Innate lymphocyte responses in inflammatory diseases

(1)

From the Center for Infectious Medicine, Department of Medicine, Huddinge Karolinska Institutet, Stockholm, Sweden

INNATE LYMPHOCYTE RESPONSES IN INFLAMMATORY DISEASES

Johanna Emgård

Stockholm 2022

(2)

All previously published papers were reproduced with permission from the publisher.

Published by Karolinska Institutet.

Printed by Universitetsservice US-AB, 2022

Cover illustration: By Johanna Emgård

(3)

Innate lymphocyte responses in inflammatory diseases THESIS FOR DOCTORAL DEGREE (Ph.D.)

By

Johanna Emgård

The thesis will be defended in public at Karolinska Institutet, room 9Q level 9, Alfred Nobels Allé 8, Huddinge, 25^th of November 2022

Principal Supervisor:

Professor Johan Sandberg Karolinska Institutet

Department of Medicine, Huddinge Center for Infectious Medicine Co-supervisor(s):

Professor Anna Norrby-Teglund Karolinska Institutet

Department of Medicine, Huddinge Center for Infectious Medicine

Associate professor Niklas Björkström Karolinska Institutet

Department of Medicine, Huddinge Center for Infectious Medicine

Opponent:

Associate professor Andrew Hogan Maynooth University

Department of Biology Examination Board:

Professor Susanna Cardell University of Gothenburg

Department of Microbiology and Immunology Associate professor Keira Melican

Karolinska Institutet

Department of Neuroscience Professor Miklós Lipcsey Uppsala University

Department of Surgical Sciences

(4)

(5)

To Filippa

(6)

(7)

POPULAR SCIENCE SUMMARY OF THE THESIS

The human body is constantly exposed to external threats, such as bacteria and viruses. The immune system is an elegant system which has evolved to protect us and combat these threats.

In most cases the immune response is well coordinated but, in some cases, it fails to clear the pathogen and instead gives rise to inflammatory conditions which can damage the body. In this thesis, I discuss three such conditions; inflammatory bowel disease (IBD), sepsis, and toxic shock syndrome (TSS).

IBD is a chronic inflammatory disease of the intestine which causes huge suffering for the patient. Many different factors contribute to the disease such as the microbiota, environmental factors, genetic factors, and the immune system. This makes it a very complex disease and despite many years of intensive research, no cure has been found. In Paper I, we investigate the role of immune cells called ILC3 in IBD. We find that in the healthy gut, ILC3s are important for the formation of small lymphoid structures which are important for the normal function of the immune system in the gut. When inflammation occur however, the ILC3s move out into the tissue and recruit other immune cells which increase the inflammation.

In paper II-IV, we explore the role of another type of immune cells called MAIT cells in sepsis and TSS. Sepsis is a leading cause of death from infection. Sepsis arises when the immune system does not function properly and fails to combat the infection. Sepsis arising from infected wounds or cuts are well known, but even more common is sepsis caused by a lung or urinary tract infection. In sepsis, immune cells throughout the body are overactivated which leads to organ damage. Sepsis progresses fast, often within hours, and therefore early diagnosis is key to survival. TSS is a rare condition with high mortality rates, which is caused by toxins called superantigens, produced by bacteria called group A streptococci (GAS) and Staphylococcus aureus. Superantigens trick the immune cells to first get massively overactivated and then exhausted. The overactivated immune cells start to produce large amounts of inflammatory proteins called cytokines which damage the organs of the body and can cause death. Similar to IBD, sepsis and TSS are complex conditions, which can look different in different individuals.

Therefore, it has been difficult to find effective treatments and a better understanding of the immune response is needed. In paper II, we found that MAIT cells are very sensitive to superantigens and even though they are few in number, they are major producers of cytokines in TSS. As the cytokines cause the organ damage in TSS, these results suggest that stopping MAIT cells from producing cytokines could be a possible way to treat TSS patients. In paper III, we found that the high production of cytokines by MAIT cells can be blocked by adding antioxidants. This suggests that antioxidants could be a possible treatment in TSS. In paper IV, we found that MAIT cells were highly activated in sepsis as well, especially in the patients with severe organ damage and in patients who died. These results suggest that measuring the level of MAIT cell activation in the clinic may be used to predict how sick the patient is going to be. It is also possible that blocking the MAIT cells could be a possible way of treating the patients.

(8)

(9)

POPULÄRVETENSKAPLIG SAMMANFATTNING

Kroppen är konstant utsatt för yttre hot, så som bakterier och virus. Immunförsvaret är ett elegant system som har utvecklats för att förvara oss mot dessa hot. I de flesta fall är immunsvaret välkoordinerat, men i vissa fall misslyckas det med att avlägsna bakterien eller viruset. Detta kan leda till utveckling av inflammatoriska tillstånd som skadar den egna kroppen. I den här avhandlingen kommer jag att diskutera tre sådana tillstånd, inflammatorisk tarmsjukdom (IBD), sepsis och toxiskt chocksyndrom (TSS).

IBD är en kronisk, inflammatorisk sjukdom i tarmen som orsakar mycket lidande för patienten.

Många olika faktorer bidrar till sjukdomsutvecklingen, så som tarmfloran, miljön, genetiken, och immunförsvaret. Detta gör att sjukdomen är mycket komplex och trots många år av intensiv forskning har ännu inget botemedel hittats. I Delarbete I undersöker vi hur immunceller som kallas ILC3 bidrar till utvecklingen av IBD. Vi upptäckte att ILC3 behövs för bildande av små kluster av immunceller i tarmen. Dessa kluster är viktiga för att immunförsvaret i tarmen ska fungera. Vid inflammation upptäckte vi att ILC3 vandrar ut i vävnaden och rekryterar andra immunceller vilket förvärrar inflammationen.

I Delarbete II-IV undersöker vi hur en annan typ av immunceller, MAIT-celler, bidrar till sjukdomsförloppet vid sepsis och TSS. Sepsis, tidigare känt som blodförgiftning, är en av de vanligaste dödsorsakerna till följd av en infektion. Sepsis kan utvecklas om immuncellerna inte fungerar som de ska och misslyckas med att bekämpa infektionen. Många känner till att sepsis kan uppstå från infekterade sår (som i Astrid Lindgrens ”Emil i Lönneberga”), men de vanligaste orsakerna är lung- och urinvägsinfektioner. Vid sepsis överaktiveras immunceller i hela kroppen och organen tar skada. Sepsis utvecklas snabbt, ofta inom loppet av några timmar och därför är tidig diagnos avgörande för överlevnad. TSS är ett ovanligt men mycket dödligt tillstånd som orsakas av gifter som kallas superantigener. Dessa produceras av bakterier som kallas grupp A streptokocker (GAS) och Staphylococcus aureus. Superantigenerna lurar immuncellerna så att de först blir kraftigt överaktiverade och därefter utmattade.

Överaktiveringen leder till produktion av stora mängder inflammatoriska proteiner som kallas cytokiner. De höga nivåerna av cytokiner skadar kroppens organ och kan leda till död. Precis som IBD är sepsis och TSS mycket komplexa tillstånd som kan se olika ut i olika individer.

Därför är det svårt att hitta effektiva botemedel och vi behöver en bättre förståelse för hur immunförsvaret fungerar i dessa sjukdomar. I Delarbete II upptäckte vi att MAIT-celler är extra känsliga för superantigener och trots att de är få i antal är de bland de största producenterna av cytokiner. Att stoppa MAIT-cellerna från att producera cytokiner skulle därför kunna vara en möjlig behandlingsstrategi vid TSS. I Delarbete III upptäckte vi att man kan blockera MAIT-cellernas cytokinproduktion genom att tillsätta antioxidanter.

