4 as the versity Manchester, UK, on the regulation of im- mammalian LPS receptor – a discovery recog- munity and inflammation following infection nized 13 years later by the 2011 Nobel Prize in with enteric pathogens. He then joined Philipp Physiology or Medicine. In 2000, he moved to Scott’s lab at the University of Pennsylvania the Scripps Research Institute in La Jolla. There, (UPenn) to interrogate the role of intestinal epi- he developed probably the largest mouse muta- thelial cells in regulating homeostasis in the in- genesis program, and applied a forward genetic testine. He joined the faculty at the UPenn in approach to decipher the signaling pathways 2005 and was promoted to Associate Professor in activated by TLRs. In 2011, he returned to UT 2010. He has developed a successful research pro- Southwestern where he is currently Director of gram focused on dissecting the pathways that the Center for Genetics of Host Defense. Prior to regulate innate and adaptive immune cell func- the Nobel Prize, he received multiple awards, tion at barrier surfaces. In employing gnotobiotic such as the Shaw Prize (2011), the Balzan Prize mice, he also developed an active program to an- (2007) and the William B. Coley Award (2005). alyze immune cell interactions with commensal In 2008, he was elected member of the National and pathogenic bacteria and viruses at mucosal Academy of Sciences (USA) and the Institute of sites. Medicine.

Bruce Beutler, MD Maria Cristina Cuturi, MD

B.B. is a 2011 Nobel Prize Laureate and best M.C.C. obtained her MD at the University of known for his work on TLR4 sensing LPS. He re- Uruguay. She started her post-doc training at the ceived his MD from the University of Chicago in Immunology Department, University of Barce- 1981. From 1983–1986, he was post-doc and As- lona. She then joined G. Trinchieri, Wistar Insti- sistant Professor at the Rockefeller University, tute at the University of Pennsylvania, USA. In 1985, she moved as assistant scientist to INSERM Marc Y. Donath, MD U211/France. In 1994, she took over a tenured po- sition at INSERM U643. M.C.C. is known for her M.Y.D. is MD and Head of the Clinic for Endo- work in transplantation tolerance, and cellular crinology, Diabetes and Metabolism at the Uni- immunotherapy in solid organ transplantation. versity Hospital of Basel, Switzerland. His re- search focuses on the mechanisms and therapy of decreased insulin production in type 2 diabetes. Vishva Dixit, MD He has published several studies supporting the concept that a specific inflammatory process un- V.D. received his MD at the University Nairobi, derlies ␤-cell failure and apoptosis in the patho- Kenya, in 1980. He worked from 1981 to 1986 at genesis of diabetes. In particular, he uncovered the Department of Laboratory Medicine and the role of IL-1 ␤ in type 2 diabetes. He then initi- from 1986 to 1991 at the Department of Pathol- ated and conducted clinical studies aiming at di- ogy, Washington University School of Medicine. rect modulation of the immune system using IL-1 Subsequently, from 1986 to 1997 he worked at the antagonism for the treatment of type 2 diabetes. University of Michigan Medical School, first as Associate Professor and later as Professor at the Department of Pathology. In 1997, he accepted Richard Flavell, PhD the Directorship of Molecular Oncology at Gen- entech, and in 2007 took over the Department of R.F. is a British molecular biologist and immu- Physiological Chemistry. His laboratory unrav- nologist. He studied at the University of Hull, UK, eled not only apoptotic and NF- ␬B signaling as well as at the University of Amsterdam. Subse- pathways but also regulatory components of the quently, he went to Charlie Weismann in Zurich innate immune system, including adaptors that for his PhD thesis. He then worked in Amsterdam activate proinflammatory caspases. Worldwide from 1973 to 1979. In 1979, R.F. was appointed as attention received, for example, his observation head of the unit Structure and Function of Genes that caspase 11 ‘controls’ activation of caspase 1. at Mill Hill, London. In 1982, he left academia to become President of Biogen Corporation; in 1984, he became member of the Royal Society. Vishwa Deep Dixit, DVM, PhD In 1988, R.F. was appointed as founding Chairman of the Department of Immunobiology In 1994, V.D.D. obtained his DVM at the Hara- at Yale, USA, that consists of 13 primary faculty yana Agricultural University, Hisae, India. In members and retains strong support of the How- 2000, he received his PhD at University in Han- ard Hughes Medical Institute (HHMI), with 4 nover (Germany). He is now Associate Professor HHMI-funded investigators. Worldwide atten- of Immunobiology at Pennington Biomedical, tion received not only his work on the structure Baton Rouge, La., USA. His laboratory focuses on of human globine genes, but also on the struc- the molecular basis of immune senescence and ture-function relationship of CD40, and more re- inflammation in obesity and aging. In particular, cently, of NLRP genes that control gut homeosta- he is interested to identify the endogenous danger sis between commensal and pathogenic bacteria. signals that are recognized by specific sensors In 2002, he became member of the National like the NLRP-3 inflammasome, which in turn Academy of Science (USA). In 2012, he received regulate interleukin-1 (IL-1) and IL-18 produc- the prestigious William B. Coley Award for dis- tion. tinguished research in basic immunology.

