SLPI and Soluble BTLA As Immunological Markers in Severe Bacterial Infections

SLPI and soluble BTLA as immunological markers in severe bacterial infections

To my family

Örebro Studies in Medicine 211

ANNA LANGE

SLPI and soluble BTLA as immunological markers in severe bacterial infections

Title: SLPI and soluble BTLA as immunological markers in severe bacterial infections Publisher: Örebro University 2020 www.oru.se/publikationer

Print: Örebro University, Repro 04/2020

ISSN 1652-4063 ISBN 978-91-7529-335-6 Abstract

Anna Lange (2020): SLPI and soluble BTLA as immunological markers in severe bacterial infections. Örebro Studies in Medicine 211.

Clinical presentation, and outcome of infections are affected by host-, and etiology- (focus of infection and pathogen) related factors. The immune response is controlled by a network of regulating pathways. This thesis focuses on Secretory Leukocyte Protease Inhibitor (SLPI), a protease inhibitor with anti-inflammatory properties, and the previously non-studied soluble isoform of B and T lymphocyte attenuator (sBTLA), a membrane-associated regulatory protein. Plasma concentrations of SLPI and sBTLA were assessed in relation to etiology, severity, mortality, and markers of inflammation and immunosuppression, in i) community- acquired pneumonia (CAP) (SLPI), ii) intensive care unit (ICU) treated severe sepsis and septic shock (sBTLA), and iii) dynamically in BSI (SLPI and sBTLA). Main findings were: higher expression of SLPI in pneumonia, compared to other sources, higher initial concentrations in Streptococcus pneumoniae, and Staphylococcus aureus BSI, compared to Escherichia coli BSI, and higher SLPI concentrations in sepsis compared to non-septic BSI. Interestingly, men with pneumonia had higher plasma levels of SLPI, both in CAP and BSI. Likewise, sBTLA was associated with severity, but preferentially at higher organ failure scores. High sBTLA was associated with increased risk of early death (28 days) in ICU-treated septic patients, and with mortality at 90 days and one year in BSI. In particular, failure to normalize sBTLA on day 7, was indicative of worse long-term outcome. SLPI was associated with decreased monocytic HLA-DR expression, and sBTLA with decreased lymphocyte count, which might indicate a connection to sepsis-associated immunosuppression. In conclusion, SLPI and sBTLA show association with severity, and markers of immune dysfunction, in sepsis and BSI. SLPI differs depending on etiology, while sBTLA may have prognostic implications. Our results propose that the pathobiological role of sBTLA, and the possible utility of SLPI and sBTLA in sepsis immune-profiling, should be further addressed in future studies. Keywords: SLPI, sBTLA, sepsis, bloodstream infection, pneumonia Anna Lange, School of health and Medical Sciences, Faculty of Medicine and Health, Örebro University, SE-70182 Örebro, Sweden, [email protected]

Table of Contents

LIST OF PAPERS ...... 11 ABBREVIATIONS ...... 12 INTRODUCTION ...... 13 Sepsis from a historical perspective ...... 13 Sepsis pathophysiology and clinical features ...... 14 Sepsis definitions ...... 14 Sepsis epidemiology ...... 16 Sepsis therapy ...... 16 Protective immune responses: innate immunity ...... 17 Sensing danger ...... 17 Macrophages ...... 19 Neutrophils ...... 20 NK cells ...... 22 Soluble factors...... 22 Cytokines ...... 22 Complement ...... 23 Antimicrobial peptides and proteins...... 23 Dendritic cells ...... 24 Protective immune responses: adaptive immunity ...... 25 T cells ...... 25 B cells ...... 26 Co-signaling proteins in regulation of immune responses...... 27 Dysregulated immune response in sepsis ...... 28 Therapies targeting immune dysregulation in sepsis ...... 30 Cortisone ...... 30 Intravenous immunoglobulin (IVIg) ...... 31 Blood purification ...... 31 Granulocyte-macrophage colony stimulating factor ...... 31 IFN-γ ...... 32 IL-7 ...... 32 Checkpoint inhibition ...... 32 Secretory leukocyte protease inhibitor (SLPI) ...... 33 BTLA ...... 35 AIMS ...... 37 SUBJECTS AND METHODS ...... 38 Patients ...... 38 Paper I ...... 38 Paper II ...... 38

ANNA LANGE SLPI and soluble BTLA as immunological markers 9

Papers III and IV ...... 39 Storage of plasma ...... 39 Enzyme-linked Immunosorbent assay (ELISA) ...... 40 SLPI and sBTLA detection ...... 41 Blood and other cultures ...... 42 Clinical data ...... 42 Comorbidity and severity scores ...... 42 Statistics ...... 43 Group comparisons ...... 43 Associations and correlations ...... 44 Confounding ...... 44 Missing data ...... 44 Longitudinal data analysis ...... 44 Mortality analyses ...... 45 RESULTS ...... 46 SLPI ...... 46 CAP and controls, and in relation to source of infection ...... 46 SLPI and bacterial etiology in BSI ...... 47 Severity ...... 48 Sex-related differences in SLPI expression ...... 49 Association to other biomarkers ...... 50 sBTLA ...... 51 Sepsis, ICU controls and healthy controls ...... 51 sBTLA and severity ...... 52 sBTLA and mortality ...... 52 Association to other biomarkers ...... 53 ETHICAL CONSIDERATIONS ...... 54 DISCUSSION ...... 55 CONCLUSIONS ...... 59 FUTURE PERSPECTIVES ...... 60 SAMMANFATTNING PÅ SVENSKA ...... 62 ACKNOWLEDGEMENTS ...... 64 REFERENCES ...... 65

10 ANNA LANGE SLPI and soluble BTLA as immunological markers

List of papers This thesis is based on the following papers and manuscripts, which are referred to in the text by their Roman numerals.

I. Lange Jendeberg A, Strålin K, Hultgren O (2013). ”Antimicrobial peptide plasma concentrations in patients with community-acquired pneumonia”. Scand J Infect Dis, 45 (6), 432-7.

II. Lange A, Sunden-Cullberg J, Magnuson A, Hultgren O (2017). “Solu- ble B and T Lymphocyte Attenuator Correlates to Disease Severity in Sep- sis and High Levels Are Associated with an Increased Risk of Mortality”. PLoS One 12 (1):e0169176

III. Lange A, Cajander S, Magnuson A, Sundén-Cullberg J, Strålin K, Hultgren O (2019). “Plasma Concentrations of Secretory Leukocyte Prote- ase Inhibitor (SLPI) Differ Depending on Etiology and Severity in Commu- nity-Onset Bloodstream Infection”. European Journal of Clinical Microbi- ology & Infectious Diseases 38:1425–1434.

IV. Lange A, Cajander S, Magnuson A, Strålin K, Hultgren O. “Sustained elevation of soluble B- and T- lymphocyte attenuator predicts long-term mortality in patients with bacteremia and sepsis”. Submitted.

Papers I-III are reprinted with permission from the publishers.

ANNA LANGE SLPI and soluble BTLA as immunological markers 11

Abbreviations AMPs Antimicrobial peptides BSI Bloodstream infection BTLA B and T Lymphocyte Attenuator CAP Community-acquired pneumonia CD Cluster of differentiation CI Confidence interval CRP C-reactive protein CTLA-4 Cytotoxic T Lymphocyte Associated Antigen 4 DAMP Damage-associated molecular pattern DC Dendritic cell DIC Disseminated intravascular coagulation ELISA Enzyme-linked immunosorbent assay HC Healthy controls HVEM Herpes Virus Entry Mediator HLA Human leukocyte antigen HR Hazard Ratio ICU Intensive care unit IL Interleukin NE Neutrophil elastase NK Natural Killer (Cell) NFƙB Nuclear factor kappa B PAMP Pathogen-associated molecular pattern PD-1 Programmed Death 1 PD-L1 Programmed death ligand 1 PD-L2 Programmed death ligand 2 PRR Pattern recognition receptor SLPI Secretory Leukocyte Protease Inhibitor SOFA Sequential Organ-failure Assessment Score TH T helper (cell) TLR Toll-like receptor TNF Tumor necrosis factor UTI Urinary tract infection

12 ANNA LANGE SLPI and soluble BTLA as immunological markers

Introduction

Sepsis from a historical perspective The lethal consequences of the symptoms related to severe infections were well known already in ancient Greece, where the word sepsis (sepo, “I rot”), appear in poems by Homer. This comprehension grew into the mi- asma (“bad air”) theory, which remained dominant for centuries, and according to which disease emerged from rotting organic matter, and spread through the air [1]. The germ theory started to gain ground in the mid-19th century [2]. During this period, Semmelweiss and Snow made seminal observations on puerperal fever and cholera, but despite the evidence, the sci- entific society did not welcome their theories. It was Pasteur’s work on the growth of microorganisms, and further research by Koch and Lister, that led to the foundation of microbiology as we know it today [1]. Penicillin was discovered in 1928, and was put into clinical use during world war II [3]. As it turned out, the germ theory could however not fully explain the pathobiology of sepsis, since some people with sepsis died from organ failure and shock, despite adequate antibiotics. Significant discoveries regarding the pathological mechanisms in sepsis were made already in the 1890’s, with the findings on the pyrogenic and shock-inducing properties of endotoxin, the uncovering of the bacteriolytic activity of complement on bacteria, and the first descriptions of experimental disseminated intravascular coagulation (DIC) in acute infection [4-6]. Major steps for understanding the interplay between innate and adaptive immunity were: the dendritic cell (1973), the portrayal of the T cell receptor structure and function (1987), and the revelation of the co-stimulatory signal required for T cell activation [7-9]. The importance of the Toll protein in fruit fly immunity (1996) was the key to understanding how pathogens activate the immune system, and was followed by the identification of TLR-4, its ligand (endotoxin), and the downstream effects of their interaction [10-12] . The study of cytokines (from Greek “cell movement”) took off in the 1980’s. Observations of high circulating levels of pro-inflammatory cytokines in sepsis, motivated studies on cortisone and anti-cytokine strategies (tumor necrosis factor, interleukin), as adjunctive sepsis treatment. Despite evidence from experimental models, there was lack of proof for efficacy in humans with respect to decreasing short-term mortality [13].

ANNA LANGE SLPI and soluble BTLA as immunological markers 13

Expanding knowledge on the pathology of sepsis immune dysregulation has led to a paradigm shift in sepsis research, with focus on persisting inflammation, and sustained immune suppression, which may have long-term consequences [14-16].

Sepsis pathophysiology and clinical features Common sources of infection in sepsis are the respiratory and urinary tracts, and abdominal foci [17, 18]. Cultures can identify the causative pathogen in about half of sepsis cases [17-19]. Symptoms and signs in the individual host relates to previous health status, the causative pathogen, the source of infection, and organ dysfunction [20]. Common early signs are fever or hy- pothermia, tachycardia, tachnypnea, and altered mental status. Immune activation, and protective mechanisms of immune cells, will be described below, as well as the processes related to immune dysregulation. In short, there is a spread of inflammation throughout the body, with generalized endothelial activation, vascular leakage and vasodilation. In addition, there is coagulopathy, which compromises microvascular blood flow, and in the worst scenario results in DIC and bleeding manifestations. To- gether, these reactions lead to organ dysfunction, hypoperfusion, and some- times chock [15, 20, 21]. While these manifestations are associated with early peaks of mortality, there is also activation of immunosuppressive pathways, which can lead to profound weakening of the immune system, with risk for secondary, often opportunistic infections [15].

Sepsis definitions The first consensus definitions for severe sepsis and septic shock were proposed in 1992 (Sepsis-1), and were based on suspected infection, in combination with two or more systemic signs of inflammation, the systemic inflammatory response syndrome (SIRS) criteria. Severe sepsis was sepsis with either organ dysfunction, hypoperfusion or hypotension, and septic shock was defined as sepsis with hypotension despite “adequate” fluid resuscitation [22]. When diagnostic criteria were revised in 2001 (sepsis-2), the list of signs and symptoms of systemic inflammation and organ dysfunction was expanded [23]. Definitions were challenged by increasing insights into sepsis pathobiology, and evidence of limited specificity and sensitivity of SIRS criteria [24, 25]. In 2016, new definitions were proposed (Sepsis-3), which abandoned

14 ANNA LANGE SLPI and soluble BTLA as immunological markers

SIRS for organ failure-based criteria, aimed to identify patients with a greater risk to die [21]. The threshold to define sepsis was increased: “a life- threatening organ dysfunction due to a dysregulated host response to an infection, which is identified by an increase of the Sequential Organ Failure Assessment (SOFA) score by 2 or more points from the baseline”. The term severe sepsis, was declared redundant, and septic shock was defined as “a vasopressor requirement to maintain a mean arterial pressure of 65 mg Hg or more, with metabolic deterioration specified as serum lactate concentration greater than 2 mmol/L, in the absence of hypovolemia”, figure 1. The Sepsis-3 definition of septic shock has been shown to capture more severely ill patients than Sepsis-2 [26]. Due to difficulties to calculate SOFA score at the bedside, a screening tool, “quick-SOFA” (qSOFA) was proposed, where suspected infection plus two of three signs: i) altered mental status, ii) respiratory rate ≥22, and iii) systolic blood pressure <100 mmHg, should raise the suspicion of sepsis. The prognostic capacity of qSOFA, as well as its sensitivity, has been questioned in subsequent studies [27, 28].

Figure 1. Sepsis-3 definitions. Reprinted with permission from [29]

ANNA LANGE SLPI and soluble BTLA as immunological markers 15

Sepsis epidemiology It is difficult to achieve a true estimate of the global sepsis incidence, due to lacking data from middle and low-income countries [30]. ICU bed availa- bility differs, and assessment of sepsis incidence outside the ICU-setting is problematic, on account of non-consistent diagnose-coding [31]. Sepsis management has improved, and mortality has decreased over the past two decades [32-34]. Mortality is still significant, however, and results from recent RCTs indicate a 90-day mortality in septic shock in adults between 28 and 46% [35-37]. Risk factors for sepsis include older age, male sex, and pre-existing chronic diseases. Genetic factors may contribute, especially at younger age [31]. The long-term consequences of sepsis, in terms of morbidity and mortality, has received growing attention [14, 38]. In May 2017, the World Health Organization adopted a resolution to reduce the global health burden of sepsis through improved prevention, diagnosis and management. Among several measures that were proposed in the resolution are: campaigns to increase the awareness of sepsis in the public, and among health care work- ers, to ensure timely diagnosis and treatment, and to include sepsis as a priority research area [39].