Antioxidanter skulle därför kunna vara en möjlig behandling vid TSS. I Delarbete IV såg vi att MAIT-celler även var kraftigt aktiverade i sepsis, framför allt i de patienter med svårast organskada och i de patienter som dog. Dessa resultat indikerar att man genom att mäta MAIT- cellers aktiveringsgrad eventuellt skulle kunna förutse hur sjuk en patient kommer bli. Det är också möjligt att MAIT-celler skulle kunna fungera som en måltavla vid behandling av sepsis.

(10)

(11)

ABSTRACT

The immune system is a highly sophisticated system that has evolved to protect the host from external and internal threats. However, under some circumstances the immune response can get dysregulated and cause inflammatory diseases. In this thesis, I focus on three sets of disorders resulting from dysregulated immune responses: sepsis, toxic shock syndrome (TSS), and inflammatory bowel disease (IBD). They are highly heterogenous in their presentation and immune response profile and so far, there are no treatment options specifically targeting these conditions. The aim of this thesis is therefore to gain deeper insight into the immune mechanisms underlying these disorders. We have focused on two types of immune cells which are early responders in inflammation and share features of both the innate and adaptive immune system: group 3 innate lymphoid cells (ILC3) and mucosal-associated invariant T (MAIT) cells.

The aim of paper I was to explore the role of the receptor GPR183 on ILC3s in the development of lymphoid structures in the colon at steady state and in promoting cell migration and tissue reorganization during colonic inflammation. Using mouse models and human IBD samples, we found that GPR183 on ILC3s senses oxysterols and is essential for the formation of cryptopatches and innate lymphoid follicles. In inflamed colon of both mice and humans, the oxysterol production in the surrounding tissue increased, resulting in migration of ILC3s out of the lymphoid tissues and towards the top of the colonic folds where they promoted recruitment of other immune cells.

In paper II and III, we aimed to investigate the contribution of MAIT cells to the pro- inflammatory cytokine storm in TSS. In paper II, ex vivo stimulation experiments revealed that despite their low frequencies, MAIT cells were major producers of pro-inflammatory cytokines in streptococcal TSS (STSS). Superantigens produced by group A streptococci (GAS) bound to the b chain of the MAIT TCR, resulting in activation of a large fraction of the MAIT cell pool. MAIT cells were also found to be highly activated in patients with STSS. In paper III, we next explored the role of redox signaling in the strong MAIT cell response in TSS. We found that the MAIT cell cytokine response was inhibited by antioxidant treatment while other T cells were unaffected. Antioxidants reduced the total IL-18 production by PBMCs, and blocked IL-12 and IL-18-mediated MAIT cell activation.

In paper IV, we investigated the MAIT cell phenotype in early samples from sepsis patients in the emergency room to evaluate the potential of MAIT cells as prognostic or diagnostic markers. MAIT cells were activated and reduced in frequency early during the sepsis response.

Expression of the activation marker CD69 on MAIT cells was associated with lymphopenia, organ dysfunction and increased mortality.

To summarize, this thesis reveals important novel insights on the roles of ILC3s in colonic inflammation and MAIT cells in sepsis and TSS.

(12)

(13)

LIST OF SCIENTIFIC PAPERS

I. Emgård J*, Kammoun H*, García-Cassani B, Chesné J, Parigi SM, Jacob JM, Cheng HW, Evren E, Das S, Czarnewski P, Sleiers N, Melo-Gonzalez F, Kvedaraite E, Svensson M, Scandella E, Hepworth MR, Huber S, Ludewig B, Peduto L, Villablanca EJ, Veiga-Fernandes H, Pereira JP, Flavell RA, Willinger T (2018) Oxysterol Sensing through the Receptor GPR183 Promotes the Lymphoid-Tissue-Inducing Function of Innate Lymphoid Cells and Colonic Inflammation. Immunity 48(1):120-132.e8

II. Emgård J, Bergsten H, McCormick JK, Barrantes I, Skrede S, Sandberg JK, Norrby-Teglund A (2019) MAIT Cells Are Major Contributors to the Cytokine Response in Group A Strepptococcal Toxic Shock Syndrome. PNAS 116(51):25923-25931

III. Emgård J, Norrby-Teglund A, Sandberg JK. Reactive oxygen species are required for MAIT cell activation in response to group A streptococcal and staphylococcal superantigens. Manuscript

IV. Emgård J, Filipovic I, Unge C, Palma Medina LM, Bergsten H, Parke Å, Moll K, Dzidic M, Öcenzi V, Björkström N, Svensson M, Sandberg JK, Strålin K, Norrby-Teglund A. Early MAIT cell activation is associated with sepsis clinical outcome. Manuscript

*equal contribution

(14)

SCIENTIFIC PAPERS NOT INCLUDED IN THE THESIS

I. Lourda M, Dzidic M, Hertwig L, Bergsten H, Palma Medina LM, Sinha I, Kvedaraite E, Chen P, Muvva JR, Gorin JB, Cornillet M, Emgård J, Moll K, García M, Maleki KT, Klingström J, Michaëlsson J, Flodström-Tullberg M, Brighenti S, Buggert M, Mjösberg J, Malmberg KJ, Sandberg JK, Henter JI, Folkesson E, Gredmark-Russ S, Sönnerborg A, Eriksson LI, Rooyackers O, Aleman S, Strålin K, Ljunggren HG, Björkström NK, Svensson M, Ponzetta A, Norrby-Teglund A, Chambers BJ; Karolinska KI/K COVID-19 Study Group (2021) High-dimensional profile reveals phenotypic heterogeneity and disease-specific alterations of granulocytes in COVID-19. PNAS 118(40):e2109123118

II. Parrot T, Gorin JB, Ponzetta A, Maleki KT, Kammann T, Emgård J, Perez- Potti A, Sekine T, Rivera-Ballesteros O; Karolinska COVID-19 Study Group, Gredmark-Russ S, Rooyackers O, Folkesson E, Eriksson LI, Norrby-Teglund A, Ljunggren HG, Björkström NK, Aleman S, Buggert M, Klingström J, Strålin K, Sandberg JK (2020) MAIT cell activation and dynamics associated with COVID-19 disease severity. Science Immunology 5(51):eabe1670

III. Sekine T, Perez-Potti A, Rivera-Ballesteros O, Strålin K, Gorin JB, Olsson A, Llewellyn-Lacey S, Kamal H, Bogdanovic G, Muschiol S, Wullimann DJ, Kammann T, Emgård J, Parrot T, Folkesson E; Karolinska COVID-19 Study Group, Rooyackers O, Eriksson LI, Henter JI, Sönnerborg A, Allander T, Albert J, Nielsen M, Klingström J, Gredmark-Russ S, Björkström NK, Sandberg JK, Price DA, Ljunggren HG, Aleman S, Buggert M (2020) Robust T cell Immunity in Convalescent Individuals with Asymptomatic or Mild COVID-19. Cell 183(1):158-168.e14.