Speakers at the Symposium 121

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) A n n e H a l l e , M D Department of Immunology, K.H. and Prof. T. Taniguchi discovered IRF-7 as master regulator A.H. received her MD at the Charité, Humboldt of type 1 interferon-dependent immune respons- University Berlin. In 2005, she started her post- es. Since 2008, his seminal discoveries revealed doc training at Charité’s Department of Experi- that T cells not only ‘learn’ in the thymus, but mental Neurology, and subsequently changed to also in the gut. Accordingly, ATP as well as gut’s D. Golenbock at the University of Massachusetts, ‘segmented filamentous bacteria’ drive Th17 cell USA. Since 2011, she works at the Center for Ad- differentiation in gut’s lamina propria, while in- vanced European Studies and Research (CAE- digenous Clostridium species favor induction SAR) in Bonn as MPI (Max-Plank Institute) re- of colonic Tregs. It follows that the commensal search group leader. There, she tackles the ques- microorganisms of the gut intimately interact tion how innate pattern recognition receptors with – and shape – the immune system of the re- drive chronic inflammation in Alzheimer dis- spective host. ease.

Veit Hornung, MD Michael Hoch, PhD V.H. is one of the first-rate talents out of the aca- M.H. studied Biology at the University of Heidel- demic school of Prof. Hartmann and Endres berg and received his PhD in Developmental Bi- (Munich) known to foster MD students. In 2003, ology at the LMU Munich in 1992. At the Bio- he received his MD at the LMU Munich and sub- physical Chemistry of the MPI, and later at MPI’s sequently undertook his post-doc training at Department of Molecular Developmental Biolo- both the University Munich (with G. Hartmann) gy, he was first post-doc and then group leader. In and the University of Massachusetts Medical 1999, he accepted professorship and chair of Mo- School (with E. Latz and D. Golenbock). Together lecular Developmental Biology at the Limes In- with G. Hartmann, he defined optimal RNA li- stitute, University Bonn. He uses the fruit fly gands for TLR7 and TLR8, and subsequently Drosophila as a genetic model to screen for novel identified 5 ؅ triphosphate RNA as ligand of the evolutionary conserved regulators of energy ho- cytosolic RNA helicase RIG-I. Together with E. meostasis and growth, such as key regulators of Latz and D. Golenbock, he was first to report that insulin signaling and body fat metabolism. A the cytosolic DNA sensor AIM2 activates the in- more recent focus lies on mechanism of cross- flammasome pathway. V.H. is now Professor at regulation of metabolism and innate immunity. the Institute of Clinical Chemistry and Clinical Here, the transcription factor FOXO3 appears to Pharmacology (headed by G. Hartmann) at the be a key player. Since 2005, he acts as speaker of University Bonn. the Collaborative Research Center (SFB) 645.