Sepsis therapy The significant improvement in early sepsis survival over the past two decades, owes much to the impact of seminal studies that pinpointed the importance of early detection, circulatory resuscitation, and prompt administration of antibiotics in septic shock [40, 41]. These interventions remain the mainstay treatment of sepsis, despite questioning of protocolized resuscitation, and the necessity of immediate antimicrobials, in recent studies [42]. The Surviving sepsis campaign (SSC) was initiated in 2002, to increase awareness and to improve outcome in severe sepsis and septic shock, through implementation of evidence-based resuscitation goals [43]. The current SSC guidelines advocate for the following interventions, within one hour of sepsis diagnosis: i) measurement of serum lactate, ii) blood cultures obtained before antibiotic are given, iii) administration of broad-spectrum antibiotics, iv) intravenous crystalloid fluids in the presence of hypotension or lactate ≥4, and v) application of vasopressors to maintain a mean arterial pressure ≥65, if inadequate response to fluid resuscitation. The SSC guide-

16 ANNA LANGE SLPI and soluble BTLA as immunological markers lines also include directions on source control, repeated lactate measurement, on the use of low dose corticosteroids, and on blood glucose management, along with recommendations on mechanical ventilation, renal re- placement therapy, and nutrition [44].

Protective immune responses: innate immunity The skin and mucosal membranes are at the frontline of immunity, dealing with constant exposure to pathogens in our environment. These surfaces are lined with protective components, such as antimicrobial peptides, acids, and gut microbiota, which protect from colonization of harmful bacteria.

Sensing danger The immune system is alerted by microbial molecules, so called pathogen- associated molecular patterns (PAMPs), which bind to pattern recognition receptors (PRRs) at the surface, or in intracellular compartments of innate immune cells, epithelial cells, and many other cell types [45, 46], figure 2. PAMPs are compounds at the surface of, or inside pathogens, that are shared among species, and which signal danger to the immune system [47]. There is redundancy in the pathogen-sensing mechanisms, where several PRRs can detect structures from the same microorganism. The prototypic PAMP is endotoxin (LPS), which binds to TLR-4. Other TLRs that are important for bacterial sensing are TLR-2, which recognizes lipoproteins, lipo- teicoic acid and peptidoglycan, and TLR-5, which binds to flagellin [46]. Importantly, not only foreign structures elicit innate immune responses. Intracellular molecules are released by damaged och stressed cells, in tissue injury, for example in wounds, or necrotic cell death secondary to sepsis. These so called damage-associated molecular pattern molecules (DAMPs), e.g. uric acid, adenosine triphosphate, and High Mobility Group Box 1, are sensed through mechanisms similar to PAMP-recognition [48].

ANNA LANGE SLPI and soluble BTLA as immunological markers 17

Figure 2. Cellular location of pattern recognition receptors. Toll-like receptors (TLR) are located within cytoplasmic and endosomal membranes. C-type lectins are located within the cytoplasmic membrane and detects pathogen-derived carbohy- drates. Nod-like receptors (NLR), retinoic-acid inducible gene (RIG)-like receptors (RLR), and Cytosolic DNA sensors (CDS) are present in the cytosol. Reprinted with permission from Abbas et al., Cellular and Molecular immunology (2017).

PRR-ligation activates intracellular signalling pathways that promote production of inflammatory cytokines, chemotactic factors, antimicrobial peptides, and interferons, figure 3. First, adaptor molecules, i.e. MyD88, MAL, TRIF and TRAM, associate with the cytosolic part of the PRR receptor. MyD88 is involved in most TLR-induced pathways (not TLR-3), and is crucial in the early response to bacterial infections [46]. Myd88 and IRAK4 (a protein that is recruited directly downstream of Myd88) deficiency, are rare genetic disorders that predispose to recurrent life-threatening bacterial infections at young age [49]. Further downstream events terminate in degradation of the inhibitor of ĸB (IĸB), which releases nuclear factor kappa B (NFĸB), so it can translocate into the nucleus and induce transcription of pro-inflammatory proteins, e.g. TNF- , IL-1, and IL-6 [46].

18 ANNA LANGE SLPI and soluble BTLA as immunological markers

Figure 3. TLR4-signalling. LPS ligation to TLR4 activates both MyD88 and TRIF- dependent pathways. The MyD88 pathway activates production of pro-inflammatory cytokines, while the TRIF pathway result in interferon production. Reprinted with permission from [46].

Macrophages Macrophages express a wide array of PRRs. When activated, they release pro-inflammatory cytokines and chemokines that induce diverse reactions, such as recruitment and activation of neutrophils and other immune cells, endothelial cell activation, fever, and liver production of acute phase proteins. Macrophages can be of embryonic or hematopoetic origin. Many special- ised tissue-resident macrophages, e.g. Langerhans cells in the skin, Kupffer cells in the liver, microglial cells in the central nervous system, and alveolar and peritoneal cavity macrophages, derive from the foetal liver and yolk sac, and self-renew under steady state conditions. Blood-derived monocytes

ANNA LANGE SLPI and soluble BTLA as immunological markers 19 maintain other tissue macrophage populations, for example in the gastroin- testinal tract, and constitute an important source of macrophages in inflammation. Monocyte-derived macrophages are of special importance for the resolution of inflammation, by clearing apoptotic cells and inflammatory debris, and by production of anti-inflammatory cytokines, e.g. IL-10 and TGF- [50]. Another central role of macrophages in infections is their pathogen-clearing role as tissue-resident phagocytes. Macrophages located in the marginalβ zone of the spleen, have enhanced capacity for phagocytosis, and specialise in capturing blood-borne pathogens [51].

Neutrophils Neutrophils are key effector cells of innate immunity. Severe prolonged neu- tropenia, or functional neutrophil defects, such as chronic granulomatous disease, result in recurrent bacterial and fungal infections [52]. Neutrophil production and release from the bone marrow is controlled by granulocyte colony-stimulation factor (G-CSF) [52]. The bone marrow reserve is five to six times larger than the circulating pool, and can be released in severe infections [53]. Neutrophils survive a relatively short time (hours to days) in the circulation, but once activated, by various substances that interact with receptors in the membrane, e.g. PAMPs, IgG, cytokines, and integrins, their lifespan increases [52, 54]. Vascular endothelial cells near a site of pathogen invasion are activated by PAMPs that bind to PRRs on their surface, or by cytokines released from resident cells in the surrounding tissue. The activated endothelium recruits neutrophils via upregulation of adhesion molecules, which leads to neutrophil rolling along the endothelium, adhesion, and transmigration, preferentially at endothelial junctions, but also transcellularly [54]. Extravasated, activated neutrophils cross tissues with help from integrins and granule-derived proteolytic enzymes, following a chemotactic gradient to the site of infection. They kill bacteria via: i) phagocytosis, ii) degranulation of granule contents, and iii) the extrusion of neutrophil extracellular traps (NETs), as outlined in figure 4. In NETosis, neutrophils release their DNA content covered by histones, granule-derived proteases, and antimicrobial proteins, trapping bacteria in an environment with high concentration of toxic substances [55].

20 ANNA LANGE SLPI and soluble BTLA as immunological markers

Figure 4. Neutrophil effector mechansims. Neutrophils fight off bacterial invasion by release of toxic granule contents, phagocytosis, and via extrusion of their DNA covered by bactericidal proteins. Adapted from [56] and reprinted with permission.

Neutrophils contain a plethora of antimicrobial substances, which work within phagolysosomes, or extracellularly after degranulation. These substances are stored in cytoplasmic granules, named according to when they appear in the neutrophil differentiation process: i) primary ii) secondary, and iii) tertiary. In addition to granules, there are so called secretory vesicles, which are smaller, and produced last in granulopoesis [53], table 1.

Table 1. Neutrophil granule content and propensity for exocytosis

Primary Secondary Tertiary Secretory (Azurophilic) (Specific) (Gelati- vesicles nase) Content Myeloperoxidase Lactoferrin Gelatinase A1 antitrypsin Proteinase 3 hCAP-18 Arginase 1 Alkaline phos- Neutrophil Elas- SLPI Lysozyme phatase tase α-1-antitrypsin Cathepsin G Lactoferrin Cathepsin C Gelatinase Azurocidin Pentraxin 3 α-Defensins Lysozyme (HNP) BPI Lysozyme Exocytosis Low + ++ +++ HNP: Human neutrophil peptide, BPI: Bactericidal permeability-increasing Protein, SLPI: Secretory leukocyte protease inhibitor.

ANNA LANGE SLPI and soluble BTLA as immunological markers 21

An internal control mechanism of neutrophils, which prevents collateral tissue damage, is the presence of protease inhibitors, e.g. 1-antitrypsin and SLPI, in granules. Regulation is also mediated by activating and inhibitory receptors and ligands, which can be rapidly recruited to αthe neutrophil surface, from the membranes of intracellular granules [57].

NK cells Natural killer (NK) cells are granulated innate cell lymphocytes lacking specific antigen receptors, which comprise 5-10% of circulating lymphocytes, and mediate viral defense and tumor surveillance. Activation of NK cells is complex, and involves activating and inhibitory receptors, and cytokine signals from macrophages. As IFN- -producers, NK cells activate macrophages, and control virus infections, and other intracellular or phagocytosed pathogens [58, 59]. NK cells γrecognize stressed (infected and tumor) cells, not via MHCI-associated antigen, but via other ligands on their surface, and induce apoptotic cell death by similar mechanisms to cytotoxic T cells, described below [60]. The role of NK cells in the pathogenesis of sepsis is unclear. Animal studies show that NK cells migrate rapidly to the site of an infection, and that NK-cell depletion is protective in experimental sepsis. Human studies report increased circulating numbers of NK cells during sepsis, but apoptotic loss in peripheral tissues [29].

Soluble factors

Cytokines Cytokines are small proteins that provide communication between immune cells. They regulate immune responses via specific receptors, and signal in an endocrine, paracrine or autocrine manner, at very low concentrations. There are over one hundred known cytokines, including the interferons (IFN), interleukins (IL), chemokines (e.g. CC and CXC) (which orchestrate immune cell migration), mesenchymal growth factors, tumor necrosis factor (TNF) and adipokines. Distinguishing features of cytokines are: i) pleiot- ropism, meaning that a specific cytokine has several effects depending on the target cell, and ii) redundancy, which denotes that cytokines overlap in function [61]. Cytokines are often classified as either pro- or anti-inflammatory, where IL-1, IL-6 and TNF are important pro-inflammatory cytokines, and IL-4, IL-10 and TGF- are typical anti-inflammatory. The same

22 ANNA LANGE SLPI and soluble BTLA as immunological markers cytokine may however have both pro-and anti-inflammatory effect in different circumstances, and depending on the target cell.

Complement The complement system consists of circulating soluble liver-derived proteins, which may kill pathogens non-specifically, and eliminate them via op- sonization. The alternative complement pathway has a sentinel function, on account of its activation directly on microbes, which lack the complement inhibiting surface molecules that are expressed on host cells. The classical complement pathway, requires antibodies attached to pathogens to be activated, and the lectin pathway, is activated by mannose-binding lectin (MBL) that bind to sugars on the surface of bacteria and fungi [5]. All three pathways converge on cleavage of C3, which generates C3a, which recruits and activates inflammatory cells, and C3b, which is an op- sonic factor that also amplifies the complement activating loop, generating more C3b. Receptors for C3b is expressed on macrophages and activated neutrophils, to facilitate phagocytosis. C3b also participate in cleavage of C5, which releases the potent pro-inflammatory molecule C5a, and activates formation of membrane attack complex (MAC), which forms lytic pores in bacterial membranes [5]. In sepsis, there is systemic hyperactivation of complement, with a burst of C3a and C5a production. Blockade of C5a with monoclonal antibodies has been successful in animal models of sepsis, and results from a phase II RCT on the effect on septic organ dysfunction is pending (NCT02246595) [62].

Antimicrobial peptides and proteins The first antimicrobial peptide (AMP), cecropin, was sequenced in 1981 [63] Since then, >3000 antimicrobial peptides (AMPs) and proteins have been characterized, forming part of innate immunity in all multicellular organisms. There are 139 human AMPs, often referred to as host defence peptides (HDPs), due to their immune-regulatory functions [64]. AMP size varies from 10 to 150 amino acid residues [65]. They are secreted by epithelial cells, or work inside white blood cells on phagocytosed organisms, and are attracted to negatively charged structures on microbial membranes via their positive charge [65, 66]. AMPs form pores, or desta- bilize microbial membranes, by inserting or attaching to their surface. In addition, they interfere with intracellular processes, such as enzyme activity, and DNA or protein synthesis [67]. Antimicrobial spectrum varies, and may include bacteria, viruses, fungi and parasites. The immunological effects of

ANNA LANGE SLPI and soluble BTLA as immunological markers 23

AMPs are elusive, and include being chemoattractants, intervening with TLR-signaling, induction of cytokine and chemokine production, inflam- masome activation, cell maturation, and regulation of cell death processes [68, 69]. Two well characterized human AMP groups are defensins ( and ), and cathelicidin. Among the -defensins, HNP 1-4 are mainly produced in neutrophils, and stored in primary granules, while HD-5 and 6 areα secretedβ by cells in the intestine [65,α 68]). Beta-defensins, are produced by epithelial cells, in response to pro-inflammatory stimuli [69] Cathelcidin derives from the pro-peptide hCAP-18, which is cleaved to release LL-37 [65]. hCAP-18 is a component of secondary granules of neutrophils, and is inducibly expressed in many epithelial cells [69]. AMPs have long been viewed as promising candidate drugs to treat infections, but their systemic use is limited by their instability and toxicity [70]. Apart from their own bactericidal effect, AMPs can work synergisti- cally with antibiotics, by increasing membrane permeability [68, 71]. Bac- terial AMP resistance mechanisms have been reported [67]. A few AMPs are in clinical practice, for topical treatment of ear, eye and skin infections (polymyxin B and bacitracin), and systemically in complicated soft tissuse and joint infections (daptomycin), and infections with resistant gram-negative bacteria (colistin) [72]. Many experimental and clinical studies have been pursued, on both natural and synthetic AMPs, for a wide array of infections, inflammatory conditions (rosasea, acne), and for wound treatment (venous leg and diabetic foot ulcers) [70, 71, 73].