IV. Sobkowiak MJ, Davanian H, Heymann R, Gibbs A, Emgård J, Dias J, Aleman S, Krüger-Weiner C, Moll M, Tjernlund A, Leeansyah E, Sällberg Chen M, Sandberg JK (2019) Tissue-resident MAIT cell populations in human oral mucosa exhibit an activated profile and produce IL-17. European Journal of Immunology 49(1):133-143

V. Dias J, Boulouis C, Sobkowiak MJ, Lal KG, Emgård J, Buggert M, Parrot T, Gorin JB, Leeansyah E, Sandberg JK (2018) Factors Influencing Functional Heterogeneity in Human Mocosa-Associated Invariant T Cells. Frontiers in Immunology 9:1602

(15)

LIST OF ABBREVIATIONS

5-OE-RU 5-(2-oxoethylideneamino)-6-D-ribitylaminouracil 5-OP-RU 5-(2-oxopropylideneamino)-6-d-ribitylaminouracil 7a,25-OHC 7α,25-hydroxycholesterol

AP-1 APC ASC BCG BCR CCR CD CDX CH25H COVID-19 CRP CYP7B1 DAMP ELISA FACS GAS GATA3 GCL GFP GGS GM-CSF GPR183 GSH GSSG GWAS HIV HLA

Activator protein-1 Antigen-presenting cell

Apoptosis-associated speck-like protein containing a CARD Mycobacterium bovis bacillus Calmette-Guérin

B cell receptor

C-C motif chemokine receptor Crohn’s disease

Cluster of differentiation X (X=number) Cholesterol 25-hydroxylase

Coronavirus disease 2019 C-reactive protein

Cytochrome P450, family 7, subfamily b, polypeptide 1 Danger-associated molecule patterns

Enzyme-linked immunosorbent assay Fluorescence-activated cell sorting Group A streptococci

GATA-binding protein 3 Glutamate cysteine ligase Green fluorescent protein Group G streptococci

Granulocyte-macrophage colony-stimulating factor G-protein coupled receptor 183, also known as EBI2 Glutathione

Glutathione disulfide

Genome-wide association study Human immunodeficiency virus Human leucocyte antigen

(18)

HLA-DR HMGB1 IBD ICU IFN IL IL-XR ILC ILF IV IVIg JAML KO LAG-3 LPS LT LTi cell LukED MACS MAIT cell MHC MLN MR1 mtDNA mTORC1 NAC NADPH NCR NET NF-kB NFAT

Human leucocyte antigen – DR isotype High-mobility group protein B1

Inflammatory bowel disease Intensive care unit

Interferon Interleukin

Interleukin receptor for interleukin X (X=number) Innate lymphoid cell

Isolated lymphoid follicle Intravenous

Intravenous immunoglobulin G

Junctional adhesion molecule-like molecule Knock out

Lymphocyte-activation gene 3 Lipopolysaccharide

Lymphotoxin

Lympoid tissue inducer cell Leukotoxin ED

Magnetic-activated cell sorting Mucosal-associated invariant T cell Major histocompatibility complex Mesenteric lymph node

MHC class I-related protein 1 Mitochondrial DNA

Mammalian target of rapamycin complex 1 N-acetyl cysteine

Nicotinamide adenine dinucleotide phosphate Natural cytotoxicity receptors

Neutrophil extracellular traps Nuclear factor kB

Nuclear factor of activated T cells

(19)

NK cell NLRP3

NOD2 NOX NSTI OXPHOS PAMP PBMC PBS PD-1 PLZF PRR qPCR Rag1 RORgt ROS

SARS-CoV-2 SEB

SLC7A5 SOD SOFA Spe STAT4 STSS T-BET TCR TEMRA

Th TIM-3 TLR

Natural killer cell

Nucleotide-binding domain, leucine-rich-containing family, pyrin domain-containing-3

Nucleotide-binding oligomerization domain-containing 2 NADPH oxidase

Necrotizing soft tissue infection Oxidative phosphorylation

Pathogen-associated molecular pattern Peripheral blood mononuclear cell Phosphate-buffered saline

Programmed cell death-1

Promyelocytic leukemia zinc finger Pattern recognition receptor

Quantitative polymerase chain reaction Recombination activating gene 1 Retinoid-related orphan receptor-γt Reactive oxygen species

Severe acute respiratory syndrome coronavirus-2 Staphylococcal enterotoxin B

Solute Carrier Family 7 Member 5 Superoxide dismutase

Sequential organ failure sssessment Streptococcal pyrogenic exotoxin

Signal transducer and activator of transcription 4 Streptococcal toxic shock syndrome

T-box transcription factor, also known as TBX21 T cell receptor

Effector memory T cells re-expressing CD45RA T helper

T cell immunoglobulin and mucin-3 Toll-like receptor

(20)

TNF TRAV Treg TSS TSST-1 UC WHO WT

Tumor necrosis factor

T Cell Receptor Alpha Variable Regulatory T cell

Toxic shock syndrome

Toxic shock syndrome toxin-1 Ulcerative colitis

World Health Organisation Wild type

(21)

1 INTRODUCTION

1.1 INNATE AND ADAPTIVE IMMUNITY

The immune system is an elegant system evolved to protect the body from external and internal threats. When exposed to an insult, such as pathogens, the immune system elicits an inflammatory response which serves to eliminate the pathogen, remove damaged cells, and initiate tissue repair. The immune system is traditionally divided into an innate and an adaptive arm. The innate immune system consists of physical barriers, such as skin and the mucosal lining of the gut and lungs, innate immune cells, such as neutrophils, monocytes, macrophages, dendritic cells, mast cells, basophils, and eosinophils, and humoral components, such as the complement system (Figure 1). Furthermore, soluble proteins including cytokines and interferons serve to limit pathogen invasion if barriers are breached. The innate immune system is old and highly evolutionarily conserved across all vertebrates. Even invertebrates have an innate defense system sharing many features with the innate immune system in vertebrates, such as phagocytic cells and cytokines (1). Innate immune cells are important players during the early phase of inflammation. They recognize a broad range of pathogens and stimuli, produce inflammatory cytokines, and kill and engulf damaged or infected cells. If the innate immune system fails to clear the pathogen, adaptive immune cells are also recruited and elicit an antigen-specific response to the pathogen. The adaptive immune system is evolutionarily younger than the innate system and is only found in vertebrates (2). B and T cells mediate the cellular response of adaptive immunity (Figure 1). B and T cells recognize antigens with high specificity due to the somatic recombination of their B cell receptor (BCR) and T cell receptor (TCR). Each B or T cell clone recognizes a specific antigen. The multitude of T and B cell clones in the body therefore enables the recognition of a vast number of antigens by the adaptive immune system. The BCR exists in a membrane bound form but can also be secreted as an effector antibody. Antibodies are key components of humoral adaptive immunity. After specific antigen recognition, the specific B or T cell clones rapidly proliferate to generate large numbers of cells recognizing that particular antigen. Once the insult is cleared, antigen-specific cells form long-term memory cells which are able to respond rapidly if the body is exposed to the same threat again. The ability to establish long-term memory is one of the key traits of the adaptive immune system. In between the innate and adaptive immune systems are cell types that share similarities with both traditional innate and adaptive cells (Figure 1). Similar to B cells and conventional T cells, such cells belong to the lymphoid lineage but in contrast they respond in an innate-like manner. Innate lymphoid cells (ILC) share many transcriptional and functional features with T helper (Th) cells; however, they lack antigen specific TCRs (3).

Unconventional T cells, such as mucosa-associated invariant T (MAIT) cells, natural killer T (NKT) cells, and gd T cells, possess conserved TCRs which recognize a limited set of non- peptide antigens. This allows them to rapidly respond in large numbers. In addition, unconventional T cells can respond to some pro-inflammatory cytokines in an innate manner.

Therefore, they are activated early upon pathogen encounter and produce effector molecules such as cytokines and cytotoxic molecules. However, the strong and broad activation of

(22)

unconventional T cells can pose a danger to the host under some circumstances. In this thesis, I will explore the role of group 3 ILCs and MAIT cells in inflammatory disorders.