Michael Karin, PhD Kenya Honda, MD, PhD Born in Tel Aviv, Israel, M.K. studied biology and Since 1994, K.H. holds an MD (Kobe University microbiology at Tel Aviv University. He then School of Medicine, Japan) and since 2001 a PhD moved to the USA and received in 1979 his PhD (Medical Science, Kyoto University, Graduate at UCLA. There, he worked in the lab of Harvey School of Medicine). As Assistant Professor at the Herschman and studied genetic regulation of

122 Speakers at the Symposium

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) metallothioneins. Several positions – including cal School, Boston, Mass.) where he became As- work together with John Baxter on nuclear recep- sociate Professor. Since 1997, he has been Direc- tors and molecular endocrinology – led to a path tor of Dumont-UCLA TRS in Los Angeles. He is back to USC, Los Angeles, Calif. Today, he is Pro- Professor of Surgery at the David Geffen School fessor of Pharmacology at the University of Cali- of Medicine and holds a Chair in Immunobiolo- fornia (UCSD; La Jolla, Calif.). He was elected in gy and Transplant Research. His research is fo- 2005 to the National Academy of Sciences (USA). cused on the pathophysiology of liver ischemia M.K. has pioneered the link between chronic in- and reperfusion injury, and on the role of regula- flammation and cancer by studying NF-␬ B and tory T cells in transplant tolerance. He is an in- its regulation in cells gone awry. Furthermore, he ternationally recognized expert in experimental explored the role of NF- ␬ B in mouse models of organ transplantation. liver cancer.

Christian Kurts, MD Daniel Kastner, MD, PhD C.K. received in 2012 the Gottfried-Wilhelm- D.K. is Chief of the Laboratory of Clinical In- Leibniz Prize and is best known for his work on vestigation in the NIAMS, Clinical Director of cross-presentation. As fellow of the German Na- NIAMS, and Deputy Director for Intramural tional Academic Foundation, he received his MD Clinical Research at the NIH. His laboratory has at the University Gottingen. As post-doc with a long-standing interest in human genetic disor- Prof. J. Miller at the WEHI in Melbourne, Vic., ders of inflammation. He led an international Australia, and as Research Fellow with Prof. Car- consortium that, in 1997, identified the gene bone, Monash Medical School, Melbourne, he causing familial Mediterranean fever. His group published seminal work on mechanisms of T cell subsequently discovered that mutations in the tolerance. In 2003, he accepted the position as p55 tumor necrosis factor receptor cause a domi- Full Professor and Head of the Department of nantly inherited periodic fever syndrome that Molecular Immunology, University of Bonn. His they named TRAPS (TNF receptor-associated research interests include: mechanisms of anti- periodic syndrome). Based on these findings, gen cross-presentation, peripheral T and B cell D.K. proposed the term ‘autoinflammatory’ to tolerance, and the role of dendritic cells in (kid- describe the family of diseases characterized by ney) diseases. He is a member of the steering seemingly unprovoked episodes of inflammation committee ‘Excellence Cluster – Immunosensa- (without high titer autoantibodies or antigen- tion, University Bonn’, funded as of January 2013 specific T cells), which are now known to be dis- by the Federal Government of Germany. orders of the innate immune system.

Walter G. Land, MD Jerzy Kupiec-Weglinski, MD, PhD W.G.L. is (together with Prof. Konrad Messmer J.K.-W. graduated at Warsaw Medical Academy and Prof. Jürgen Reulen) offspring of the famous in 1974, and received his PhD at the Polish Acad- Munich academic school headed by the late Prof. emy of Sciences in 1979. He then joined the Walter Brendel. Today, he is Emeritus Professor Transplant Research Laboratories (TRS) at the at the LMU Munich. During his active career as Brigham and Women’s Hospital (Harvard Medi- an experimental and clinical transplant surgeon