Dendritic cells Dendritic cells (DCs) are the sentinel cells of innate immunity, and essential for induction of the adaptive immune response. Conventional DCs (cDCs), the typical antigen-presenting cells, reside in most organs, but most abundantly in the skin and at mucosal barriers. They sample the surrounding tissue for self and foreign antigens through receptor-mediated phagocytosis and macropinocytosis, process the antigens, and present peptide fragments on MHC class I and II on their surface. DCs also express a wide array of PRR, and are activated in the presence of PAMPs in the tissue [74]. Activation leads to expression of co-stimulatory molecules (i.e.CD80, CD86 and CD40), and CC-chemokine receptor 7 on the DC surface, which facilitates chemokine-driven migration into lymphatic vessels, and transportation to draining lymph nodes, or other secondary lymphoid tissues, where they present antigens to naïve CD4+ and CD8+ T cells [75].

24 ANNA LANGE SLPI and soluble BTLA as immunological markers

Protective immune responses: adaptive immunity Although functionally distinct, adaptive immunity act in synergy with innate immune responses to eradicate infections, and to generate immunological memory. Adaptive immune responses have the three hallmarks: i) specificity, ii) memory and, iii) adaptability.

T cells T cells proliferate, differentiate, and undergo selection in the thymus. Cells that do not interact with MHC, or that bind too strongly to MHC-antigen complexes, are deleted. The rest differentiate into CD4+ and CD8+ T cells, and leave the thymus, after a last round of selection that removes auto-reactive cells [76]. Naïve T cells have a long life-span. They recirculate between secondary lymph organs every 12 to 24 hours, to browse for antigens. At any moment in time, the blood contains only 2 to 3 percent of the total T cell pool [77]. T cell activation, e.g. in a bacterial infection, requires a danger signal delivered by activated dendritic cells. Naïve CD4+ cells meet their MHCII- associated antigen, on a dendritic cell in a secondary lymphoid organ, and the T cell receptor (TCR) attaches. CD28 on the T cell surface then ligates with CD80/86 (B7.1/B7.2) on the dendritic cell. This provides the necessary co-stimulation, which results in cytokine (e.g. IL-2) production and full T cell activation. CD4+ cells proliferate, and differentiate into different types of functional effector cells, depending on cytokine-mediated instructions from innate immune cells [9, 78]. Superantigens are bacterial toxins that bypass antigen-presentation, and induce polyclonal CD4+ cell (up to 20 %) activation, via cross-linkage of MHCII and TCRs outside the antigen specific regions. This leads to a sudden massive release of T cell cytokines (IL- 2, IFN- , TNF- ), which can induce the so called “toxic shock syndrome” [79]. CD4+γ (helper αT) effector cells are distinguished by the cytokines they se- crete, and mediate specific functions in the protection against pathogens.

ANNA LANGE SLPI and soluble BTLA as immunological markers 25

The main effector cell subsets, their associated cytokines, and/or leading functions with regard to infections are:

i) T helper 1 (TH1): IFN-γ. Protect against intracellular pathogens, such as mycobacteria, certain fungi, and viruses. ii) T helper 2 (TH2): IL-4, IL-5, IL-9, IL-10, and IL-13. Protect against extracellular parasites. iii) T helper 17 ((TH17): IL-17. Fight of extracellular bacteria and fungi at mucosal surfaces via neutrophil recruitment [80]. iv) Follicular helper cells (TFH): Essential for development of plasma cells and memory B cells [81]. v) Regulatory T cells (Treg): IL-10, TGF- . Suppress immune responses to maintain self-tolerance and immune homeostasis [78]. β CD8+ cells are primed by two-step signals similar to CD4+ cells, but via MHCI. In contrast to MHC II, all nucleated cells express MHCI, but CD8+ cells respond best to antigens presented by DCs. Since not all viruses infect DCs, this is assumed to take place via so called cross-presentation, whereby DCs present antigen that derives from ingested parts of infected cells on MHCI [82]. Usually, CD8+ cells need CD4+ help (via DC stimulation) to differentiate into efficient cytotoxic T cells, also called cytotoxic T lymphocytes (CTL) [74]. CTLs are important, in particular in the defense against intracellular pathogens, and for cancer surveillance. They bind to infected or cancer cells, via antigen-MHCI complexes, and release granules containing perforin and granzymes. Perforin creates membrane pores which allows granzymes to enter the target cell and induce apoptosis [83] After elimina- tion of the infecting pathogen, most cells in the CD8+ clone die by apoptosis, but some survive and mature into memory CD8+ cells, which can rapidly generate CTLs in a second exposure to the pathogen [58].

B cells Like T cells, B cells can recognize an immense number of antigens. T cell- independent activation of B cells is typical for polysaccharide antigens, and results in low-affinity IgM antibodies. So-called Marginal Zone B cells, located in the spleen, specialize in mounting rapid T cell-independent responses to blood borne antigens, and have enhanced phagocytic ability [84]. Splenectomy leads to an increased risk of fulminant sepsis caused by encap- sulated bacteria [85]. T cell (TFH) dependent activation of follicular B cells occurs in response to protein antigens. This results in class-switching (IgM

26 ANNA LANGE SLPI and soluble BTLA as immunological markers to IgG, IgA and IgE), antibody affinity maturation, and development of memory cells (requires formation of the germinal center) [86].

Co-signaling proteins in regulation of immune responses Numerous activating and inhibitory interactions between T cells and APCs regulate the magnitude, and duration of the immune response. This is mediated by co-signalling molecules, which belong to either of two families based on their structure: i) the immunoglobulin superfamily (IgSF) and ii) tumor necrosis factor receptor superfamily (TNFRSF). As a rule, co-signalling proteins interact with ligands belonging to the same family. Two important co-inhibitors are Cytotoxic T lymphocyte associated protein 4 (CTLA-4) and Programmed death-1 protein (PD-1), which are depicted along with other co-inhibitors in figure 5.

Figure 5. Co-inhibiting receptors and ligands. PD-1: Programmed death 1, PD-L1: Programmed death ligand 1, HVEM: Herpes Virus Entry Mediator, BTLA: B and T lymphocyte attenuator, LIGHT: homologous to Lymphotoxin exhibits Inducible expression and competes with HSV Glycoprotein D for binding to Herpesvirus entry mediator, a receptor expressed on T lymphocytes, CTLA-4: Cytotoxic T lymphocyte antigen 4, MHC II: Major histocompatibility complex II, LAG-3: Lymphocyte activation-gene 3, CEACAM-1: Carcinoembryonic antigen-related cell adhesion molecule-1, TIM-3: T cell membrane protein-3. Modified, and reprinted with permission from [87].

ANNA LANGE SLPI and soluble BTLA as immunological markers 27

CTLA-4 in expressed on the T cell surface soon after activation, and binds to CD80/CD86 on APCs, with much higher affinity than CD28, thereby opposing further co-stimulation [88, 89]. PD-1 is structurally similar to CTLA-4, but more widely expressed, i.e. on T cells, B cells, NK cells, monocytes and dendritic cells [90]. While CTLA-4 regulate early T cell activation, PD-1 limits later immune responses, in the peripheral tissues [91]. PD-1 has two ligands, PD-L1, which is expressed by a variety of cells, including non-hematopoetic cells, and PD- L2, which is mainly, but not exclusively, expressed on APCs [90]. The importance of co-inhibitory pathways in maintaining homeostasis is demonstrated by lymphoproliferative and autoimmune manifestations in mice lacking these genes [92, 93]. CTLA-4, PD-1 and PD-L1 expression have been shown to be increased in sepsis, and this is believed to reflect a state of immune suppression [94-96]. Blockade of the PD-1 pathway in sepsis is under current investigation, as will described in the context of immune therapy.

Dysregulated immune response in sepsis The hallmark of sepsis, is a dysregulated immune response that causes organ dysfunction, rather than protecting the host. The acute phase typically features excessive inflammation, which is mediated by cytokines that activate cascades of other pro-inflammatory mediators, including, histamine, plate- let activating factor and prostaglandins [97]. This induces generalized activation of neutrophils, complement, endothelial cells, and platelets, which leads to release of tissue-toxic substances, intravascular coagulation and thrombosis, and disrupted endothelial integrity [15]. While pro-inflammation dominates initially, there is accumulating evidence of early, and progressive immune suppression in sepsis. The first observations suggestive of a contributing pathogenic role of immune suppression, were reports of reduced HLA class II antigen expression and cytokine secretion, in monocytes isolated from sepsis patients with a fatal outcome. This was named “immunoparalysis” [98]. In 1996 the term CARS (Com- pensatory Anti-inflammatory Response Syndrome) was coined, to describe a two-sided septic immune response, which could explain why anti-inflammatory sepsis trials had failed [99]. Additional evidence of sepsis-associated immune suppression came from studies showing that sepsis was associated with late-phase re-infections by opportunistic pathogens [100]. Post mor- tem studies reported unresolved septic foci, loss of immune cells, T cell an-

28 ANNA LANGE SLPI and soluble BTLA as immunological markers ergy, and upregulated inhibitory proteins, along with downregulated activating proteins, on immune cells and in tissues [101-103]. Consistent with these findings, sepsis, and septic mortality, have been shown to be associated with reactivation of latent viruses, in particular Cytomegalovirus [104]. Immune suppression is now considered to be of central importance in the immune pathology of sepsis, and to involve apoptotic loss of innate and adaptive immune cells, and disrupted effector functions, in particular affecting T cells, i.e. “T cell exhaustion” [103, 105]. Functional aspects of sepsis- associated immune suppression are [97, 103]:

i) Neutrophils: Chemotactic activity and bacterial clearance ↓ ii) DCs: HLA-DR expression ↓, IL-10 secretion ↑. Inability to induce robust T cell responses. iii) Monocytes and macrophages: Pro-inflammatory cytokine responses to TLR-stimulation ↓ HLA-DR expression ↓ iv) NK cells: IFN-γ ↓ v) CD4+: Suppressed TH1 and TH2, and TH17 responses. In- creased proportion of Treg cells. Based on the evidence of simultaneous and sustained alterations in innate and adaptive immunity, a new conceptual model of the immune dysfunction in sepsis has been proposed. While the interplay between inflammation and immunosuppression is still ill-defined, this model describes the co-existence of inflammation and immunosuppression in the acute phase of sepsis, and the sustained immunological derangement that contributes to late phase morbidity and deaths [103, 106], figure 6.

ANNA LANGE SLPI and soluble BTLA as immunological markers 29

Figure 6. Early and late immune alterations and outcome in sepsis. Reprinted with permission from [107].

Therapies targeting immune dysregulation in sepsis While recent advances in the understanding of sepsis pathobiology has motivated immune-boosting approaches, some strategies aiming to suppress immune-activation are still under debate, or investigation.

Cortisone The first RCT on adjunctive treatment in sepsis was published in 1976, and reported a significant survival benefit for a bolus dose of cortisone [108]. Subsequent studies have shown inconsistent results. A Cochrane review from 2015 reported improved survival with prolonged cortisone courses (>3 days) at lower doses (<400 mg daily), but a later metaanalysis failed to demonstrate an advantage [109, 110]. Two recently published large RCTs, again showed conflicting results. The ADRENAL study, showed no effect on short-term mortality, while the APPROCCHSS study, which added fludrocortisone to the hydrocortisone arm, showed a slight survival benefit

30 ANNA LANGE SLPI and soluble BTLA as immunological markers at 90 days [35, 37]. The current place for cortisone in sepsis therapy is in vasopressor-resistant septic shock [44].

Intravenous immunoglobulin (IVIg) Polyclonal intravenous immunoglobulin (IVIg) could theoretically neutral- ize bacterial toxins and inflammatory mediators, and modulate the immune response in sepsis, but the current evidence for improved survival is weak [111]. Recently, an RCT on necrotizing soft tissue infection showed no survival benefit [112]. Invasive Group A streptococcal (GAS) infections have high case-fatality rates, and has been shown to be associated with significantly lowered levels of protective antibodies against superantigens and M- protein, which provides a rationale for the use of IVIg in this patient group [113, 114]. A small RCT, including 21 patients with GAS toxic shock syndrome, whereof 13 had deep tissue infection, reported a non-statistically significant 3.6-fold higher 28-day mortality rate in the placebo group [115]. Due to low quality of evidence, the SSC guidelines advice against the use of IVIG in sepsis, but encourage new trials [44].

Blood purification Another strategy to remove toxic substances is blood purification, for which there is two basic approaches. High volume hemofiltration (HVHF) was recently subject to a Cochrane metaanalysis, and did not reduce mortality [116]. The second strategy, cytokine hemadsorption, was studied in a non- blinded pilot RCT (N=20) that did not show effect on the endpoint (organ failure at 48 hours) [117]. Despite this, the technique is widely used, and is followed in a prospective case study [118]. Blood levels of endotoxin can be elevated irrespective of the causative pathogen in sepsis, possibly due to increased gut permeability [119]. High endotoxin levels were shown to be associated with organ failure [120]. Pol- ymyxin B (PMB) binds irreversibly to LPS. Previous studies on extracorporeal PMB hemadsorption for septic shock have produced heterogenous results, but a recent multicenter study, showed no effect on mortality or other clinical outcomes, nor did it lower endotoxin activity [36, 121]. In conclusion, there is no evidence supporting the use of blood purification is sepsis.

Granulocyte-macrophage colony stimulating factor Granulocyte-macrophage colony stimulating factor (GM-CSF) stimulates proliferation and differentiation of myeloid bone-marrow precursors, but also survival and activation in mature neutrophils and macrophages [122].

ANNA LANGE SLPI and soluble BTLA as immunological markers 31

This provides a rationale for its use as adjunctive treatment in sepsis, i.e. to improve pathogen clearance. There is, however, no current evidence from clinical trials to support this. GM-CSF was administered after the initial inflammatory phase in four small RCTs. Results from a metaanalysis indicated an effect on time to recovery, and on HLA-DR expression, but there was no effect on short-term mortality [123].

IFN-γ Interferon gamma (IFN- ) is produced by TH1, activated CD8+, and NK cells, and activates macrophages and induces HLA-DR expression [78, 124]. Reports on sepsis-associatedγ monocyte deactivation in the 1990’s, suggested IFN- -treatment as adjuvant sepsis treatment [125]. A pilot study on septic patients showed restoration of HLA-DR expression [126]. There are still noϒ published results from randomized trials, but case reports describe positive effects in invasive fungal infections [127, 128].

IL-7 Persistent lymphopenia is associated with worse outcome in sepsis [129]. Interleukin-7 (IL-7) is a non-redundant growth factor necessary for naïve T cell survival and proliferation [130]. The immune-restorative effect of recombinant human IL-7 (CYT107), on septic shock patients with severe lymphopenia, was investigated in a recent phase II study (IRIS-7), with positive results regarding CD4+ and CD8+ cell counts [131]. A follow-up study is currently recruiting (NCT03821038).