Figure 1. Cellular components of innate and adaptive immunity. ILCs and unconventional T cells, such as MAIT cells share features of both the innate and the adaptive immune system.

1.2 INFLAMMATORY DISEASES

Inflammatory processes play an important role in the pathogenesis and outcome of many diseases. In most cases, inflammatory responses are well coordinated, and an appropriate number of immune cells required to clear the pathogen insult are recruited and activated. Under optimal conditions, the inflammation has limited impact on the homeostasis of the tissue and does not cause any major physiological alterations. However, many pathogens can trigger exaggerated immune cell activation and the resulting inflammation can cause local or even systemic tissue damage. Failure to regulate the immune response or to clear the pathogen can also result in prolonged chronic inflammation. This can lead to the development of other diseases such as cancer or autoimmune disorders. In this thesis, I will explore the role of group 3 ILCs and MAIT cells in three inflammatory conditions: inflammatory bowel disease, sepsis, and toxic shock syndrome.

(23)

1.2.1 INFLAMMATORY BOWEL DISEASE

Inflammatory bowel disease (IBD) is a global health problem with increasing prevalence in high-income countries but also in newly industrialized countries (4). The term IBD encompasses a set of chronic inflammatory disorders of the gastrointestinal tract (5). The main types of IBD are ulcerative colitis (UC), which primarily affects the lower part of the colon and rectum, and Crohn’s disease (CD), which presents as more patchy inflammation in any part of the gastrointestinal tract, including the mouth, esophagus, stomach, small intestine, colon, and anus. Common symptoms of IBD are diarrhea, rectal bleeding, fatigue, abdominal pain, and weight loss. IBD can also give rise to cancer and other chronic inflammatory disorders. The pathobiology of IBD is complex and environmental, genetic, and microbial factors contribute to the dysregulated immune responses responsible for the chronic inflammation in both CD and UC. To date there is no cure for IBD, and the current treatments rely predominantly on immunosuppressive drugs (6).

Early microbial colonization is important for the development, education, and maturation of the immune system in the gut (7). Alterations in microbial composition is therefore an important contributor to the dysregulated immune response in IBD. Dysbiosis of the microbiota with increased abundance of selected microbes and decreased complexity is seen in both CD and UC patients (8). The abundance of Proteobacteria and Bacteroides is increased in CD, while the abundance of Firmicutes is decreased. Increased intestinal permeability is also associated with CD allowing for microbial products to breach the gut wall and enter the tissue (9).

Multiple genetic variations have been associated with development of IBD. An early genome- wide association study (GWAS) performed on IBD patients identified IL-23R variants associated with CD and UC in independent cohorts (10). IL-23R encodes the receptor for interleukin-23 (IL-23) and is important for activation of ILC3 and Th17 cells (11). Mutations in NOD2, the gene encoding nucleotide-binding oligomerization domain-containing protein 2 (NOD2), have also been associated with CD pathogenesis (12-14). Loss of NOD2 function results in increased inflammation due to impaired microbial clearance, increased production of proinflammatory cytokines, and inhibition of IL-10 production (12-15). GWAS studies have also revealed an association between UC and polymorphisms in GPR183, encoding the G- protein coupled receptor 183 (GPR183) (16). GPR183 is expressed on immune cells, including dendritic cells, B cells and T cells (17-19). GPR183 recognizes the cholesterol metabolite 7α,25-hydroxycholesterol (7a,25-OHC), and guides migration of the cells towards this oxysterol (20). 7a,25-OHC is synthesized from cholesterol by the enzymes cholesterol 25- hydroxylase (CH25H) and cytochrome P450, family 7, subfamily b, polypeptide 1 (CYP7B1) and degraded by the enzyme hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta- isomerase 7 (HSD3B7) into bile acid precursors which do not bind GPR183.

(24)

1.2.2 SEPSIS

Sepsis is a rapid, life-threatening condition characterized by organ dysfunction caused by an exaggerated and dysregulated host-response to infection (21). Despite advances in treatment regimens, antibiotic usage and vaccinations, sepsis is still a main cause of death from infection.

The true burden of sepsis worldwide remains unclear due to underdiagnosis and lack of data from low-income countries. Sepsis has therefore been highlighted by the World Health Organization (WHO) as a global health priority (22). The pathobiology of sepsis is still not fully defined. Infection triggers a complex response, which is highly variable between individuals. The character of the response depends on many different factors such as the type, load and virulence of the pathogen and origin of infection, but also on host-specific factors such as age, genetics, co-morbidities, medication, nutritional status, and immune dysfunction, as well as the time of validation (23). Sepsis was earlier believed to be an hyperinflammatory condition but is now recognized to involve both pro- and anti-inflammatory responses as well as alterations in metabolic, cardiovascular, hormonal, neuronal, and coagulation pathways (21, 23, 24).The hyperinflammatory state is driven by massive immune cell activation and cytokine release, which may help eradicate the pathogen but at the same time can damage the tissue and result in organ dysfunction and failure. On the other hand, the anti-inflammatory response and immunosuppression can result in increased susceptibility to secondary infections (25). There are still no approved drugs which specifically target sepsis and the treatment is limited to administration of antibiotics, intravenous fluids, oxygen, and support of organ function (23, 26). Instead of identifying single drugs targeting all forms of sepsis, the focus has now shifted towards developing tailored treatments targeting specific subgroups of patients (26). There are many immune modulatory drugs currently in clinical trials, but they have only been successful in some patients (26, 27). Treatment with some anti-inflammatory drugs have even increased mortality in a few cases (28, 29). Therefore, there is an urgent need to identify biomarkers of the immunological state of the individual to predict if the patient would benefit from a specific immune intervention. It is also important to keep in mind that sepsis is not solely an immunopathology but also involves alterations in many other non-immunological pathways in the body (24).

The mechanisms behind the immune response in sepsis is still not fully understood. The early response is triggered by pathogen-associated molecular patterns (PAMPs) derived from the invading microbe (30). There is also generation of extracellular or intracellular danger- associated molecule patterns (DAMPs) derived from host products, such as cellular debris, DNA, reactive oxygen species (ROS), histones, mtDNA, high-mobility group protein B1 (HMGB1), S100 proteins, and cytokines. PAMPs and DAMPs activate pattern recognition receptors (PRRs) leading to inflammasome assembly and release of active IL-18 and IL-1b.

This further amplifies the innate immune response which in turn leads to activation of the adaptive immune system (30-32). The formation of inflammasomes also mediates the onset of apoptosis, cell proliferation, induction of different metabolic responses, expression of genes encoding cytokines such as tumor necrosis factor (TNF), and release of additional DAMPs which results in a cycle of sustained immune activation and dysfunction (24). Neutrophils also

(25)

play an important role in the early phase of sepsis. Neutrophils are the most abundant immune cell type in the body and rapid responders to infections. Their effector function involves phagocytosis, release of ROS and protease-mediated killing of the pathogen (33), and they also form extracellular webs of DNA and antimicrobial proteins, known as neutrophil extracellular traps (NETs). NETs serve to kill invading pathogens, but excessive NET formation results in inflammation, induction of intravascular thrombosis, tissue damage, and multi-organ failure in sepsis (31, 34, 35). Sepsis is also associated with a massive systemic release of proinflammatory cytokines by both innate and adaptive immune cells. This systemic reaction was introduced as the “cytokine storm” in early preclinical studies (27, 36, 37). Elimination of proinflammatory cytokines such as IL-1b, TNF, IL-12, and IL-18 in animal models protected against organ failure and mortality. Although experimental sepsis models have limited relevance for human sepsis, exaggerated activity of proinflammatory cytokines is still considered an important contributor to human sepsis (27).