Speakers at the Symposium 123

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) (1967–2004), he (co)pioneered clinical applica- ment of Pathology, University of Pittsburg School tion of anti-lymphocyte globulin, first success- of Medicine. He discovered that bacteria can ac- fully applied in 1968 to a heart-transplanted pa- tively respond to neuroendocrine hormones and tient in Cape Town, South Africa. He also discov- change both their rate of growth and production ered the phenomenon that a non-specific injury of virulence-associated factors. This led him to to human allograft (and not primarily its degree establish and to develop the field of microbial en- of foreignness) initiates and induces a specific docrinology, intersecting neurobiology and mi- acute allograft response. That in turn led to the crobiology. description of the feasibility of a clinically suc- cessful mycophenolate mofetil-based, calcineu- rin inhibitor-free immunosuppressive therapy. Luke O’Neill, PhD In the last years, he successfully introduced via reviews and books innate immunity into the field L.O. is Professor of Biochemistry at the Trinity of organ transplantation. College Dublin, Ireland. The major focus of his group is to provide a molecular understanding of innate immunity and inflammation. He (co)un- Eike Latz, MD raveled the molecular basis of signaling pathways driven by TLRs, Nod-like receptors (including E.L. studied Medicine in Göttingen and Berlin NLRP3), NF- ␬B, IRF family transcription factors (Germany). He then joined Prof. D. Golenbock and MAP kinases. He also studied the impact of in Boston, Mass., USA, and subsequently at the micro-RNA on these signal pathways. L.O. began University of Massachusetts (UMass) Medical his academic career as science student at the School, first as post-doc and later as Assistant Trinity College Dublin, where he is now Profes- Professor. He founded in 2007 the UMass Nano- sor. He went on to complete in 1988 his PhD in Medicine Institute. Since 2008, he is adjunct Full Pharmacology at the University London. One of Professor II at the Institute of Cancer Research the key findings emanating from L.O.’s lab was and Molecular Medicine, Norwegian University the discovery of a protein termed MAL, a master of Science and Technology. Since 2009, he is Pro- switch in TLR signaling. fessor and Director of the Institute of Innate Im- munity, University of Bonn. E.L. made multiple seminal contributions in the field of TLR signal- Eyal Raz, MD ing, cytoplasmatic sensors of nucleic acids, and NLRP-3-dependent activation of the inflamma- E.R. is Professor of Medicine at the University of some in health and diseases, such as arterioscle- California San Diego. He received his MD degree rosis and Alzheimer disease. in 1980 from the Hebrew University, Hadassah Medical School, Jerusalem, Israel. The current research in his laboratory focuses on the funda- Mark Lyte, PhD mentals of host interactions with environmental factors. His team studies host responses to micro- M.L. received his PhD at the Weizmann Institute bial ligands sensed by TLRs, and non-TLR stimu- of Science, Rehovot, Israel, in 1983. From 1983 to lation. In addition, he explores the role of mem- 1985, he worked as post-doc at the Department of bers of the transient receptor potential family by Microbiology and Immunology, Medical College which the host interacts with diverse chemical of Virginia, and from 1985 to 1986 at the Depart- and biochemical stimuli that originate from the