Checkpoint inhibition The success of checkpoint inhibitors (i.e. CTLA-4 and PD-1 blockade) in advanced cancer, and the evidence of immune cell exhaustion in sepsis, has motivated trials of checkpoint-inhibition in sepsis [132-134]. Ex vivo studies on immune cells from septic patients, and results from experimental sepsis models in mice, provide evidence that blocking the PD-1-PD-L1 pathway can restore immune cell function, and be favourable in terms of organ failure and mortality [135-137]. Three recent phase II studies investigated PD-1 and PD-L1 blockade in patients with sepsis >24 hours, and lymphopenia, and reported positive effect on immune function, defined as increased mHLA-DR expression. Im- mune-related side effects were common [138-140]. CTLA-4 blockade has not been studied in human sepsis, but is reported to reduce T cell apoptosis in mice [141].

32 ANNA LANGE SLPI and soluble BTLA as immunological markers

Secretory leukocyte protease inhibitor (SLPI) Secretory leukocyte protease inhibitor (SLPI) was first isolated from bronchial secretions, and identified as an inhibitor of neutrophil elastase (NE) [142]. It was later purified in saliva, and sequenced to a 107 amino acid residue peptide [143]. SLPI is secreted by various epithelial cells, by macrophages and dendritic cells, and is stored in the secondary granules of neutrophils [144-147]. Be- sides NE, the antiprotease activity of SLPI includes Cathepsin G produced in neutrophils, and mast cell-derived chymase and tryptase [148]. SLPI production is induced by NE, and by inflammatory stimuli, such as PAMPs and pro-inflammatory cytokines [149-151]. SLPI has a larger role in immunity than protecting epithelia from protease-mediated degradation in inflammation, figure 6. First, it has antimicrobial activity, which, like for other host defense peptides, is associated with its positive charge. The antimicrobial spectrum includes gram-positive and gram-negative bacteria, various fungi and mycobacteria [148, 152]. Addi- tionally, SLPI has implications in viral infections. It was shown that high levels of SLPI in vaginal secretions of untreated HIV positive women was associated with protection from perinatal transmission [153, 154].

Figure 6. SLPI in infection and immunity. Schematic illustration of the protective and regulating functions of SLPI. Reprinted with permission from [148].

ANNA LANGE SLPI and soluble BTLA as immunological markers 33

SLPI has attracted a lot of interest as a modifier of inflammation. A seminal study in mouse macrophages indicated that LPS-induced SLPI expression could suppress the production of pro-inflammatory cytokines [145]. It was further demonstrated that SLPI attenuates TNF- -induced neutrophil oxidative burst, and TLR-mediated responses in monocytes and macrophages [150, 155]. Later, it was identified that the antiα-inflammatory effect of SLPI was mediated via inhibition of NFĸB activation [156, 157]. In line with this evidence, SLPI deficient mice were shown to have increased susceptibility to LPS-induced shock [158]. It has been proposed that SLPI, via its anti-inflammatory mechanisms, is important for maintaining tolerance to harmless microbes at mucosal surfaces [146, 159]. Another immunological effect of SLPI is attenuation of the formation of NETs, via inhibition of NE-mediated histone cleavage in the cell nucleus, a key event in NETosis [160]. Finally, SLPI has been ascribed a potential role in neutrophil maturation and survival [161]. Many studies cover the mechanisms of SLPI secretion, and its protective role at epithelial barriers [148]. SLPI was shown to be decreased in infectious COPD exacerbations, and to be involved in the disturbed protease/antiprotease balance behind the pathology of chronic infections in cystic fibrosis [162, 163]. There is pre-clinical evidence of a local protective effect of SLPI in CNS trauma and ischemic injury, and in cardiac ischemia, which could be mediated via its anti-protease and anti-inflammatory activity [164- 166]. The circulating levels of SLPI are significantly lower (nanomolar compared to micromolar) than in mucous secretions [150, 162]. SLPI was shown to be elevated in serum and plasma of patients with pneumonia, sepsis, and experimental endotoxemia [150, 167]. SLPI was also increased in acute pancreatitis, and acetaminophen-induced liver failure [168, 169] .

34 ANNA LANGE SLPI and soluble BTLA as immunological markers

BTLA B and T lymphocyte attenuator (BTLA), also known as CD272, a member of the IgSF, was the third T cell co-inhibitor to be described, after CTLA-4 and PD-1, as a receptor induced on CD4+ cells during activation. BTLA was shown to be expressed on both B and T cells, and deletion of the BTLA gene in mice resulted in increased T cell proliferation after stimulation [170]. Further studies showed that BTLA is also expressed on naïve T cells and APCs, and that contrary to what was observed in T cells, BTLA expression decreased following B cell activation [171, 172]. Soon, in became clear that BTLA differed from other co-inhibitors in several aspects. First, opposing the general rule, a member of the TNFRSF, herpes virus entry mediator (HVEM), was identified as the BTLA counter- receptor. As the name implies, HVEM mediates herpesvirus entry into host cells [173]. Secondly, HVEM, already had two other known ligands, LIGHT [lymphotoxin-like, exhibits inducible expression, and competes with herpessimplex virus glycoprotein D (gD) for HVEM, a receptor expressed by T lymphocytes] and lymphotoxin alpha (LT ). In addition, it was shown that LIGHT and BTLA could bind simultaneously to HVEM at topographically different binding sites [173]. Later, a αfourth ligand for HVEM was identified, CD160 [174]. The HVEM signaling network can convey both activating (LIGHT and LT ) and inhibiting (BTLA and CD160) signals, but the net function of HVEM is inhibitory, as demonstrated by increased inflammatory responses in HVEMα knockout mice [175, 176]. Both BTLA and HVEM are widely expressed in tissues, and among immune cells (T cells, B cells, NK cells, DCs and macrophages) [175]. An additional layer to the complexity of this signaling system, is the dual role of HVEM as both receptor and ligand. Trans- interaction between BTLA and HVEM on different cells results in an inhibitory signal via BTLA, but a stimulating signal to the HVEM wearing cell, while cis-interaction between BTLA and HVEM on the same cell, prevents HVEM-mediated activation [177], figure 7.

ANNA LANGE SLPI and soluble BTLA as immunological markers 35

Figure 7. BTLA-HVEM interaction. BTLA and HVEM interaction can occur between cells (trans), or on the same cell (cis). Trans-interaction results in bidirectional signalling: stimulation on the HVEM side, and inhibition on the BTLA side. Cis- interaction between BTLA and HVEM blocks trans-interaction, and prevents HVEM-mediated activation. Published with permission from [178].

The BTLA-HVEM pathway is a recent focus of interest in autoimmune disease and as a target in cancer immune therapy [179, 180]. There are few reports of BTLA expression on human immune cells in sepsis, and results deviate. Two studies report increased CD4+ BTLA expression in sepsis and another found gradually decreasing BTLA expression with increasing severity [96, 181, 182]. BTLA-knockout mice were protected from septic mortality in an experimental model, but administration of anti-BTLA antibody to septic mice resulted in increased organ damage and mortality [183, 184]. Recently, it was shown that a soluble BTLA isoform, lacking the transmembrane region, is produced to greater extent relative to the full-length receptor in mice subjected to hemorrhagic shock, followed by sepsis. This study also investigated the biological significance of soluble BTLA, by studies on mouse splenocytes. It found that sBTLA induced increased IL-6 and IL-9 production, and increased proliferation of splenocytes from critically ill mice [185].

36 ANNA LANGE SLPI and soluble BTLA as immunological markers

Aims

• To study antimicrobial peptide and SLPI plasma concentrations, in patients with community-acquired pneumonia (CAP), compared to patients with non-respiratory infections, and non-infected controls, and to study AMP concentrations in relation to severity and bacterial etiology in CAP (Paper I). • To further study SLPI in relation to the focus of infection, bacterial etiology, disease severity, and immunological markers, over time, in community-onset bloodstream infection (BSI) (Paper III). • To evaluate the usefulness of soluble co-inhibitors as sepsis biomarkers, and indicators of disease severity and short term prog- nosis in ICU-treated patients (Paper II). • To further explore the utility of sBTLA as a prognostic biomarker, and to evaluate its possible association with markers indicative of inflammation and immune suppression, in a less severely ill infected patient cohort, i.e. BSI. To assess the dynamic expression of sBTLA in patients and controls, and the association to disease severity and bacterial etiology (Paper IV).

ANNA LANGE SLPI and soluble BTLA as immunological markers 37

Subjects and methods

Patients This thesis is based on three clinical studies, which differ in terms of the focus of infection, disease severity, and presence or not of bacteremia. This enabled us to look at the studied markers from different perspectives.

Paper I The study population in paper I consisted of a subgroup of patients enrolled in a larger study, including 294 patients hospitalized at the department of Infectious Diseases at Örebro University Hospital, for suspected community-acquired pneumonia (CAP) [186]. Hospitalization the preceding month was the single exclusion criterion. Criteria for CAP were acute onset of illness, radiological signs of pulmonary consolidation and at least two of the following signs or symptoms: fever of ≥38°C, dyspnea, cough, pleuritic chest pain, or abnormal lung auscultation. Out of 235 patients with CAP, 92 had a plasma sample collected on hospital admission, out of which 89 had sufficient plasma for analyses, and was included in the study, along with 21 patients with non-respiratory tract infection (non-RTI), and 63 patients scheduled for orthopedic or urological surgery (controls). While not reported in the paper, the study groups did not differ significantly with respect to age, but the non-RTI had a lower proportion of female study subjects.

Paper II The study population consisted of patients ≥18 years of age (N=129), who were treated at the intensive Care Unit at Karolinska University Hospital in Huddinge, Sweden. They had either been diagnosed with severe sepsis or septic shock within the last 24 hours before enrolment (Sepsis, N=101), or had been admitted due to non-infectious critical illness the last 24 hours (ICU controls, N=28). Sepsis was defined according to Sepsis-1 criteria, and there were no exclusion criteria. Blood and plasma donors (N=31, not age- and sex-matched), were included as a control group. The sepsis and ICU control cohorts did not differ with respect to age, and sex, however, a greater proportion of subjects in the sepsis cohort had malignancy, or were immunocompromised, whereas a greater proportion of ICU controls had diabetes. Sepsis patients, and ICU controls, were sampled for plasma at five time-points: at study enrolment (day 0), day one, and days 2, 4 and 6.

38 ANNA LANGE SLPI and soluble BTLA as immunological markers

Papers III and IV The study population of papers III and IV was based on the prospective study “Dynamics of sepsis”, conducted at Örebro University hospital between 2011 and 2014, and designed to follow bacteremic patients, with repeated sampling over four weeks, with focus on bacterial DNA and immunological markers. Patients ≥18 years, admitted to the Departments of In- fectious Diseases and Internal Medicine, in whom a blood culture drawn on hospital admission showed clinically significant bacterial growth within three days, were eligible for inclusion. Exclusion criteria were infection with HIV, hepatitis B or C, and previous inclusion in the study. The study protocol included blood sampling on hospital admission, and days 1-2 (enrolment), 3, 7±1, 14±2, and 28±4. In paper III, 109 patients from whom at least one plasma sample was available for analysis, were included, and in paper IV, 108. Thirty-one blood and plasma donors (the same as in paper II) were used as healthy controls, and plasma from days 0 and 28 were analyzed.

Storage of plasma Plasma from all study subjects was stored in biobanks at -70 to -80 °C. All samples in the individual studies, including control plasma, were treated equally.

ANNA LANGE SLPI and soluble BTLA as immunological markers 39

Enzyme-linked Immunosorbent assay (ELISA) ELISA is a technique that enables detection and quantification of a specific antigen or antibody in a clinical or experimental sample. It comes in many variants, and the choice of method may depend on the nature of the antigen, or the specificity of the antibody that one aims to quantify. Figure 8 shows two basic ELISA methods that are typically performed on a plastic multi- well plate.

Figure 8. Principles of Enzyme-linked immunosorbent assay (ELISA). Between each step, unbound antibody or antigen is removed in a rinsing step.

In the Indirect assay, the sample, e.g. plasma or serum, is added to a plate well, upon which proteins that are present in the sample will bind non-specifically to the plastic surface. Unlabeled primary antibodies, specific to the antigen of interest, are added to the well. Following a rinsing step that removes unbound antibody, secondary enzyme-conjugated antibodies, which binds to the primary antibodies, are added. After another rinse, a chromogenic substrate for the conjugated enzyme, is added. There are different enzyme systems, the most commonly used being alkaline phosphatase and horseradish peroxidase (HRP). The enzyme-reaction results in a colored product, and the intensity of the color is proportional to the amount of antigen in the sample. Multiple secondary antibodies can bind to the same primary antibody, which amplifies the enzyme reaction.

40 ANNA LANGE SLPI and soluble BTLA as immunological markers

The Capture assay (also called “Sandwich” because the antigen of interest is sandwiched between two antibodies) allows detection of very low concentrations of antigen, and is the method used in most commercial ELISA kits. Instead of antigen, the wells are coated with a fixed amount of antibodies, specific to the antigen of interest. The sample is added, and if it contains the antigen, this will bind to the immobilized antibodies. Next, a second, enzyme-labelled, detector-antibody is added, which binds to another epitope on the antigen. After this step, the capture assay can follow the principles described above, however, a common modification is the use of biotin-labelled, instead of enzyme-conjugated antibodies. Biotin is, in fact, a B-vitamin, that is used in biotechnology due to its extreme avidity to the tetrameric bacterium-derived protein streptavidin. Streptavidin, in turn, can be conjugated to an enzyme, and when added to a well containing cap- tured antigen, several streptavidin-enzyme complexes can bind to each bio- tinylated antibody, which enhances the enzymatic reaction that follows addition of the chromogenic substrate. The result of an ELISA can be read either as “positive or negative”, or be quantified, using a plate reader. To allow for determination of antigen concentration in a sample, a fitted standard curve is created for each run, based on a step-wise diluted reference (standard) sample, containing a known concentration of antigen or antibody. While ELISAs are relatively simple to perform, there are sources of concentration estimation error, which are important to take into account. Intra- assay variability, due to for example uneven coating, can be controlled for by analyzing samples in duplicate, or triplicate, and inter-assay variations can be controlled for by including internal and blank controls.

SLPI and sBTLA detection Commercial biotin and HRP based Sandwich ELISA kits were used for detection of sBTLA and SLPI. Detection ranges were SLPI: 78 to 5000 pg/mL, and BTLA: 0.45 (0.47) to 30 ng/mL.