Dysregulation of multiple immune cell subsets contributes to the immunosuppressive changes observed in sepsis. The immunosuppressive state results in increased susceptibility to secondary infections and viral reactivation (38, 39). Apoptosis of CD4+ and CD8+ T cells, B cells, and NK cells have been reported in sepsis (40-44). T cells also become exhausted with impaired capacity to produce cytokines. CD8+ T cells in sepsis patients have decreased proliferating capacity, impaired cytotoxicity, and produce less IL-2 and interferon-g (IFNg) when stimulated (40, 45). The frequencies of regulatory T cells (Tregs) have been reported to increase in sepsis patients (46, 47). B cells also acquire an exhausted phenotype, downregulate their expression of major histocompatibility complex (MHC) class II and increase their production of the anti-inflammatory cytokine IL-10 (48).

The microbial pathogenesis is also an important factor contributing to the heterogeneity of the sepsis response. Sepsis can originate from a broad range of bacterial, viral, or fungal infections.

Infections of the respiratory tract are the most common cause, followed by infections of the abdomen and urinary tract (23, 49). Until the early 1980s, infections with Gram-negative bacteria were the most common cause of sepsis, however since then the cases caused by Gram- positive bacteria have steadily increased and are now the most common cause (25).

Staphylococcus aureus and Streptococcus pneumoniae are the leading Gram-positive bacteria causing sepsis. The most common Gram-negative bacteria are Escherichia coli, Klebsiella spp, and Pseudomonas aeruginosa. Microbial components are important initiators of the sepsis response. In 1970, Levine et al (50) reported that lipopolysaccharide (LPS) was present in the blood of sepsis patients. LPS is a major component of the outer membrane of Gram-negative bacteria and is a significant mediator of Gram-negative sepsis. LPS can trigger inflammatory responses through both extracellular and intracellular pathways (51). Extracellular signaling is mediated by binding to Toll-like receptor-4 (TLR4) and results in release of pro-inflammatory cytokines. LPS can also be internalized by some cell types resulting in activation of the nucleotide-binding domain, leucine-rich-containing family, pyrin domain-containing-3 (NLRP3)-inflammasome. Peptidoglycan is a major component of the cell wall of Gram-

(26)

positive bacteria and an important contributor to Gram-positive sepsis (52). Peptidoglycan binds to TLR2 resulting in the production of pro-inflammatory cytokines. Lipoproteins are a group of surface proteins present in both Gram-positive and Gram-negative bacteria (53). They exist both in a membrane-bound form with peptidoglycan and in a soluble form and are important for bacterial virulence, immune evasion, and colonization. Lipoproteins bind to TLR2 but also require additional interactions with other TLRs, and the downstream signaling results in production of pro-inflammatory cytokines (53). Microbial toxins contribute to the tissue damage and immune dysfunction seen in sepsis. Superantigens, which will be discussed in more detail below, are mainly produced by S. aureus and group A streptococci (GAS, Streptococcus pyogenes) and cause damage to the host cell without entering it, and severely disrupt the immune response (54). Other toxins, such as hemolysins and phospholipidases, can damage the cell membrane of host cells and disrupt the response to invading pathogens (25).

Bacterial A/B toxin has a binding moiety and an enzyme moiety and can inhibit host proteins and signaling pathways, and breaks down barriers to promote invasion (25). Many bacteria also produce antioxidant enzymes such as superoxide dismutase (SOD) and glutathione peroxidase which protect them from high concentrations of ROS (25, 55).

1.2.3 TOXIC SHOCK SYNDROME

Some species of Gram-positive bacteria, in particular S. aureus and GAS, can also cause toxin- mediated diseases such as toxic shock syndrome (TSS). TSS is an acute onset, rapidly progressing condition with high morbidity and mortality rates (54). It results from a superantigen-induced massive cytokine release and immune cell expansion and is characterized by high fever, hypotension and multi-organ failure early in the course of the infection (see also section 1.2.3.3). Current therapies focus on early identification and intervention, source control, administration of intravenous fluids and oxygen, support of organ dysfunction, and administration of antibiotics, including agents inhibiting toxin production (e.g. vancomycin) (54). Intravenous immunoglobulin (IVIg) can also be utilized to neutralize superantigens and improve outcome (56, 57). Interestingly, most individuals colonized with a superantigen- producing strain never develop TSS, suggesting that host-factors also play an important role.

People lacking antibodies against superantigens seem to be more at risk of developing TSS when colonized or infected with superantigen-producing S. aureus or GAS (58-60). Variations in human leucocyte antigen (HLA) class II haplotypes are also associated with differences in streptococcal TSS (STSS) development and severity (61, 62).

1.2.3.1 Group A streptococci

Streptococcus pyogenes, also known as group A streptococci (GAS), are Gram-positive beta- hemolytic cocci which can cause a wide range of diseases (63, 64). Most commonly GAS cause mild superficial infections of the skin or throat, such as impetigo and pharyngitis. However, in some cases the bacteria can breach the epithelial barrier and cause severe invasive infections with high mortality rates, such as necrotizing soft tissue infections (NSTI) or STSS. Repeated exposure to GAS can in some cases also give rise to autoimmune conditions such as acute

(27)

glomerulonephritis, acute rheumatic fever, or rheumatic heart disease (63). Together these diseases result in over half a million deaths per year world-wide (65). STSS was first reported in 1987 (66). In the following years, there was a marked increase in cases of severe invasive GAS infections, resulting from the emergence of new highly virulent strains (67). GAS secrete a wide range of toxins including proteases, hemolysins, and superantigens, which enhance their ability to colonize and survive in the tissue as well as evading the immune system (68). STSS often arises from invasive soft-tissue infections, such as necrotizing fasciitis, myositis or cellulitis and the organ dysfunction and hypotension develop rapidly. The initiating injury may in many cases seem trivial, such as blunt trauma or muscle strain, and evidence of a deep infection is not always obvious in the beginning of the disease progression. The infection develop with exceptional speed and rapid identification of the infection focus is crucial for survival. Presence of necrotizing fasciitis or myositis require acute surgical debridement and removal of both infected and surrounding tissue (54).

1.2.3.2 Staphylococcus aureus

S. aureus is an opportunistic pathogen which is part of the commensal microbiota of nasal mucosa in approximately one-third of the human population (69). Like GAS, S. aureus are Gram-positive cocci which can cause a wide variety of diseases at different sites throughout the body, including pneumonia, skin and soft tissue infections, bacteremia, sepsis, endocarditis, osteomyelitis, food-poisoning, and medical implant-associated infections (70). S. aureus poses a threat to the society due to its exceptional ability to develop resistance to antibiotics (71). The first cases of staphylococcal TSS were reported in 1978 and the following years an increasing number of cases were reported in young, otherwise healthy women using highly absorbent tampons (72). Vaginal cultures revealed that a high percentage of the women were colonized with toxin-producing strains of S. aureus (72, 73). Changes in tampon manufacturing and usage advice have reduced the incidence of menstrual TSS and it is now estimated to account for around 40% of staphylococcal TSS cases. The remaining 60% of staphylococcal TSS cases arise from S. aureus infections of other sites, such as skin infections or pneumonia (54, 73). S.

aureus produce a broad arsenal of virulence factors enabling colonization and immune evasion, such as superantigens and cytolytic toxins (74). Menstrual and non-menstrual TSS have similar clinical features with abrupt onset of influenza-like symptoms, such as fever, severe myalgia and gastrointestinal problems. This is rapidly followed by confusion and lethargy. In contrast to STSS, the infection focus is often superficial (54).