124 Speakers at the Symposium

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) environment. His laboratory explores the conse- was elected into the American Society of Clinical quences of these host-environmental interac- Investigation. In 2012, he organized an EMBO tions on the induction of inflammatory processes conference on Diabetes and Obesity, as well as a and immune responses. These include vaccines Keystone Meeting on Neural Control of Diabetes against various infectious agents, allergic diseas- and Obesity. He also received the prestigious es, inflammatory bowel diseases and cancer. Alexander von Humboldt Professorship Award. Since 2011, he is Director of the Institute of Dia- betes and Obesity, an integral component of the Jürgen Ruland, MD German National Diabetes Center, and chairs as Professor of Medicine the Division of Metabolic J.R. received his MD at the Justus Liebig Univer- Diseases, Technical University Munich. sity Giessen (Germany) in 1996. From 1997 to 1910, he undertook a post-doc training with Prof. Tak Mak at the Amgen Research Institute and Harald von Boehmer, MD, PhD Ontario Cancer Institute, Toronto, Ont., Canada. There, using genetic approaches he discovered H.v.B. is a German immunologist best known for the role of Malt-10 in TCR- and BCR-mediated his work on T lymphocytes. He obtained his MD signaling. In 2010, he acted as founder and Head at the Ludwig Maximilian University of Munich of the Institute of Molecular Immunology, Kli- (1968), and his PhD at Melbourne University nikum r. d. Isar, Technical University Munich. (1974). He was a member of the Basel Institute Since 2012, he is Professor and Head of the Insti- (1973–1996), where he discovered the intra-thy- tute for Clinical Chemistry and Pathobiochem- mic fate of antigen-specific T cells during ‘posi- istry, Klinikum r. d. Isar. J.R. pioneered our un- tive’ versus ‘negative’ selection. He headed Unité derstanding of innate anti-fungal immunity. INSERM 373 in Paris and is now Professor of Pa- Accordingly, upon recognizing fungal ligands, thology at Harvard Medical School and Chief of C-type lectins drive protective cytokine produc- the Laboratory for Lymphocyte Biology. At Har- tion either via a SYK kinase CARD9 pathway or vard, he found out how peripheral regulatory T interleukin-1 production via activation of the in- cells (Tregs) become induced. His present focus is flammasome. to translate the induction of Tregs into a vaccina- tion protocol against human type 1 diabetes. He received multiple awards, including the Kurt A. Matthias Tschöp, MD Körber Prize, the NIH Merit Award, and the Lou- is-Janet Prize for Medicine. M.T. received his MD at the LMU Munich in 1994. He spent 4 years as resident and research fellow at the Department of Neuroendocrinology Editors at the Munich University Hospital, before accept- ing a post-doc fellowship at the Eli Lilly Research Hermann Wagner, MD, PhD Laboratories, USA. There, he discovered the role of ghrelin in the control of food uptake, metabo- H.W. received his MD at the University of Tübin- lism and bodyweight. In 2002, he moved to the gen in 1967. In 1973, he obtained his PhD from German Institute of Nutrition, Potsdam (near Melbourne University, Australia. There, he spent Berlin), and in parallel started in 2003 a research a 3-year post-doc training with Sir Gus Nossal at program in Cincinnati, Ohio, USA. In 2009, he the Walter and Eliza Hall Institute. In his PhD

Speakers at the Symposium 125

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) work, he was first to describe the induction of cy- nition of short interfering RNAs, and RNA totoxic T cells in vitro. Back in Mainz, Germany, structures sensed by RIG-I. In 1994, G.H. re- with Prof. P. Klein (Medical Microbiology, Mainz ceived his MD from the University of Ulm, Ger- University), he headed the laboratory unit Cellu- many. From 1998 to 1999, he trained as post-doc lar Immunology. In 1983, he became Director of with A. Krieg, University of Iowa, USA. In Mu- the Department of Medical Microbiology and Im- nich, he headed the research group Therapeutic munology, University of Ulm. There, he studied Oligonucleotides within the Division of Clinical the immunobiology of bacterial super-antigens. Pharmacology (Prof. S. Endres), LMU Munich. In 1989, he moved from Ulm to the Technical In 2005, he was appointed Professor and Direc- University Munich as Director of the Institute tor of the Division of Clinical Pharmacology of of Medical Microbiology, Immunology and Hy- the University Hospital in Bonn; in 2007, he be- giene. Based on his observations (1999) that bacte- came Chair of the Institute of Clinical Chemis- rial DNA activated innate immune cells in a try and Clinical Pharmacology with the Central MYD88-dependent manner, a subsequent collab- Laboratory of the University Hospital Bonn, oration with S. Akira, Japan, unraveled in 2000 Germany. In 2000, G.H. received the Paul-Mar- that DNA is sensed by TLR9. In 2004, with his col- tini Award, in 2004 the Biofuture Award, and in leagues S. Bauer (now Marburg), H. Häcker (now 2009 the GOBio Award, both funded by the USA), C. Kirschning (now Essen) and H. Hoch- BMBF. Jointly with Percy Knolle, he successfully rein (Munich), he subsequently identified the nat- applied for funds from the state NRW which al- ural ligands of TLR7 and TLR8, and in 2012 that lowed to found the first Institute of Innate Im- of TLR13. As in Ulm (SFB Lympho-Haemopoese), munity in Germany at the University Hospital in H.W. initiated as speaker also in Munich a Collab- Bonn, now led by Prof. E. Latz. Jointly with sci- orative Research Center fostering Immunology. entists of the Medical Faculty, the Faculty of Since 2009, H.W. is TUM Emeritus of Excellence. Mathematics and Life Sciences, the Max-Planck- associated Center for Advanced European Stud- ies (CAESAR) and German Center for Neurode- Gunther Hartmann, MD, PhD generative Diseases, he initiated and now acts as speaker of the new Excellence Cluster Immuno- G.H. is a 2012 Gottfried-Wilhelm Leibniz Award Sensation (started in November 2012) which is Laureate and best known for his work on TLR9- part of the excellence program of the German stimulating CpG oligonucleotides, TLR7 recog- Research Council.