Sample dilution Plasma samples were diluted before analyses according to the manufactur- ers’ instructions: for SLPI detection 1:100, and for BTLA 1:5 in paper II, and as a result of lower BTLA concentrations in the less severely ill cohort of paper IV, we here diluted plasma 1:2.

ANNA LANGE SLPI and soluble BTLA as immunological markers 41

Blood and other cultures Culturing procedures were specified in papers I, III and IV. Blood cultures were performed on all patients and controls in I, and on patients in III and IV, each consisting of 20 mL blood distributed equally between one aerobic and one anaerobic bottle, and incubated in a Bactec blood culturing system. In paper I, nasopharyngeal aspirates were retrieved from patients and controls, and sputum samples from CAP patients. Other cultures in paper I, III and IV were performed according to clinical suspicion, and were analyzed with accredited diagnostic methods.

Clinical data Clinical data, as well as information on culture results (other than blood cultures in III) were extracted retrospectively from patient records.

Comorbidity and severity scores Scoring systems can be used as clinical decision tools, and for research pur- poses. Different scores were used in papers I-IV to describe the cohorts, to correlate the studied markers to severity, and to control for severity in mortality analyses. The Pneumonia Severity Index (PSI), and CURB-65 scores were developed to identify patients with CAP at high risk of death in 30 days, to help decide for inpatient or outpatient treatment. PSI score assessment is cum- bersome, and divides patients into five risk classes based on age, sex, co- existing disease, abnormal physical and laboratory findings, and chest x-ray findings [187]. CURB-65 is based on the presence of Confusion, Urea ni- trogen concentration >7mmol/L, Respiratory rate ≥30, systolic Blood pressure <90 or diastolic ≤60, and age ≥65. A simplified version of the CURB- 65 score, CRB-65, includes only clinical markers [188]. PSI was shown to have higher sensitivity, but lower specificity compared to CURB-65 [188]. APACHE II is a risk score for in-hospital mortality in critical illness, assessed during the first 24 hours after ICU admission [189]. It is based on age, previous health status (severe organ failure, recent surgery, immune suppression), and 12 physiological and laboratory parameters. The Sequential organ Failure assessment score (SOFA) was developed for dynamic assessment of severity of organ dysfunction in sepsis, and was later shown to correlate to survival also in other critical illness [190], Table 2.

42 ANNA LANGE SLPI and soluble BTLA as immunological markers

Table 2. Sequential organ Failure assessment score (SOFA), adapted from [21].

SOFA score Organ system 0 1 2 3 4

Respiration1 ≥53.3 <53.3 <300 (40) <26.7 with <13.3 with respiratory respiratory support support

Coagulation2 ≥150 <150 <100 <50 <20

Liver3 <20 20-32 33-101 102-204 204

Cardiovascular MAP MAP<70 Dopamine Dopamine Dopamine ≥70 <5 or 5.1-15/ ep- >15/epi- doubuta- inephrine nephrine mine (any ≤0.1/ nore- >0.1/nore- dose)b orepineph- pinephrine rine ≤o.1 >0.1

Central nervous 15 13-14 10-12 6-9 <6 system4

Renal5 <110 110-170 171-299 300-440 >440 (<500) (<200)

1 2 9 3 4 PaO2/FiO2, kPa, Platelets x 10 /L, Bilirubin, µmol/L, Glasgow Coma Scale (3- 15), MAP= mean arterial pressure mmHg, b µg/kg/min for at least 1 hour, 5 µmol/L (urine output, ml/day)

Statistics

Group comparisons In paper I, t-test was used to compare mean antimicrobial peptide concentrations in the three study groups. P-values were corrected for multiple comparison with the Bonferroni method. In paper II, we performed Kruskal Wallis test, followed by pair-wise Mann Whitney, and post hoc Bonferroni-Holm correction for multiple comparisons.

ANNA LANGE SLPI and soluble BTLA as immunological markers 43

In papers III and IV, study group, and subgroup differences, were assessed with linear regression, with log-transformed SLPI and sBTLA as outcome, and adjustment for relevant confounders, i.e. age, sex and severity.

Associations and correlations Parametric (I) and non-parametric (II and III) correlation analyses, at individual time-points, were performed between the studied markers, and known markers of inflammation and immunosuppression, and presented as

Pearsons’s r (I), and Spearman’s rho (rs) (II and III). In paper IV, overall, and time-stratified, associations to other biomarkers were assessed with Linear Mixed Model analyses, with random intercept, log10 sBTLA as outcome, and days after hospital admission as fixed factor.

Confounding Adjustments for possible confounders were made in papers II-IV. In IV, the studied marker (sBTLA) was negatively associated with age on day 1-2, hence age-adjustment was made in all subsequent analyses. Sex and age are possible confounders in mortality analyses, and were included in the adjusted Cox models in II and IV. Baseline-comorbidities were categorized according to number of comorbidities (III), or into comorbidity score groups (IV) in adjusted analyses. We also adjusted for initial disease severity, APACHE II score in (II), and day 1 ∆SOFA in (IV).

Missing data In papers II-IV, we had to deal with missing data over time. In II, missing values were imputed, to allow for time-varying mortality analysis, by multiple imputation, with baseline data as predictors for missing values [191]. In papers III and IV, we performed missing data analysis regarding baseline factors, for patients who were sampled at all 6 time-points, and those with incomplete plasma series (III), and those with ≤1 missing sample or ≥2 (IV).

Longitudinal data analysis Dynamic expression of SLPI and sBTLA, over time, and concentration differences depending on bacterial etiology, at individual time-points (III and IV), were evaluated with Linear Mixed Model analyses, with random intercept. The respective marker was the outcome variable in the Mixed Model analysis, and time, etiology, and their interaction term, were fixed factors.

44 ANNA LANGE SLPI and soluble BTLA as immunological markers

Mortality analyses We used unadjusted and adjusted Cox Proportional Hazard models, with time to mortality as outcome, to study associations of the studied markers with short-and long-term mortality (papers II and IV). The studied markers were analyzed on continuous log scale, and categorized into two (IV) or three (II) groups based on concentration, to study dose-dependent as well as threshold-related associations. In IV the threshold sBTLA used in mortality analyses was defined as the nearest even concentration above the high- est concentration among controls. Cox regression was performed for day 1- 2 and 7 sBTLA. Mortality for categorized markers were presented with the Kaplan-Meier method. In II, we performed time-varying Cox Regression, after imputation of missing data, to account for the change of sBTLA over time, which may change the risk of the event (mortality) [192], figure 9.

Figure 9. Effect on time-varying risk on mortality. Time-varying Cox regression calculates a weighted average Hazard Ratio (HR) from the different time windows defined by change in exposure. Reprinted with permission from [192].

ANNA LANGE SLPI and soluble BTLA as immunological markers 45

Results A few other immunological mediators and potential biomarkers were studied in papers I and II, but only results regarding SLPI and sBTLA are presented here. I refer to papers I-II regarding detailed results on LL-37, HNP 1-3, BPI, sPD-1, and sCTLA-4.

SLPI

CAP and controls, and in relation to source of infection In paper I, SLPI showed increased concentrations in CAP compared to controls, and just outside the limit of statistical significance compared to non- RTI (p=0.06). In (III) we found that pneumonia was significantly associated with higher SLPI concentrations than other sources of BSI, figure 10.

Figure 10. SLPI in pneumonia and other sources of infection in BSI. SLPI concentrations were significantly higher in pneumonia (n=26) compared to other sources of BSI (n=71), reprinted with permission from paper III.

46 ANNA LANGE SLPI and soluble BTLA as immunological markers

SLPI and bacterial etiology in BSI No significant etiology-related differences in SLPI-concentrations could be seen in CAP in paper I. In paper III, we found significantly different expression of SLPI depending on bacterial etiology in BSI: S. pneumoniae and S. aureus were associated with higher SLPI compared to E. coli on day 1-2, and 3, and S. aureus also later in the course of the study (day 14 and 28), figure 11.

Figure 11. SLPI concentrations from admission to day 28 related to etiology of Bloodstream infection. SLPI was significantly higher in S. aureus compared to E. coli etiology on days 1-2 (p<0.01), 3 (p=0.02), 14 (p=0.03) and 28 (p=0.04), and in S. pneumoniae compared to E.coli on day 1-2 (p<0.01) and 3 (p<0.01). Reprinted with permission from paper III.

ANNA LANGE SLPI and soluble BTLA as immunological markers 47

Severity SLPI was not correlated to PSI or CURB-65 in CAP (paper I), but in BSI (paper III), day 1-2 SLPI was linearly associated with day 1 SOFA score, figure 12. SLPI was significantly increased in sepsis (n=54), compared to non-septic BSI (n=55) on day 1-2. ROC analyses showed anΔ AUC of 0.77 (95 % CI 0.68-0.87) for SLPI on day 1-2 to identify sepsis.

Figure 12. Association of SLPI and SOFA. SLPI on day 1-2 shows linear association with day 1 ΔSOFA, here presented on a categorical scale.

48 ANNA LANGE SLPI and soluble BTLA as immunological markers

Sex-related differences in SLPI expression In study I we found that SLPI concentrations were significantly higher in male (n=45) compared to female (n=43) patients with CAP, which motivated further studies on sex-related differences in SLPI expression in BSI, figure 13. Here, we again found that male patients with pneumonia and BSI, hade higher age-, and severity-adjusted mean SLPI compared to fe- males, a difference not seen in the full BSI cohort.

Figure 13. SLPI in relation to sex and source of infection. In paper I (upper figure), male patients with pneumonia (N= 42) had significantly higher mean SLPI concentrations compared to female (N= 43) (mean 97 (SD 40) vs. 50 (SD 25) ng/mL). In paper II (lower figure) male sex (N = 9) was associated with significantly higher mean SLPI concentrations on day 1-2 compared to female sex (N = 17). Bars represent mean (upper figure), and median (lower figure). Reprinted with permission from papers I and III.

ANNA LANGE SLPI and soluble BTLA as immunological markers 49

Association to other biomarkers Finally, SLPI was studied in relation to established markers of inflammation and markers indicative of immune suppression. In paper I, SLPI was weakly correlated to CRP, but not to white blood cell count. On day 1-2 and day 7 of BSI (III), SLPI was positively correlated to CRP and neutrophil count, and negatively to mHLA-DR, figure 14. There was also a negative correlation to lymphocyte count on day 7.

Figure 14. Correlations between SLPI and mHLA-DR. Significant negative correlations were found on day 1-2 (A), and day 7 (B). Reprinted with permission from paper III.

50 ANNA LANGE SLPI and soluble BTLA as immunological markers sBTLA

Sepsis, ICU controls and healthy controls In paper II, sBTLA showed higher initial concentrations in septic patients compared to patients with other critical illness, and healthy controls, figure 15. AUC for sBTLA to diagnose sepsis in critically ill patients was 0.63 (95% CI 0.492-0.76, p= 0.048)

Figure 15. sBTLA in Sepsis, ICU controls and healthy controls at baseline. Soluble BTLA was significantly higher in sepsis compared to other critical illness and healthy controls. Bars represent the median. Reprinted with permission from paper II.

Half of septic patients in paper II had a hospital-acquired infection, as compared with the community-onset of BSI in IV. In paper II, the most common focus of infection was abdominal (42%), followed by lungs (24%) and urinary tract (UTI) (15%). In paper IV, 27% had a lung or airway infection, 24 % a UTI, whereas abdominal source was less prevalent (7%). We found that sBTLA was associated with age, but only at early assessment-points, and not across all study groups. In paper II, among patients with non-septic critical illness, those >65 years had lower plasma concentration than patients <65 at baseline, but this was not seen in septic patients, and in paper IV sBTLA was negatively associated with age in BSI on day 1-2.

ANNA LANGE SLPI and soluble BTLA as immunological markers 51 sBTLA and severity In paper II, average baseline SOFA score was 10 in septic patients, and sBTLA correlated to disease severity (SOFA), with significantly higher concentrations in septic shock compared to sepsis (Sepsis-3). In paper IV, average SOFA was 2, and we found that sBTLA was increased at higher SOFA scores (≥4) compared to lower (≤3), but was not a good marker to identify sepsis in BSI (AUC 0.54, 95% CI 0.43-0.66). sBTLA and mortality Mortality at 28 days was 18% in the sepsis cohort of paper II, and 4.6% in paper IV. There was no linear association to an increased risk of 28-day mortality for baseline sBTLA in paper II, but high sBTLA, defined as the 67th percentile of sBTLA concentrations among sepsis patients, was associated with a nearly 5-fold increased risk of 28-day mortality relative to low sBTLA (defined as the 33rd percentile) in the adjusted analysis. The time- varying Cox-regression showed a stronger association for sBTLA to early mortality, indicating that that the dynamic expression of sBTLA is an important factor for the association. In paper IV, sBTLA on day 1-2 and day 7 was associated with 90-day and 1-year mortality, and the strongest association was found for the failure to normalize BTLA by day 7, figure 16.

Figure 16. Survival curves for sBTLA. Paper II: Severe sepsis and septic shock (left figure). High sBTLA (≥21 ng/mL) at baseline, was associated with increased risk of 28-day mortality, compared to low concentrations (≤9 ng/mL). Paper IV: Bloostream infection (right figure). Failure to normalize sBTLA by day 7 (red line) was associated with increased risk of 1-year mortality. Reprinted with permission.

52 ANNA LANGE SLPI and soluble BTLA as immunological markers

Association to other biomarkers Baseline sBTLA concentrations were not correlated to CRP and white blood cell count in paper II. In paper IV, biomarkers on all time-points of assessment were included in the analyses, which showed that sBTLA was associated with CRP and decreased lymphocyte count, figure 17. sBTLA was not associated with SLPI (non-published results).

Figure 17. Associations for sBTLA and other markers in BSI (IV). There was a significant overall positive association to CRP, p<0.01 (A) and a significant overall negative association to lymphocyte count, p<0.01 (B). Reprinted with permission.

ANNA LANGE SLPI and soluble BTLA as immunological markers 53

Ethical considerations Ethical approval was achieved for the three studies included in this thesis. Written informed consent was obtained from patients, or, if unable to sign, from a relative. Results are presented on a group level to prevent breaches of privacy. To allow for biomarker analysis at baseline, the “Dynamics of sepsis” study (study population in papers III and IV) had ethical approval to retrieve an initial blood sample on hospital admission, without informed consent. This sample was drawn in tandem with venipuncture for blood culture, therefore not inflicting extra discomfort on the patient. If the patient was not enrolled in the study, the sample was saved without other identification information than age and bacterial agent.