1.2.3.3 Superantigens

Superantigens produced by GAS or S. aureus have been implicated as key players in the pathogenesis of TSS (63, 75). Superantigens can, without prior cellular processing, bind certain TCR Vb segments and crosslink the TCR with MHC class II molecules and thereby bypass the conventional antigen processing and presentation. This results in activation of a large fraction of the T cell pool (54). The activation results in massive production of cytokines, such as IFNg, lymphotoxin-a, TNF, and IL-2 by T cells, and TNF, IL-1b, and IL-6 by antigen-presenting

(28)

cells (54). This leads to subsequent downstream activation of other immune cells further amplifying the cytokine release. This proinflammatory cascade is commonly referred to as a cytokine storm. The cytokine storm can cause vascular leakage, shock and multiorgan failure.

The cytokine response to superantigens involves activation of the transcription factor nuclear factor (NF)-kB (76) and the level of NF-kB activation has been linked to the risk of mortality (77). Since superantigens are specific to one or several Vb chains, skewing of the T cell response has been observed in superantigen-associated diseases, i.e., TSS and Kawasaki disease, with activation, expansion, or depletion of T cells carrying certain Vb chains (78, 79).

In menstrual TSS, almost all cases are primarily driven by the staphylococcal superantigen toxic shock syndrome toxin-1 (TSST-1) (59). TSST-1 is also responsible for around half the cases of non-menstrual TSS, while most of the remaining cases are dependent on staphylococcal enterotoxin B (SEB) (80). The streptococcal superantigen streptococcal pyrogenic exotoxin A (SpeA) have been associated with invasive GAS infections in many studies (81-83) but SpeC, SpeJ, and SpeM have also been found to be more prevalent in invasive GAS infections compared to non-invasive infections (81, 82, 84).

The release of superantigens leads to exhaustion of the responding T cells and provides an immune evasion strategy for the bacteria (85). Superantigens also facilitates bacterial colonization of the nasal mucosa and establishment of nasopharyngeal infection (86, 87).

1.2.4 LYMPHOCYTE METABOLISM IN INFLAMMATION

Inflammatory responses involve cell proliferation and differentiation, massive production of effector molecules, as well as tissue remodeling. This requires a major switch in immune metabolism and nutrient uptake to meet the increasing energy demand as well as the need of intermediates for biosynthesis of proteins, nucleic acids, and lipids. In their resting state, lymphocytes mainly rely on oxidative phosphorylation (OXPHOS) for their low energy demands (88, 89). However, upon activation they rapidly switch from catabolic to anabolic metabolism and upregulate aerobic glycolysis, where glucose is converted into lactate despite sufficient oxygen availability. In addition, there is an increase in OXPHOS activity as well as in uptake of glucose, amino acids, and other nutrients. T cells also increase glutamine metabolism, which is required for cell proliferation (90). The anabolic metabolism is dependent on specific transcription factors such as nuclear factor of activated T cells (NFAT) (91), Myc (92), and mammalian target of rapamycin complex 1 (mTORC1) (93, 94) signaling.

1.2.4.1 ROS signaling in T cells

The metabolic alterations during lymphocyte activation result in generation of ROS, such as hydrogen peroxide (H2O2) and superoxide (O2^•-) (88). ROS can be generated both in the intracellular and extracellular space through both enzymatic and non-enzymatic reactions. O2^•-

is mainly generated by Complex I, II, and III of the mitochondrial electron transport chain by transfer of electrons to molecular oxygen, and by membrane-bound NADPH oxidase (NOX) complexes (95-98). Complexes I and II emit O2^•- into the mitochondrial matrix where it

(29)

immediately is converted into H2O2 by superoxide dismutase 2 (SOD2). Complex III releases O2^•-both into the mitochondrial matrix and to the intermembrane space from where it can exit into the cytosol (98, 99). However, O2^•-is a very unstable molecule and can also spontaneously transform into H2O2. At high levels, ROS can damage DNA, proteins and lipids, but at low levels ROS can serve as signaling molecules important for immune cell functions (88, 98, 100).

ROS are also released by the immune system in defense against pathogens (101, 102). In T cells, ROS are involved in the production of many different inflammatory cytokines. ROS are required for signaling downstream of the TCR leading to NFAT, activator protein-1 (AP-1), and NF-kB activation and subsequent cytokine production (98, 103-105). Although debated, ROS are also suggested to be involved in the formation of the NLRP3 inflammasome (106- 108). The NLRP3 inflammasome is assembled in response to various danger signals and cellular stress. Upon sensing such signals, the NLRP3 protein recruits apoptosis-associated speck-like protein containing a CARD (ASC) and pro-caspase-1 which is cleaved into caspase- 1 (109). Caspase-1 then further processes pro-IL-18 and pro-IL-1b into active cytokines (109- 111). In a model of mice deficient in superoxide generation, the production of IL-12p70, IFNg and TNF was reduced, and so was the expression of IL-12R and T-bet on CD4+ T cells (104).

Both endogenous ROS in the T cells and exogenous ROS produced by dendritic cells contributed to the ROS-dependent cytokine production in CD4+ T cells (104). ROS are also required for signal transducer and activator of transcription 4 (STAT4) and NF-kB signaling downstream of IL-12R and IL-18R resulting in IFNg production (112). However, too high levels of ROS inhibit T cell activation by blocking mTOR, Myc and NFAT signaling (100).

1.2.4.2 Antioxidant systems

In order to maintain the redox balance and prevent damage caused by ROS, the cell has several enzymatic and non-enzymatic antioxidant systems. The glutathione system is one of the most important lines of defense against oxidative stress in mammalian cells. Glutathione (GSH) is an endogenous antioxidant which is synthesized in two steps, first by ligation of cysteine and glutamate to g-glutamylcysteine by glutamate cysteine ligase (GCL) and then by addition of glycine catalyzed by glutathione synthetase (113). The reaction catalyzed by GCL is the rate- limiting step in GSH synthesis due to the usually low availability of cysteine (114). GSH plays a major role in maintaining T cell effector functions, promote T cell growth, and metabolic reprogramming by buffering ROS (100). Two important structural features of the GSH molecule are its γ-glutamyl bond, which makes it very stable and resistant to intracellular degradation, and its sulfhydryl group which gives it reductive power (113). GSH acts as an antioxidant through reactions catalyzed by glutathione peroxidase, which reduce H2O2 while GSH is oxidized into glutathione disulfide (GSSG). GSSG is reduced back to GSH by glutathione reductase, a reaction which requires nicotinamide adenine dinucleotide phosphate (NADPH) (114). In addition to its antioxidative capacity, GSH can also regulate signal transduction, metabolism, inflammation, and apoptosis through post-translational modifications called S-glutathionylation (114). For example, S-glutathionylation of IkBa or IkB kinase b inhibits NF-kB activation and migration to the nucleus (115, 116). NF-kB is

(30)

important for the transcription of many inflammatory factors including IFNg, TNF, IL-12, pro- IL-18 and NLPR3 (117, 118). GSH further control NLRP3 inflammasome formation by S- glutathionylation of caspase-1 which prevents the inflammasome assembly (119). The activity of other caspases can also be regulated by S-glutathionylation. For example, S- glutathionylation of caspase-3 inhibits its proteolytic activity and prevents apoptosis (120). In conventional T cells, the ROS-buffering capacity of GSH is an important regulator of metabolic reprogramming in inflammatory responses (100).