126 Speakers at the Symposium

Hartmann G, Wagner H (eds): Innate Immunity: Resistance and Disease-Promoting Principles. Else Kröner-Fresenius Symp. Basel, Karger, 2013, vol 4, pp 120–126 (DOI: 10.1159/000346503) Author Index

Artis, D. 59 Honda, K. 43 Nagano, Y. 43 Atarashi, K. 43 Hornung, V. 15 O’Neill, L.A.J. 73 Beckert, H. 86 Jentgens, E. 103 Bülow, M. 103 Ji, H. 29 Pahernik, S. VIII Pfluger, P.T. 96 Curtis, A.M. 73 Karin, M. 4 Cuturi, M.-C. 35 Kupiec-Weglinski, J.W. 29 Roth, S. 11 Kurts, C. 112 Ruland, J. 11 Dixit, V.D. 91 Donath, M.Y. 80 Land, W.G. 24 Suzuki, S. 103 Loch, G. 103 Elinav, E. 48 Lyte, M. 53 Takeyama, H. 103 Tanoue, T. 43 Flavell, R.A. 48 Mankan, A.K. 15 Thomas, C. 11 Messmer, K. 24 Tittel, A.P. 112 Garbi, N. 112 Monticelli, L.A. 59 Tschöp, M.H. 96 Grivennikov, S. 4 Moreau, A. 35 Mori, T. 103 Wagner, H. 1 Halle, A. 86 Müller, T.D. 96 Wang, K. 4 Hartmann, G. 1 Henao-Meija, J. 48 Zhai, Y. 29 Heuser, C. 112 Zinke, I. 103 Hoch, M. 103