54 ANNA LANGE SLPI and soluble BTLA as immunological markers

Discussion The major findings in this thesis are divergent expression of SLPI related to etiology of infection (source and pathogen), and that sBTLA is associated with mortality. Additionally, both SLPI and sBTLA showed association to severity, and markers indicative of immune suppression in sepsis. Notably sBTLA and SLPI were not associated. We found that plasma SLPI was higher in pneumonia than in non-infected controls, and compared to other sources of infection. This is in line with two previous studies [167, 193]. Contrary to our findings in CAP, one study found that bacteremia increased SLPI significantly, and that SLPI concentrations correlated to the extent of lung tissue involved [167]. Pneumo- nia is the most common site of infection in ICU-treated sepsis and is associated with worse outcome, which highlights the importance of additional clarification of the underlying immunopathology [31, 194]. Few studies have compared immunological markers in different septic etiology, however a recent study of ICU-treated sepsis, showed that pneumonia was associated with failure to recover mHLA-DR expression compared to UTI and abdominal focus [194]. In paper I, we found that BPI was statistically significantly increased in CAP compared to HC, while LL-37 and HNP 1-3 were not. Both BPI and -defensins have been shown to be increased in sepsis, and BPI correlates with severity [195, 196]. Results on LL-37 in sepsis are divergent, both no differenceα in serum levels compared to HC, and an association with severity, have been reported [195, 197]. In line with our findings in paper I, another study found that LL-37 was not statistically significantly increased in pneumonia compared to HC [198]. Interestingly, we found that men with pneumonia had higher SLPI concentrations compared to women, in both CAP and BSI. The significance of this is uncertain. Male sex is not a clear risk factor for short-term mortality in pneumonia [199]. In addition to the source-related difference, we found that S. aureus and S. pneumoniae BSI were associated with higher mean SLPI compared to E. coli etiology, at early time-points for S. pneumoniae, and at both early and late time-points for S. aureus. S. pneumonia, S. aureus, and E. coli etiology were shown to have similar odds-ratios for mortality, in ICU-treated sepsis [31]. Apart from epidemiological data, there are sparse reports on differences related to individual pathogens in sepsis. Instead, etiology is often di- vided into gram-negative and gram-positive. While some virulence factors

ANNA LANGE SLPI and soluble BTLA as immunological markers 55 are common, e.g. LPS in gram-negatives, others differ depending on the specific organism, e.g. superantigen production by S. aureus and S. pyogenes [200]. Moreover, host-recognition pathways differ as described in the introduction. Experimentally, the most widely used sepsis-model is cecal ligation and puncture, which induces a polymicrobial abdominal sepsis, but clinically, gram-positive organisms account for almost half of positive cultures is sepsis [31, 201]. This may partly account for disparate results in experimental and clinical sepsis studies. Studies that investigate host responses in relation to sepsis etiology are hence motivated. SLPI concentrations were linearly associated with SOFA score, and higher in sepsis compared to non-septic BSI. A small study of surgical ICU patients showed that SLPI was correlated to multi organ dysfunction score [150]. Blood levels of SLPI were also recently shown to be elevated in acute kidney failure, and in acetaminophen-induced acute liver failure [168, 169]. One theory is that circulating SLPI originates from spill-over from inflamed lung epithelium in pneumonia, however, SLPI is secreted by many epithelial cells in inflammation, and is present in the secondary granules of neutrophils [147, 150, 167]. In acetaminophen-induced acute liver failure, sub- analysis of patients who underwent liver transplantation, showed higher SLPI levels in the hepatic vein compared to the portal vein, and increased SLPI expression in biliary epithelial cells, and macrophages of necrotic liver areas [169]. An extrapolation to these findings, is that circulating SLPI might reflect both the amount of epithelial affection (i.e. high in pneumonia), and the extent of organ failure, indicated by SOFA score. Possibly, S. pneumoniae and S. aureus BSI is associated with more epithelial affection than E. coli in the context of UTI, the dominant source of E. coli bacteremia in our study. Studies in another context, with other sources of E. coli bacteremia, e.g. abdominal infections, might produce different results. Plasma concentrations of the previously non-studied protein sBTLA were higher in severe sepsis and septic shock, compared to other critical illness, and controls. These findings were repeated in another study of critically ill patients, in which sBTLA was shown to have even better predictability of sepsis at baseline (AUC 0.76, compared to 0.63 in our study) [185]. In line with these results, we also found increased sBTLA in patients with BSI, and here sBTLA remained elevated compared to controls after 4 weeks. We found that sPD-1 was decreased in severe sepsis and septic shock compared to controls, and was not associated with mortality, while sCTLA- 4 was low or undetectable. Disparate findings regarding sPD-1 were recently presented, where sPD-1 was shown to correlate to sepsis severity, and

56 ANNA LANGE SLPI and soluble BTLA as immunological markers to have high predictive capacity for early mortality in ICU-treated patients [202]. Differences in methodology and patient selection might account for the divergent results. Soluble CTLA-4 has been shown to be increased in several autoimmune diseases, and to be a favorable marker in advanced melanoma [203, 204]. As a severity marker, sBTLA was weighted towards more severe disease. It was significantly elevated in septic shock compared to sepsis (Sepsis-3 definitions), but could not discriminate sepsis from non-septic BSI. There are no studies for reference regarding soluble BTLA in sepsis, but one study on membrane-expression of BTLA on CD4+ cells showed a gradually decreasing percentage of cells expressing BTLA, with increasing sepsis severity [182]. Two other studies on BTLA expression on CD4+ isolated from septic patients showed no difference, and increased expression respectively [96, 181]. It was shown, that in mice, sBTLA originates from an alternative mRNA splice product that is produced to a greater degree in critical illness [185]. This could account for the deviant findings regarding soluble and membrane-associated BTLA in sepsis. In BSI patients, we found that sBTLA was associated with decreased lymphocyte count, and that SLPI correlated to reduced mHLA-DR, on day 1-2 and day 7, which suggests that sBTLA and SLPI may be linked to separate aspects of immune dysfunction. Lymphocyte count and mHLA-DR are commonly used surrogate markers for immune suppression in sepsis trials, but are not used in clinical practice [107]. Lowered circulating lymphocyte count is a typical finding in the early sepsis phase, and is inferior to both CRP and procalcitonin in detecting sepsis in a critical care setting [205]. Persistent lymphopenia was however shown to be associated with increased 28-day mortality in bacteremic sepsis [129]. Monocytic HLA-DR expression is a well-studied marker of sepsis-associated immunosuppression. The trajectory on mHLA-DR in sepsis associates with secondary infections and mortality, and increased mHLA-DR over time indicates recovery [194]. sBTLA was associated with mortality in both ICU-treated septic patients, and in BSI. In the former group we found that high baseline sBTLA concentrations were associated with an almost five times increased risk of 28-day mortality, compared to low concentrations. We found a time-varying association, indicating that the change of sBTLA over the week-long study period was of importance. In BSI, we found associations to mortality at 90 days and 1 year, for sBTLA measured already at baseline. In line with the time-varying association seen in paper II, we however saw the strongest association to mortality in patients who failed to normalize sBTLA on day 7.

ANNA LANGE SLPI and soluble BTLA as immunological markers 57

The applicability of soluble co-regulatory proteins as prognostic biomarkers in cancer has been addressed in three recent studies, out of which two high- lighted the significance of high sBTLA [206-208]. The cause of delayed deaths after sepsis is multifactorial. Age and pre- existing comorbid conditions, may contribute, but the septic episode itself accounts for a significant increased mortality up to two years afterwards [14]. Reasons to this may be persisting organ dysfunction, and unresolved immune responses. In addition to mortality, sepsis is associated with an increased risk of functional impairment [209]. Two recent studies followed sepsis survivors up to one year, and found signs of persistent immune dis- turbances [210, 211]. One limitation of this thesis is the missed samples over time. Reasons were; missed sampling, death, transfer to other ICU’s and medical/surgical wards (paper II), with a risk of missing out both the ones who were most ill, and those who recovered quickly. In papers III and IV, there was a larger proportion of missed samples, due to the working schedules of study nurses, and failure of patients to attend follow-up visits. This could cause selection bias by the mechanisms described above. Missing data analyses showed that patients who provided fewer samples in papers III and IV, were older, and we tried to account for this by adjusting for age in between-group comparisons. Another limitation is the relatively small sizes of the study groups, and the unmatched control group of studies II-IV. Many sepsis studies focus on ICU-treated patients, but we here show a consistency in findings across study groups of varying degrees of severity. Another strength is that our results regarding sBTLA in critical illness were reproduced in another study. In support of the validity of our findings, are also the succeeding reports that sBTLA is associated with mortality in cancer. This thesis suggests sBTLA as a possible prognostic tool in sepsis. Asso- ciations of SLPI with decreased mHLA-DR, and sBTLA with decreased lymphocyte count, suggests that these proteins may be involved in sepsis-associated immune suppression. The new immune-boosting adjuvant treatments in sepsis come at the price of immune-related side effects, which underscores the necessity to find reliable immunological markers that can identify suitable subjects. I propose that sBTLA and SLPI should be further evaluated in the context of immune profiling, and that the pathobiological role of the candidate biomarker sBTLA, and its utility as a prognostic marker, war- rants further investigation.

58 ANNA LANGE SLPI and soluble BTLA as immunological markers

Conclusions

• SLPI concentrations were increased in CAP, but did not correlate to severity, or to bacterial etiology. An interesting finding was that SLPI concentrations were higher in men compared to women with CAP.

• SLPI was increased in pneumonia compared to other sources of infection in BSI, and was increased in S. pneumoniae, and S. aureus BSI compared to E. coli BSI. SLPI correlated to disease severity, to neutrophil count and CRP, and to decreased mHLA-DR expression. Again, we found that SLPI was higher in men than in women with pneumonia.

• Soluble BTLA was increased in sepsis, compared to other critical illness, and was associated with severity. Soluble BTLA showed a time-varying association to mortality, but high sBTLA measured already at baseline was associated with increased risk of death before 28 days in patients with sepsis.

• In BSI, sBTLA remained elevated compared to controls four weeks after hospital admission. Soluble BTLA was increased in more severe disease, and was associated with late mortality (90 days and 1 year). The association with mortality was particularly notable for sustained sBTLA concentrations above normal on day 7. Finally, sBTLA was associated with CRP and decreased lymphocyte count. Soluble BTLA was not associated with bacterial etiology.

ANNA LANGE SLPI and soluble BTLA as immunological markers 59

Future perspectives There are major gaps in the knowledge of the timeline of the immunological events in sepsis. There is compelling evidence that immune suppression contribute to the pathobiology of sepsis, but it remains to be clarified whether it is attributable to sepsis itself, or to factors present prior to the septic episode. The best way to answer the first question, would be via a prospective cohort study that follows patients with increased risk to develop sepsis, for example persons over a certain age, with repeated sampling, to follow immunological markers over several years. This would be very resource intensive, and require multiple participating centers. A middle way to address the chicken-or-the-egg enigma could be to focus on a specific patient group at risk for infective complications, e.g. patients who are subject to elective surgery, e.g. arthroplasty of the hip or knee. A blood sample taken at pre-operative assessment could be followed by sampling directly after surgery, before discharge from hospital, and at follow- up visits to nurse, physiotherapy or physician. Immunological markers, e.g. sBTLA, could then be studied in relation to subsequent infectious events. A feasible approach to study the dynamics of sBTLA in relation to the clinical trajectory, is of course to start at the event, i.e. to follow patients admitted to the ICU due to sepsis and other critical conditions, during the ICU and hospital stay, and afterwards. A suitable schedule for blood samples would be day 0, 4, and 7 of hospital stay, and follow up at 1, 3, 6 and 12 months. In addition to sBTLA, it would be interesting to study other soluble co- regulatory proteins that might be involved in the sepsis-related immunopathology. There are multiplex immunoassay systems that allow simultaneous assessment of several soluble proteins, which minimizes the amount of sample needed, and requires little manpower. The quest to find adjuvant therapies that boost the immune system has just started, but to continue in the right direction, it is crucial to increase knowledge on the interconnect- edness of the immunological markers we measure. A major question for sBTLA is how it participate in the complex HVEM signaling network, and if it contributes to septic morbidity, or is merely a marker of disturbed homeostasis. This needs studies at a cellular and molecular level that I do not have sufficient knowledge to discuss, but hopefully the findings regarding sBTLA that we, and others have made in sepsis and cancer, will motivate such mechanistic studies.

60 ANNA LANGE SLPI and soluble BTLA as immunological markers

Serine proteases inhibitors, including SLPI, have implications in many diseases, e.g. cancer, autoimmune disorder, inflammatory and infectious lung disease, and viral infection. The last role is a topic of particular interest, in light of the ongoing COVID-19 pandemic. Cellular entry of corona- viruses is dependent on priming of a viral surface (Spike) protein by host cell proteases, and for Sars-Cov-2 it was just recently shown to be mediated by Cathepsin B and L, and the protease transmembrane protease serine 2 (TMPRSS2) [212]. That study showed that Sars-Cov-2 entry could be blocked by addition of an inhibitor of TMPRSS2, i.e. a protease inhibitor. It was previously shown that SLPI might be involved in protection against influenza A virus infection via inhibition of TMPRSS2, and that disrupted protease-antiprotease balance could increase infectivity [213]. It would be highly interesting to investigate how SLPI participates in Sars-Cov-2 protection, and if disturbed SLPI-secretion is associated with more severe COVID- 19 disease.