1.3 INNATE LYMPHOCYTE RESPONSES 1.3.1 INNATE LYMPHOID CELLS

ILCs are a recently discovered group of innate lymphocytes which are important in the initiation, regulation and resolution of inflammatory processes. They respond rapidly to cytokines and microbial stimuli and can produce large amounts of cytokines (121). Even though they are rare compared to adaptive lymphocytes they are enriched at barrier surfaces in the body, such as skin, airways and in the intestine.

Figure 2. Classification of innate lymphoid cells and illustration of their similarities to T helper cell subsets.

1.3.1.1 ILC phenotypes

ILCs share many features of T helper (Th) cells, such as transcriptional programs and cytokine responses (Figure 2). However, in contrast to Th cells they lack rearranged antigen- specific receptors. ILCs can be divided into five major subsets based on their developmental trajectories: NK cells, ILC1, ILC2, ILC3 and lymphoid tissue inducer (LTi) cells (3). They are further grouped based on their cytokine and transcription factor profiles; group 1 ILCs (ILC1 and NK cells) which produce the Th1 cytokine IFNγ and require the transcription factor T-bet (122), group 2 ILCs (ILC2) which produce the Th2 cytokines IL-5 and IL-13

(31)

and are dependent on the transcription factor GATA3 (123-125), and group 3 ILCs (ILC3 and LTi cells) which produce the Th17 cytokines IL-17 and IL-22 and express the transcription factor retinoid-related orphan receptor-γt (RORγt) (126-128). Furthermore, there is a high plasticity between ILC subsets, and they have the ability to adapt to the surrounding by altering their phenotype and function in response to cytokines.

Group 3 ILCs is a heterogenous group which, in addition to RORγt are dependent on IL-7Ra for their development. All subsets within the group respond to the cytokines IL-23 and IL-1b (11, 127, 129, 130). Apart from NK cells, LTi cells was the first ILC subset to be discovered in 1997 (131). LTi cells and ILC3 share a similar transcription factor dependency and cytokine profile, but it is now clear that LTi cells develop from a progenitor distinct from the progenitors of other ILCs and of NK cells. LTi cells play an important role in lymphoid tissue development and influence the development, activation, and function of B cells and T cells (132-135). LTi cells in mouse and human share many similarities, however mouse LTi cells express CD4 while human LTi cells do not (122). Non-LTi ILC3s can be subdivided based on the expression of natural cytotoxicity receptors (NCR) NKp46 in mice and NKp44 in human. NCR+ ILC3s produce IL-22 but not IL-17A, while NCR- ILC3s secrete both IL-22 and IL-17, as well as IFNg (127).

Figure 3. Development of SILTs in the intestine.

1.3.1.2 Group 3 ILC functions in the gut and in gut inflammation

Different subsets of group 3 ILCs localize to different areas of the intestine. C-C motif chemokine receptor 6 (CCR6)- ILC3s are found scattered throughout the lamina propria (128, 136, 137), whereas CCR6+NKp46- LTi cells seed the gut during fetal development and are localized within lymphoid structures known as cryptopatches and innate lymphoid follicles (ILFs) (131, 138-140). Cryptopatches and ILFs are collectively referred to as solitary isolated lymphoid tissues (SILTs) and differ from larger lymphoid structures such as Peyer’s patches and mesenteric lymph nodes (MLN), both developmentally and functionally (141).

Cryptopatches are composed of LTi cells and stromal cells surrounded by dendritic cells (Figure 3). Upon exposure to the microbiota, B cells are recruited and the cryptopatch develops

(32)

into an ILF. SILTs develop postnatally through the activity of adult LTi cells producing lymphotoxin (LT) (138, 142, 143).

ILC3s are important for maintaining the balance between the commensal bacteria and the immune system and also play a role in maintaining the intestinal barrier function (121, 144).

The protective role of ILC3s in the intestine is mainly mediated by moderate secretion of cytokines such as IL-22, IL-17, and granulocyte-macrophage colony-stimulating factor (GM- CSF) which trigger production of antimicrobial peptides by epithelial cells, regulate T cell responses by expression of MHC class II and support the tolerance function of dendritic cells.

In mice, IL-22 produced by ILC3s has also been found to stimulate stem cells to prevent tissue damage (145). Consequently, dysregulation of these interactions has been implicated in IBD.

The role of ILCs in intestinal pathology was originally described in an innate colitis model in mice where ILCs produced high amounts of IL-17 in response to IL-23 (146). Next, IL-23- responsive ILCs were also found to be enriched in the intestine of patients with IBD (11).

Although moderate production of IL-17 by ILC3 in response to IL-23 has been shown to have a protective effect in the intestine, dysregulation of ILC3s can result in overproduction of IL- 17 which leads to recruitment of proinflammatory neutrophils. The neutrophils impair the epithelial barrier and increases intestinal permeability by disrupting E-cadherin and junctional adhesion molecule-like molecule (JAML), thus exacerbating inflammation in IBD (147). IL- 22 also serves as a double egged sword in IBD. Although protective at moderate levels, too high production of IL-22 and GM-CSF by NKp46+ ILC3 could lead to inflammation in the anti-CD40 colitis model in mice (148). In addition, elevated expression of IL-22 by human ILC3 has been reported in patients with CD and UC (149).

The plasticity of ILC3s have also been implicated in IBD. CD patients have a higher proportion of CD14+ dendritic cells producing IL-12 (150). IL-12 stimulation of ILC3 results in downregulation of RORγt expression and upregulation of T-bet expression, thus acquiring a more ILC1-like phenotype with increased production of IFNγ. This transition from ILC3 to ILC1 was reversible and driven by IL-23 together with IL-1b and retinoic acid (150). In line with this, increased proportions of ILC1, at the expense of ILC3, have been reported in IBD (151).

1.4 MAIT CELLS

Mucosal-associated invariant T (MAIT) cells are the most abundant subset of unconventional T cells in human. They were first discovered in 2003 as MHC class I-related protein 1 (MR1)- restricted T cells expressing a TCR with an invariant Va7.2-Ja33 arrangement (152). These cells were abundant in human gut mucosa and were thus named mucosal-associated invariant T cells. (152). MAIT cells are now known to be abundant at many different mucosal and non- mucosal barrier sites, such as the liver (153), skin (154), adipose tissue (155), lungs (156), gastrointestinal tract (153, 157), female genital mucosa (158), and oral mucosa (159), as well as in circulation (153, 160). The frequency of circulating MAIT cells in adult humans varies substantially among individuals ranging from 0.1% to 20% and decline in older age (160).

(33)

1.4.1 MAIT CELL DEVELOPMENT AND PHENOTYPE

MAIT cells develop in the thymus through selection by MR1-expressing CD8⁺CD4⁺ thymocytes (161, 162). In the first development stage, CD161^-CD27^- MAIT cells are located in the thymus, display a naïve phenotype and lack expression of promyelocytic leukemia zinc finger (PLZF). At stage two, MAIT cells acquire a CD161^-CD27⁺ phenotype and start to migrate out of the thymus under the control of PLZF. At stage three, MAIT cells mature in the periphery and become CD161⁺CD27^pos-lo (163). Seeding of MAIT cells in peripheral tissues occurs before birth, however the MAIT cell frequency in the periphery is lower at birth compared to adults and increases during the first years of life (161, 163-165). Mouse studies have also suggested that exposure of riboflavin-synthesizing microbes early in life is required for MAIT cell development (166).