127

Subject Index

Adipose tissue macrophage, see Macrophage Card9, C-type lectin receptor signaling 12 Agouti-related peptide, body weight regulation 97 β-Cell, see Diabetes type 2 Airway hyperreactivity, innate lymphoid cell Cholecystokinin, gut-brain communication in body modulation 62, 63 weight regulation 98, 99 Alzheimer’s disease Colorectal cancer epidemiology 86 inflammation pathways interleukin-1β role 87, 88 colitis-associated cancer microglia response 87 STAT3 signaling 7 misfolded proteins Th17 cells in promotion 7 inflammasome activation 88, 89 interleukin-6 5–8 overview 86, 87 nuclear factor-κB 5 γ-Aminobutyric acid, microbial endocrinology 55, 56 tumor necrosis factor 5 Amylin, gut-brain communication in body weight inflammatory bowel disease association 4, 5, 50 regulation 98 C-type lectin receptor Amyloid-β, see Alzheimer’s disease danger recognition 11, 12 Antimicrobial peptides effector pathways 12, 13 Drosophila immune response brain receptors 109 Dectin, see C-type lectin receptor FOXO regulation 106–108 Dendritic cell overview 105 bacterial pyelonephritis studies in mouse model types and functions 104 tolerogenic cells animal studies Bacterial pyelonephritis allograft studies of tolerance-induction dendritic cell response and prospects for study 113, mechanisms 37, 38 117, 118 generation ex vivo 36, 37 epidemiology 112 human studies mouse models clinical application prospects 38, 39, 41 innate immune response 113, 114 clinical trials 38 neutrophilia good manufacturing practice 39 bacterial clearance improvement 114, 115 kidney transplant studies of cell therapy 40 CD11c.DTR transgenic mouse studies 116 safety assays 39, 40 CD11c.LuciDTR transgenic mouse studies rationale for organ transplant use 36 116 Diabetes type 2 granulocyte release from bone marrow in inflammasome sensing of metabolic stress 50, 51, 81 induction 115 inflammation recruitment of polymorphonuclear cells by evidence for β-cell failure role 81 dendritic cells 115 insulitis in obesity and diabetes 82, 83 overview 113 overview of role 80 Bcl10, C-type lectin receptor signaling 12 interleukin-1β inflammation regulation 82 lung cells islet expression regulation by nutrients 81, 82 epithelial repair promotion 63, 64 therapeutic targeting 83 identification in mice and humans 61, 62 Dopamine, microbial endocrinology 55 markers and classification 60 Drosophila innate immunity, see Antimicrobial Insulin peptides, Gut microbiota Drosophila innate immunity signaling 105, 106 gut-brain communication in body weight regulation Epinephrine, microbial endocrinology 54 97, 98 Insulitis, see Diabetes type 2 Ghrelin, gut-brain communication in body weight Interleukin-1β regulation 97 Alzheimer’s disease role 87, 88 Glucagon, gut-brain communication in body weight C-type lectin receptor pathway 12, 13 regulation 99 diabetes type 2 Glucagon-like peptide 1, gut-brain communication in inflammation regulation 82 body weight regulation 99, 100 islet expression regulation by nutrients 81, 82 Gut microbiota therapeutic targeting 83 Clostridia maturation overview 43, 44 caspase 1-independent release 18, 19 regulatory T cell induction 45, 46 inflammasome complex 17, 18 Drosophila innate immunity regulation 108, 109 neutrophil secretion 20, 21 innate lymphoid cells in homeostasis 67 overview 17 microbial endocrinology 53–57 Interleukin-6, colorectal cancer role 5–8 regulatory T cell function in gut 44, 45 Interleukin-10, innate immunity activation after liver transplantation 32 Helminth infection, innate lymphoid cell response 64 Interleukin-22, innate lymphoid cell expression 66–68 Heme oxygenase-1, tolerogenic dendritic cell Ischemia-reperfusion injury induction 37 innate immunity activation after liver transplantation Hepatitis, hepatocellular carcinoma association 4 interleukin-10 signaling 32 Histamine, microbial endocrinology 56 NLRP3 31, 32 T cell role 32, 33 Inflammasome Toll-like receptors 30, 31 activation and function 48, 49 organ transplant failure 29, 30 Alzheimer’s disease and misfolded protein activation 88, 89 Kidney infection, see Bacterial pyelonephritis interleukin-1β maturation 17, 18 Kidney transplantation intestinal mucosal immune response 49, 50 superoxide dismutase clinical trials 24, 25, 27 metabolic disorder regulation 50, 51, 81 tolerogenic dendritic cell therapy 40 obesity role 91–94 Inflammatory bowel disease Lectin receptor, see C-type lectin receptor colorectal cancer association 4, 5, 50 Lipopolysaccharide, microRNA induction 75, 76 inflammasome 49, 50 Liver transplantation, see Ischemia-reperfusion injury Innate lymphoid cell commensal bacteria homeostasis role 67 Macrophage, adipose tissue macrophage and obesity 91 functional overview 59, 60 Malt1, C-type lectin receptor signaling 12 group 2 cells Metabolic syndrome, inflammasome mediation 51 airway inflammation regulation 62, 63 Microbial endocrinology, see Gut microbiota helminth immune response 64 Microbiota, see Gut microbiota identification 60, 61 MicroRNA therapeutic targeting 64, 65 functional overview 73, 74 group 3 cells lipopolysaccharide induction 75, 76 enteric pathogen protection 66 miR-155 in cancer 76 extraintestinal tissue regeneration promotion 66 p300 interactions 77 identification 65 therapeutic targeting 76 therapeutic targeting 67, 68 Toll-like receptor modulation 74–77