ANNA LANGE SLPI and soluble BTLA as immunological markers 61

Sammanfattning på svenska Svårighetsgrad och prognos vid infektioner beror på en samverkan mellan värdfaktorer (ålder och grundsjukdomar), infektionens lokalisation och orsakande organism. I ett optimalt immunsvar begränsas infektionen till det infekterade organet och frisk vävnad skyddas från skadliga effekter av inflammation. Vid sepsis ser man en spridd inflammationsreaktion, med kon- sekvenser i form av organpåverkan och ibland blodtrycksfall med chock. Samtidig aktivering av reglerande immunologiska mekanismer kan, om de blir för starka, medföra en försvagning av immunförsvaret som gör det svårt att läka ut den aktuella infektionen och öka risken för nya, så kallat oppor- tunistiska infektioner. Den kortsiktiga överlevnaden vid sepsis har förbätt- rats, men senare års forskning har visat på ökad sjuklighet och dödlighet på sikt och kvarstående störningar i immunförsvaret misstänks bidra till detta. Syftet med denna avhandling var att studera hur de lösliga proteinerna Secretory Leukocyte protease Inhibitor (SLPI) och B and T Lymphocyte At- tenuator (sBTLA) uttrycks i blodcirkulationen vid svåra infektioner och sepsis. SLPI utsöndras av celler i kroppens slemhinnor och bildas också inuti vita blodkroppar. Som namnet anger, är SLPI en hämmare av proteaser (proteinnedbrytande ämnen) som frisätts av vita blodkroppar vid inflammation. SLPI har också en bakteriedödande effekt och är intressant ur ett immunreglerande perspektiv, då det motverkar bildningen av ämnen som driver på inflammationsreaktionen vid infektioner. BTLA uttrycks på immuncellernas yta och är närbesläktad med de så kallade checkpoint-re- ceptorerna Programmed death-1 (PD-1) och Cytotoxic T lymphocyte-associated Antigen 4 (CTLA-4), som har fått stor betydelse som angreppspunk- ter vid cancer-immunterapi. Checkpoint-receptorer hämmar lymfocyternas aktiveringsgrad och tros också vara inblandade i den immunförsvagning som kan ske vid sepsis. De lösliga formerna av PD-1 och CTLA-4 har dock inte tidigare studerats vid sepsis och sBTLA varken vid sepsis, eller andra tillstånd. I avhandlingsarbetet har jag studerat plasmakoncentrationen av SLPI- och sBTLA i förhållande till infektionslokal, orsakande bakterie, svårighets- grad mätt som organsvikt och mortalitet (dödlighet), samt relationen till andra immunmarkörer. I avhandlingens delarbete I och III visades att SLPI-koncentrationen var högre vid lunginflammation (pneumoni), än vid andra infektionslokaler och att pneumokock- och Staphylococcus aureus -bakteriemi (växt i blodet) var förknippade med högre SLPI-nivåer än Escherichia Coli -bakteriemi. Vidare

62 ANNA LANGE SLPI and soluble BTLA as immunological markers sågs högre SLPI-nivåer vid sepsis än vid bakteriemi utan sepsis och att män med pneumoni, med eller utan bakteriemi, hade högre nivåer av SLPI än kvinnor. I delarbete II och IV visades att sBTLA-koncentrationen var associerad till graden av organsvikt och var högre vid sepsis, än vid icke-infektiös kri- tisk sjukdom. Höga sBTLA-nivåer var kopplade till ökad risk för tidig (28 dagar) död hos intensivvårdade sepsispatienter och till sena dödsfall (90 dagar och 1 år) hos patienter med bakteriemi, särskilt hos de som inte hade normaliserat sina sBTLA-nivåer dag 7 efter inläggning på sjukhus. Slutligen fann vi att både SLPI och sBTLA samvarierade med markörer tydande på försvagning av immunförsvaret, SLPI i förhållande till en anti- genpresenterande molekyl på monocyter, HLA-DR och sBTLA i förhål- lande till lymfocytnivån. Sammanfattningsvis visar denna avhandling en association mellan SLPI respektive sBTLA och svårighetsgrad vid sepsis och bakteriemi, samt en intressant koppling till kända indikatorer för ett försvagat immunförsvar vid sepsis. SLPI-uttrycket skiljer sig åt beroende på infektionslokal och orsakande bakterie, medan sBTLA är associerat till ökad risk för död. Forskning kring immunterapi vid försvagat immunförsvar vid sepsis pågår, men verk- tygen att identifiera lämpliga kandidater är trubbiga. Fynden i denna avhandling motiverar närmare studier avseende om SLPI och sBTLA skulle kunna vara användbara i detta sammanhang.

ANNA LANGE SLPI and soluble BTLA as immunological markers 63

Acknowledgements I would like to direct my sincerest gratitude to my supervisor Olof Hultgren, for encouraging my interest in immunology and for bearing with me through this process, being carefully enthusiastic about positive results, and encouraging me in moments of doubt. A special thank you also to my co-supervisor Jonas Sundén-Cullberg for coming up with the idea to look at soluble co-signaling molecules, for providing the ICU-material for study II, and for sharing your time and knowledge on sepsis. I would also like to thank: Sara Cajander, co-author of papers III and IV, for encouragement, inspi- ration, and invaluable support, and for being a dear colleague and friend. Kristoffer Strålin, for providing ideas and material to paper I, for design- ing the Dynamics of sepsis study, and for valuable participation in analysis and drafting of papers I, II and IV. Anders Magnuson, co-author of papers II-IV, for shedding light on regression, imputation and much more. Ulla Andersson, and the staff at the Department of Clinical Research La- boratory at Örebro University for laboratory assistance. Lars-Göran Jansson, for assistance with figures and tables. Past and present bosses: Henrik Eliasson, Martin Widlund, and Inger Nordin-Olsson, for giving me the opportunity to do research and travel. My other colleagues at the departments of Infectious Diseases, and Com- municable Disease Control, who have made it possible for me to write this thesis, and who do a fantastic job, always, and especially right now during the COVID-19 pandemic. You are amazing! Special thanks to Lotta Lybeck, Anja Rosdahl, Ingrid Ziegler and Simon Athlin, for positive attitude, and for sharing moments of gossip, and frustration, but also fruitful discussions. To Gunlög Rasmussen for being a dear friend and colleague, and for giving me the space and time to write this thesis. Åsa Athlin for providing eggs, bubbles, one-liners, and alternative definitions to Cox regression. Karolina Evenhamre for walks, talks, runs, and tips on good things in life. Other friends, you know who you are! My parents for always believing in me. Last, and most of all, I would like to thank Johan for your love and pa- tience, and my wonderful kids, Clara, Ivar and Olle, for always reminding me of what is most important in life.

64 ANNA LANGE SLPI and soluble BTLA as immunological markers

References

1. Funk DJ, Parrillo JE, Kumar A. Sepsis and septic shock: a history. Crit Care Clin. 2009;25(1):83-101, viii. 2. Henle J. Von den miasmen und contagien und von der miasmatisch- contaginösen krankheiten. Pathologische Untersuchungen. Berlin1840. p. 1-82. 3. Gaynes R. The Discovery of Penicillin—New Insights After More Than 75 Years of Clinical Use. Emerg Infect Dis. 2017;23(5):849- 53. 4. Beutler B, Rietschel ET. Innate immune sensing and its roots: the story of endotoxin. Nature reviews. Immunology. 2003;3(2):169- 76. 5. Merle NS, Noe R, Halbwachs-Mecarelli L, Fremeaux-Bacchi V, Roumenina LT. Complement System Part II: Role in Immunity. Front Immunol. 2015;6:257. 6. Manasse P. Ueber hyaline Ballen und Thromben in den Gehirngefässen bei acuten Infectionskrankheiten. Archiv für pathologische Anatomie und Physiologie und für klinische Medicin. 1892;130(2):217-33. 7. Steinman RM, Cohn ZA. Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation, tissue distribution. The Journal of experimental medicine. 1973;137(5):1142-62. 8. Allison JP, Lanier LL. Structure, function, and serology of the T- cell antigen receptor complex. Annu Rev Immunol. 1987;5:503-40. 9. Harding FA, McArthur JG, Gross JA, Raulet DH, Allison JP. CD28-mediated signalling co-stimulates murine T cells and prevents induction of anergy in T-cell clones. Nature. 1992;356(6370):607-9. 10. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA. The dorsoventral regulatory gene cassette spätzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell. 1996;86(6):973-83. 11. Medzhitov R, Preston-Hurlburt P, Janeway CA, Jr. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature. 1997;388(6640):394-7. 12. Poltorak A, He X, Smirnova I, Liu MY, Van Huffel C, Du X, et al. Defective LPS signaling in C3H/HeJ and C57BL/10ScCr mice: mutations in Tlr4 gene. Science (New York, N.Y.). 1998;282(5396):2085-8.

ANNA LANGE SLPI and soluble BTLA as immunological markers 65

13. Marshall JC. Why have clinical trials in sepsis failed? Trends Mol Med. 2014;20(4):195-203. 14. Prescott HC, Osterholzer JJ, Langa KM, Angus DC, Iwashyna TJ. Late mortality after sepsis: propensity matched cohort study. BMJ. 2016;353:i2375. 15. van der Poll T, van de Veerdonk FL, Scicluna BP, Netea MG. The immunopathology of sepsis and potential therapeutic targets. Nat Rev Immunol. 2017;17(7):407-20. 16. Yende S, Austin S, Rhodes A, Finfer S, Opal S, Thompson T, et al. Long-Term Quality of Life Among Survivors of Severe Sepsis: Analyses of Two International Trials. Crit Care Med. 2016;44(8):1461-7. 17. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, et al. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med. 2006;34(2):344-53. 18. Esper AM, Moss M, Lewis CA, Nisbet R, Mannino DM, Martin GS. The role of infection and comorbidity: Factors that influence disparities in sepsis. Crit Care Med. 2006;34(10):2576-82. 19. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med. 2003;348(16):1546-54. 20. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med. 2013;369(9):840-51. 21. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801-10. 22. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest. 1992;101(6):1644-55. 23. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med. 2003;31(4):1250-6. 24. Vincent J-L, Opal SM, Marshall JC, Tracey KJ. Sepsis definitions: time for change. Lancet (London, England). 2013;381(9868):774- 5. 25. Kaukonen K-M, Bailey M, Pilcher D, Cooper DJ, Bellomo R. Systemic inflammatory response syndrome criteria in defining severe sepsis. The New England journal of medicine. 2015;372(17):1629-38.

66 ANNA LANGE SLPI and soluble BTLA as immunological markers

26. Shankar-Hari M, Harrison DA, Rubenfeld GD, Rowan K. Epidemiology of sepsis and septic shock in critical care units: comparison between sepsis-2 and sepsis-3 populations using a national critical care database. BJA: British Journal of Anaesthesia. 2017;119(4):626-36. 27. Raith EP, Udy AA, Bailey M, McGloughlin S, MacIsaac C, Bellomo R, et al. Prognostic Accuracy of the SOFA Score, SIRS Criteria, and qSOFA Score for In-Hospital Mortality Among Adults With Suspected Infection Admitted to the Intensive Care Unit. JAMA. 2017;317(3):290-300. 28. Haydar S, Spanier M, Weems P, Wood S, Strout T. Comparison of QSOFA score and SIRS criteria as screening mechanisms for emergency department sepsis. The American journal of emergency medicine. 2017;35(11):1730-3. 29. Guo Y, Patil NK, Luan L, Bohannon JK, Sherwood ER. The biology of natural killer cells during sepsis. Immunology. 2018;153(2):190-202. 30. Fleischmann C, Scherag A, Adhikari NKJ, Hartog CS, Tsaganos T, Schlattmann P, et al. Assessment of Global Incidence and Mortality of Hospital-treated Sepsis. Current Estimates and Limitations. Am J Respir Crit Care Med. 2016;193(3):259-72. 31. Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Virulence. 2014;5(1):4-11. 32. Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA. 2014;311(13):1308-16. 33. Zimmerman JE, Kramer AA, Knaus WA. Changes in hospital mortality for United States intensive care unit admissions from 1988 to 2012. Crit Care. 2013;17(2):R81-R. 34. Stevenson EK, Rubenstein AR, Radin GT, Wiener RS, Walkey AJ. Two decades of mortality trends among patients with severe sepsis: a comparative meta-analysis*. Crit Care Med. 2014;42(3):625-31. 35. Venkatesh B, Finfer S, Cohen J, Rajbhandari D, Arabi Y, Bellomo R, et al. Adjunctive Glucocorticoid Therapy in Patients with Septic Shock. N Engl J Med. 2018;378(9):797-808. 36. Dellinger RP, Bagshaw SM, Antonelli M, Foster DM, Klein DJ, Marshall JC, et al. Effect of Targeted Polymyxin B Hemoperfusion on 28-Day Mortality in Patients With Septic Shock and Elevated Endotoxin Level: The EUPHRATES Randomized Clinical Trial. JAMA. 2018;320(14):1455-63.

ANNA LANGE SLPI and soluble BTLA as immunological markers 67

37. Annane D, Renault A, Brun-Buisson C, Megarbane B, Quenot JP, Siami S, et al. Hydrocortisone plus Fludrocortisone for Adults with Septic Shock. N Engl J Med. 2018;378(9):809-18. 38. Inoue S, Suzuki-Utsunomiya K, Okada Y, Taira T, Iida Y, Miura N, et al. Reduction of immunocompetent T cells followed by prolonged lymphopenia in severe sepsis in the elderly. Crit Care Med. 2013;41(3):810-9. 39. Reinhart K, Daniels R, Kissoon N, Machado FR, Schachter RD, Finfer S. Recognizing Sepsis as a Global Health Priority - A WHO Resolution. The New England journal of medicine. 2017;377(5):414-7. 40. Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368-77. 41. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med. 2006;34(6):1589-96. 42. Angus DC, Barnato AE, Bell D, Bellomo R, Chong CR, Coats TJ, et al. A systematic review and meta-analysis of early goal-directed therapy for septic shock: the ARISE, ProCESS and ProMISe Investigators. Intensive Care Med. 2015;41(9):1549-60. 43. Dellinger RP, Carlet JM, Masur H, Gerlach H, Calandra T, Cohen J, et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Intensive Care Med. 2004;30(4):536-55. 44. Rhodes A, Evans LE, Alhazzani W, Levy MM, Antonelli M, Ferrer R, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive Care Med. 2017;43(3):304-77. 45. Leiva-Juárez MM, Kolls JK, Evans SE. Lung epithelial cells: therapeutically inducible effectors of antimicrobial defense. Mucosal Immunol. 2018;11(1):21-34. 46. Mogensen TH. Pathogen recognition and inflammatory signaling in innate immune defenses. Clin Microbiol Rev. 2009;22(2):240- 73, Table of Contents. 47. Gong T, Liu L, Jiang W, Zhou R. DAMP-sensing receptors in sterile inflammation and inflammatory diseases. Nature reviews. Immunology. 2019:10.1038/s41577-019-0215-7. 48. Vénéreau E, Ceriotti C, Bianchi ME. DAMPs from Cell Death to New Life. Front Immunol. 2015;6:422. 49. Picard C, von Bernuth H, Ghandil P, Chrabieh M, Levy O, Arkwright PD, et al. Clinical features and outcome of patients with