Adult MAIT cells express different transcription factors that shapes their functionality. PLZF is expressed in all MAIT cells and is essential for their development (163) and is also associated with elevated expression of the cytokine receptors IL-12R and IL-18R (167). Expression of T- bet and Eomes is associated with a Th1 phenotype and production of IFNg and TNF, as well as cytotoxic molecules granzyme B, perforin, and granulysin (168, 169). Expression of RORgt and STAT3 is associated with a Th17 phenotype and production of IL-17 and IL-22 (158, 170).

1.4.2 MAIT CELL ACTIVATION

There are two main pathways of MAIT cell activation: TCR-dependent and TCR-independent.

In many situations there is synergy between these two pathways (171). In addition, MAIT cells can be activated in a largely unspecific manner by superantigens (85). The MAIT cell activation pathways and effector functions are illustrated in Figure 4.

1.4.2.1 Antigen presentation and TCR-mediated activation

In humans, MAIT cells express a semi-invariant TCR with a highly conserved Va7.2-Ja33 arrangement coupled to a limited repertoire of b chains. The MAIT cell TCR therefore recognizes a limited set of microbial antigens presented by the antigen-presenting molecule MR1. The most potent MAIT cell-activating antigens presented by MR1 are the vitamin B2 (riboflavin) metabolites 5-(2-oxopropylideneamino)-6-D-ribitylaminouracil (5-OP-RU) and 5- (2-oxoethylideneamino)-6-D-ribitylaminouracil (5-OE-RU) which are generated during riboflavin biosynthesis in a wide range of microbes (172, 173). As this pathway is absent in mammalian cells, the detection of these antigens allows for effective discrimination between host and pathogen. In addition, vitamin B9 (folic acid) metabolites also act as MR1-ligands (173, 174). However, they do not have the ability to stimulate MAIT cell activation. Le Bourhis et al (175) reported that MAIT cells respond to a wide range of riboflavin-synthesizing microbes, but not to GAS or Enterococcus faecalis that lack riboflavin metabolism. However, Meermeier et al (176) found that a certain MR1-restriced subset of MAIT-like cells expressing TRAV12.2 could get activated by GAS in an MR1-dependent manner. Lepore et al (177) also identified a new subset of MR1-restricted cells which were able to respond to endogenous

(34)

Figure 4. Summary of MAIT cell activation pathways and effector functions. The TCR-dependent activation occurs in response to 5-OP-RU from riboflavin-synthesizing bacteria which is presented by MR1 on the antigen-presenting cell (APC). TCR-independent activation occurs in response to innate cytokines which are released when viruses or bacteria bind to PRRs on the APC. Superantigens activate MAIT cells by crosslinking the MAIT cell TCR with MHC class II on the APC resulting in both activation through the TCR and release of IL-12 and IL-18.

MR1-presented ligands. In addition, MR1 can bind and present synthetic drugs, which can induce activation of MAIT cells (178). These findings therefore suggest that there might be significant ligands for MR1 other than the riboflavin metabolite antigens. Differences in TCR Vb usage enable the MAIT cells to fine tune the response to different microbes (179). For example, MAIT cells responding to Escherichia coli display a different Vb profile than MAIT cells responding to Candida albicans.

The MAIT cell response to riboflavin metabolites results in production of cytokines like TNF, IFNg, and IL-17, and cytotoxic molecules, such as granzyme B, perforin and granulysin (168, 179-181). The pattern of cytokine production by MAIT cells differs depending on tissue site.

The antimicrobial activity of MAIT cells also depends on location. Circulating MAIT cells produce mainly IFNg, TNF, and IL-2 and low levels of IL-17A (85, 179, 182). Tissue-resident MAIT cells are more prone to produce IL-17A and IL-22 (158, 183). MAIT cells in adipose

(35)

tissue have been reported to produce IL-10 and IL-17 (155). Tissue-resident MAIT cells in oral mucosa are biased towards production of IL-17 upon activation (159). However, triggering of the MAIT cell TCR is not sufficient for full activation of MAIT cell functions (169, 171).

MAIT cells also require co-stimulation by CD28, TLRs or cytokines (184, 185). The need for co-stimulation is probably important to prevent unwanted MAIT cell activation in the mucosa where they are in close proximity to commensal riboflavin-synthesizing bacteria (186).

1.4.2.2 Response to innate cytokines

Besides the TCR-mediated activation, MAIT cells can be activated independently of MR1 by the cytokines IL-12 and IL-18 in an innate-like manner (187). MAIT cells express high levels of the cytokine receptors IL-18R, IL-12R, IL-15R, and IFNaR (171, 175, 188). Ussher et al (187) compared activation of MAIT cells cultured with THP-1 cells fed with either fixed E.

coli, which produce riboflavin metabolites, or fixed E. faecalis, which lack riboflavin metabolism. In their system, E. faecalis could induce a robust IFNg-dominated response in MAIT cells in a TCR-independent, IL-12 and IL-18-mediated manner. The MAIT cell response to fixed E. coli was TCR-mediated at early timepoints but shifted to a TCR- independent response at later timepoints in the co-culture. This could be due to a limited availability of MR1 ligand in fixed bacteria (189). The TCR-independent, cytokine-driven activation of MAIT cells allows for responses to viral infections and to bacteria that lack riboflavin metabolism. However, cytokine-mediated TCR-independent MAIT cell stimulation activates a limited set of MAIT cell effector functions dominated by IFNg, granzyme B and low levels of TNF (171).

1.4.2.3 Response to superantigens

In addition to the MR1-restricted and cytokine-driven activation mechanisms of MAIT cells, Shaler et al (85) found that MAIT cells can be activated by the staphylococcal superantigen SEB resulting in a massive release of the proinflammatory cytokines IFNg, TNF, and IL-2, but not IL-17A. The activation was partially Vb13.2-dependent and largely IL-12 and IL-18 dependent. They also showed that the activation was dependent on the interaction of SEB with MHC class II and not MR1. The hyperactivation of MAIT cells was followed by a state of anergy with induced expression of the exhaustion markers lymphocyte-activation gene 3 (LAG-3), T cell immunoglobulin and mucin-3 (TIM-3), and/or programmed cell death-1 (PD- 1). In addition, the MAIT cells became unresponsive to stimulation with Klebsiella pneumoniae and E. coli. Hyperactivation and exhaustion of MAIT cells may therefore be an immune evasion mechanism utilized by S. aureus, and MAIT cells may have a pathogenic, rather than a protective role, in staphylococcal TSS.

1.4.3 MAIT CELL METABOLISM

MAIT cells at resting state are metabolically quiescent. Upon activation they rapidly engage the glycolytic pathway and increase the uptake of glucose in a stimulus-dependent manner (190). The increase in glucose uptake is lowest when the cells are stimulated with IL-12 and

Innate lymphocyte responses in inflammatory diseases

From the Center for Infectious Medicine, Department of Medicine, Huddinge Karolinska Institutet, Stockholm, Sweden

INNATE LYMPHOCYTE RESPONSES IN INFLAMMATORY DISEASES

Innate lymphocyte responses in inflammatory diseases THESIS FOR DOCTORAL DEGREE (Ph.D.)

Johanna Emgård

POPULAR SCIENCE SUMMARY OF THE THESIS

POPULÄRVETENSKAPLIG SAMMANFATTNING

ABSTRACT

LIST OF SCIENTIFIC PAPERS

SCIENTIFIC PAPERS NOT INCLUDED IN THE THESIS

CONTENTS

LIST OF ABBREVIATIONS

1 INTRODUCTION