129 NALP, history of study 2 inflammasome role 91–94 Natural killer cell, see also Innate lymphoid cell insulitis in obesity and diabetes 82, 83 innate immunity activation after liver transplant cell Organ transplantation, see Dendritic cell, Ischemia- 33 reperfusion injury, Kidney transplantation Neuropeptide Y, body weight regulation 97 Oxyntomodulin, gut-brain communication in body Neutrophil, see also Bacterial pyelonephritis weight regulation 100, 101 functional overview 15–17 interleukin-1β maturation and secretion 17–21 p300, microRNA interactions 77 NLRC4, colorectal cancer role 50 Pyelonephritis, see Bacterial pyelonephritis NLRP3, see also Inflammasome Alzheimer’s disease activation 88, 89 Regulatory T cell C-type lectin receptor cross talk 13 function in gut 44, 45 innate immunity activation after liver transplantation gut microbiota induction in intestine 45, 46 31, 32 interleukin-1β maturation role 18, 19 Serotonin, microbial endocrinology 55 intestinal mucosal immune response 49 STAT3, colitis-associated cancer role 7 obesity role 92–94 Step, Drosophila innate immunity 105 NLRP6 Superoxide dismutase, clinical trials in kidney history of study 2, 3 transplantation 24, 25, 27 intestinal mucosal immune response 49, 50 Syk, activation by C-type lectin receptors 12 Non-alcoholic fatty liver disease, inflammasome mediation 50, 51 T cell, innate immunity activation after liver Norepinephrine, microbial endocrinology 54, 55 transplantation 32, 33 Nuclear factor-κB, colorectal cancer role 5 Th17 cell, colitis-associated cancer promotion 7 Tolerogenic dendritic cells, see Dendritic cell Obesity Toll-like receptors gut-brain communication in body weight regulation Drosophila innate immunity 104 amylin 98 functional overview 75 cholecystokinin 98, 99 history of study 1, 2, 26, 27 ghrelin 97 innate immunity activation after liver transplantation glucagon 99 30, 31 glucagon-like peptide 1 99, 100 microRNA control 74–77 insulin 97, 98 Tumor necrosis factor, colorectal cancer role 5 overview 96, 97 oxyntomodulin 100, 101

130 Else Kröner-Fresenius Symposia Editor: S. Pahernik ISSN 1421–5721

1 Molecular Mechanisms of Adult Stem Cell Aging Editor: K.L. Rudolph, Ulm VIII + 108 p., 21 fig., 19 in color, hard cover, 2010. ISBN 978–3–8055–9243–7

2 Nanomedicine – Basic and Clinical Applications in Diagnostics and Therapy Editor: C. Alexiou, Erlangen VIII + 202 p., 80 fig., 56 in color, 11 tab., hard cover, 2011. ISBN 978–3–8055–9818–7

3 Advances in Stem Cell Aging Editor: K.L. Rudolph, Ulm VIII + 126 p., 20 fig. in color, hard cover, 2012. ISBN 978–3–318–02170–7

4 Innate Immunity: Resistance and Disease-Promoting Principles Editors: G. Hartmann, Bonn; H. Wagner, Munich VIII + 130 p., 13 fig., 10 in color, hard cover, 2013. ISBN 978–3–318–02347–3

Our understanding of the complex innate immune response is increasing rapidly. Its role in the protection against viral or bacterial pathogens is essential for the survival of an organism. However, it is equally important to avoid unregulated inflammation because innate immune responses can cause or promote chronic autoinflammatory diseases such as gout, atherosclerosis, type 2 diabetes or certain aspects of the met- abolic syndrome. In this book leading international experts in the field of innate immunity share their findings, define the ‘state of the art’ in this field and evaluate how insight into the molecular basis of these diseases could help in the design of new therapies. A tre- mendous amount of work on the innate immune response has been done over the last fifteen years, culminating in the 2011 Nobel Prize in Physiology/Medicine award- ed for the discoveries of Toll genes in immunity in flies, membrane-bound Toll-like receptors in mammals, and dendritic cells as initiators of adaptive immunity.