68 ANNA LANGE SLPI and soluble BTLA as immunological markers

IRAK-4 and MyD88 deficiency. Medicine (Baltimore). 2010;89(6):403-25. 50. Davies LC, Jenkins SJ, Allen JE, Taylor PR. Tissue-resident macrophages. Nat Immunol. 2013;14(10):986-95. 51. Borges da Silva H, Fonseca R, Pereira RM, Cassado Ados A, Álvarez JM, D'Império Lima MR. Splenic Macrophage Subsets and Their Function during Blood-Borne Infections. Front Immunol. 2015;6:480. 52. Németh T, Sperandio M, Mócsai A. Neutrophils as emerging therapeutic targets. Nat Rev Drug Discov. 2020. 53. Cowland JB, Borregaard N. Granulopoiesis and granules of human neutrophils. Immunol Rev. 2016;273(1):11-28. 54. Kolaczkowska E, Kubes P. Neutrophil recruitment and function in health and inflammation. Nat Rev Immunol. 2013;13(3):159-75. 55. Brinkmann V. Neutrophil Extracellular Traps in the Second Decade. J Innate Immun. 2018;10(5-6):414-21. 56. Rosales CaU-Q, Eileen. Neutrophils 2018 November 5th. 57. Favier B. Regulation of neutrophil functions through inhibitory receptors: an emerging paradigm in health and disease. Immunol Rev. 2016;273(1):140-55. 58. Kaech SM, Cui W. Transcriptional control of effector and memory CD8+ T cell differentiation. Nat Rev Immunol. 2012;12(11):749- 61. 59. Paolini R, Bernardini G, Molfetta R, Santoni A. NK cells and interferons. Cytokine Growth Factor Rev. 2015;26(2):113-20. 60. Kärre K. Natural killer cell recognition of missing self. Nat Immunol. 2008;9(5):477-80. 61. Kany S, Vollrath JT, Relja B. Cytokines in Inflammatory Disease. Int J Mol Sci. 2019;20(23). 62. Karasu E, Nilsson B, Köhl J, Lambris JD, Huber-Lang M. Targeting Complement Pathways in Polytrauma- and Sepsis-Induced Multiple-Organ Dysfunction. Front Immunol. 2019;10:543. 63. Steiner H, Hultmark D, Engström A, Bennich H, Boman HG. Sequence and specificity of two antibacterial proteins involved in insect immunity. Nature 292: 246-248. 1981. Journal of immunology (Baltimore, Md. : 1950). 2009;182(11):6635-7. 64. The antimicrobial peptide Database [updated Feb 24 2020. Available from: http://aps.unmc.edu/AP/. 65. Wang G. Human antimicrobial peptides and proteins. Pharmaceuticals (Basel). 2014;7(5):545-94. 66. Kumar P, Kizhakkedathu JN, Straus SK. Antimicrobial Peptides: Diversity, Mechanism of Action and Strategies to Improve the Activity and Biocompatibility In Vivo. Biomolecules. 2018;8(1):4.

ANNA LANGE SLPI and soluble BTLA as immunological markers 69

67. Brogden KA. Antimicrobial peptides: pore formers or metabolic inhibitors in bacteria? Nature reviews. Microbiology. 2005;3(3):238-50. 68. Hancock RE, Haney EF, Gill EE. The immunology of host defence peptides: beyond antimicrobial activity. Nat Rev Immunol. 2016;16(5):321-34. 69. Hemshekhar M, Anaparti V, Mookherjee N. Functions of Cationic Host Defense Peptides in Immunity. Pharmaceuticals (Basel). 2016;9(3). 70. Raheem N, Straus SK. Mechanisms of Action for Antimicrobial Peptides With Antibacterial and Antibiofilm Functions. Frontiers in microbiology. 2019;10:2866-. 71. Sierra JM, Fusté E, Rabanal F, Vinuesa T, Viñas M. An overview of antimicrobial peptides and the latest advances in their development. Expert Opin Biol Ther. 2017;17(6):663-76. 72. Grégoire N, Aranzana-Climent V, Magréault S, Marchand S, Couet W. Clinical Pharmacokinetics and Pharmacodynamics of Colistin. Clin Pharmacokinet. 2017;56(12):1441-60. 73. Grönberg A, Mahlapuu M, Ståhle M, Whately-Smith C, Rollman O. Treatment with LL-37 is safe and effective in enhancing healing of hard-to-heal venous leg ulcers: a randomized, placebo-controlled clinical trial. Wound repair and regeneration : official publication of the Wound Healing Society [and] the European Tissue Repair Society. 2014;22(5):613-21. 74. Bedoui S, Heath WR, Mueller SN. CD4(+) T-cell help amplifies innate signals for primary CD8(+) T-cell immunity. Immunol Rev. 2016;272(1):52-64. 75. Worbs T, Hammerschmidt SI, Förster R. Dendritic cell migration in health and disease. Nature reviews. Immunology. 2017;17(1):30-48. 76. Kondo K, Ohigashi I, Takahama Y. Thymus machinery for T-cell selection. Int Immunol. 2019;31(3):119-25. 77. van den Broek T, Borghans JAM, van Wijk F. The full spectrum of human naive T cells. Nature reviews. Immunology. 2018;18(6):363-73. 78. Wan YY, Flavell RA. How diverse--CD4 effector T cells and their functions. J Mol Cell Biol. 2009;1(1):20-36. 79. Fraser JD, Proft T. The bacterial superantigen and superantigen- like proteins. Immunol Rev. 2008;225:226-43. 80. Mazzoni A, Maggi L, Liotta F, Cosmi L, Annunziato F. Biological and clinical significance of T helper 17 cell plasticity. Immunology. 2019;158(4):287-95.

70 ANNA LANGE SLPI and soluble BTLA as immunological markers

81. Crotty S. T follicular helper cell differentiation, function, and roles in disease. Immunity. 2014;41(4):529-42. 82. Nair-Gupta P, Blander JM. An updated view of the intracellular mechanisms regulating cross-presentation. Front Immunol. 2013;4:401. 83. Voskoboinik I, Whisstock JC, Trapani JA. Perforin and granzymes: function, dysfunction and human pathology. Nat Rev Immunol. 2015;15(6):388-400. 84. Bronte V, Pittet MJ. The spleen in local and systemic regulation of immunity. Immunity. 2013;39(5):806-18. 85. Sinwar PD. Overwhelming post splenectomy infection syndrome - review study. Int J Surg. 2014;12(12):1314-6. 86. Hoffman W, Lakkis FG, Chalasani G. B Cells, Antibodies, and More. Clin J Am Soc Nephrol. 2016;11(1):137-54. 87. De Sousa Linhares A, Leitner J, Grabmeier-Pfistershammer K, Steinberger P. Not All Immune Checkpoints Are Created Equal. Front Immunol. 2018;9:1909. 88. Walunas TL, Lenschow DJ, Bakker CY, Linsley PS, Freeman GJ, Green JM, et al. CTLA-4 can function as a negative regulator of T cell activation. Immunity. 1994;1(5):405-13. 89. Coyle AJ, Gutierrez-Ramos JC. The expanding B7 superfamily: increasing complexity in costimulatory signals regulating T cell function. Nat Immunol. 2001;2(3):203-9. 90. Keir ME, Butte MJ, Freeman GJ, Sharpe AH. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol. 2008;26:677-704. 91. Buchbinder EI, Desai A. CTLA-4 and PD-1 Pathways: Similarities, Differences, and Implications of Their Inhibition. Am J Clin Oncol. 2016;39(1):98-106. 92. Waterhouse P, Penninger JM, Timms E, Wakeham A, Shahinian A, Lee KP, et al. Lymphoproliferative disorders with early lethality in mice deficient in Ctla-4. Science. 1995;270(5238):985-8. 93. Nishimura H, Nose M, Hiai H, Minato N, Honjo T. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity. 1999;11(2):141-51. 94. Guignant C, Lepape A, Huang X, Kherouf H, Denis L, Poitevin F, et al. Programmed death-1 levels correlate with increased mortality, nosocomial infection and immune dysfunctions in septic shock patients. Crit Care. 2011;15(2):R99. 95. Zhang Y, Li J, Lou J, Zhou Y, Bo L, Zhu J, et al. Upregulation of programmed death-1 on T cells and programmed death ligand-1 on monocytes in septic shock patients. Crit Care. 2011;15(1):R70.

ANNA LANGE SLPI and soluble BTLA as immunological markers 71

96. Boomer JS, Shuherk-Shaffer J, Hotchkiss RS, Green JM. A prospective analysis of lymphocyte phenotype and function over the course of acute sepsis. Crit Care. 2012;16(3):R112. 97. Delano MJ, Ward PA. The immune system's role in sepsis progression, resolution, and long-term outcome. Immunol Rev. 2016;274(1):330-53. 98. Volk HD, Thieme M, Heym S, Döcke WD, Ruppe U, Tausch W, et al. Alterations in function and phenotype of monocytes from patients with septic disease--predictive value and new therapeutic strategies. Behring Inst Mitt. 1991(88):208-15. 99. Bone RC. Sir Isaac Newton, sepsis, SIRS, and CARS. Crit Care Med. 1996;24(7):1125-8. 100. Otto GP, Sossdorf M, Claus RA, Rödel J, Menge K, Reinhart K, et al. The late phase of sepsis is characterized by an increased microbiological burden and death rate. Crit Care. 2011;15(4):R183. 101. Torgersen C, Moser P, Luckner G, Mayr V, Jochberger S, Hasibeder WR, et al. Macroscopic postmortem findings in 235 surgical intensive care patients with sepsis. Anesth Analg. 2009;108(6):1841-7. 102. Boomer JS, To K, Chang KC, Takasu O, Osborne DF, Walton AH, et al. Immunosuppression in patients who die of sepsis and multiple organ failure. JAMA. 2011;306(23):2594-605. 103. Hotchkiss RS, Monneret G, Payen D. Sepsis-induced immunosuppression: from cellular dysfunctions to immunotherapy. Nat Rev Immunol. 2013;13(12):862-74. 104. Walton AH, Muenzer JT, Rasche D, Boomer JS, Sato B, Brownstein BH, et al. Reactivation of multiple viruses in patients with sepsis. PLoS One. 2014;9(2):e98819. 105. Wherry EJ, Kurachi M. Molecular and cellular insights into T cell exhaustion. Nat Rev Immunol. 2015;15(8):486-99. 106. Gentile LF, Cuenca AG, Efron PA, Ang D, Bihorac A, McKinley BA, et al. Persistent inflammation and immunosuppression: a common syndrome and new horizon for surgical intensive care. J Trauma Acute Care Surg. 2012;72(6):1491-501. 107. Venet F, Monneret G. Advances in the understanding and treatment of sepsis-induced immunosuppression. Nat Rev Nephrol. 2018;14(2):121-37. 108. Schumer W. Steroids in the treatment of clinical septic shock. Ann Surg. 1976;184(3):333-41. 109. Annane D, Bellissant E, Bollaert PE, Briegel J, Keh D, Kupfer Y. Corticosteroids for treating sepsis. Cochrane Database Syst Rev. 2015(12):Cd002243.

72 ANNA LANGE SLPI and soluble BTLA as immunological markers

110. Rochwerg B, Oczkowski SJ, Siemieniuk RAC, Agoritsas T, Belley- Cote E, D'Aragon F, et al. Corticosteroids in Sepsis: An Updated Systematic Review and Meta-Analysis. Crit Care Med. 2018;46(9):1411-20. 111. Alejandria MM, Lansang MA, Dans LF, Mantaring JB, 3rd. Intravenous immunoglobulin for treating sepsis, severe sepsis and septic shock. Cochrane Database Syst Rev. 2013;9:Cd001090. 112. Madsen MB, Hjortrup PB, Hansen MB, Lange T, Norrby-Teglund A, Hyldegaard O, et al. Immunoglobulin G for patients with necrotising soft tissue infection (INSTINCT): a randomised, blinded, placebo-controlled trial. Intensive Care Med. 2017;43(11):1585-93. 113. Mehta S, McGeer A, Low DE, Hallett D, Bowman DJ, Grossman SL, et al. Morbidity and mortality of patients with invasive group A streptococcal infections admitted to the ICU. Chest. 2006;130(6):1679-86. 114. Basma H, Norrby-Teglund A, Guedez Y, McGeer A, Low DE, El- Ahmedy O, et al. Risk factors in the pathogenesis of invasive group A streptococcal infections: role of protective humoral immunity. Infect Immun. 1999;67(4):1871-7. 115. Darenberg J, Ihendyane N, Sjolin J, Aufwerber E, Haidl S, Follin P, et al. Intravenous immunoglobulin G therapy in streptococcal toxic shock syndrome: a European randomized, double-blind, placebo- controlled trial. Clin Infect Dis. 2003;37(3):333-40. 116. Borthwick EM, Hill CJ, Rabindranath KS, Maxwell AP, McAuley DF, Blackwood B. High-volume haemofiltration for sepsis in adults. Cochrane Database Syst Rev. 2017;1:Cd008075. 117. Hawchar F, Laszlo I, Oveges N, Trasy D, Ondrik Z, Molnar Z. Extracorporeal cytokine adsorption in septic shock: A proof of concept randomized, controlled pilot study. J Crit Care. 2019;49:172-8. 118. Friesecke S, Trager K, Schittek GA, Molnar Z, Bach F, Kogelmann K, et al. International registry on the use of the CytoSorb(R) adsorber in ICU patients : Study protocol and preliminary results. Med Klin Intensivmed Notfmed. 2017. 119. Opal SM, Scannon PJ, Vincent JL, White M, Carroll SF, Palardy JE, et al. Relationship between plasma levels of lipopolysaccharide (LPS) and LPS-binding protein in patients with severe sepsis and septic shock. The Journal of infectious diseases. 1999;180(5):1584- 9. 120. Klein DJ, Derzko A, Foster D, Seely AJE, Brunet F, Romaschin AD, et al. Daily variation in endotoxin levels is associated with increased

ANNA LANGE SLPI and soluble BTLA as immunological markers 73

organ failure in critically ill patients. Shock (Augusta, Ga.). 2007;28(5):524-9. 121. Terayama T, Yamakawa K, Umemura Y, Aihara M, Fujimi S. Polymyxin B Hemoperfusion for Sepsis and Septic Shock: A Systematic Review and Meta-Analysis. Surg Infect (Larchmt). 2017;18(3):225-33. 122. Hamilton JA. Colony-stimulating factors in inflammation and autoimmunity. Nat Rev Immunol. 2008;8(7):533-44. 123. Mathias B, Szpila BE, Moore FA, Efron PA, Moldawer LL. A Review of GM-CSF Therapy in Sepsis. Medicine (Baltimore). 2015;94(50):e2044. 124. Schroder K, Hertzog PJ, Ravasi T, Hume DA. Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol. 2004;75(2):163-89. 125. Docke WD, Randow F, Syrbe U, Krausch D, Asadullah K, Reinke P, et al. Monocyte deactivation in septic patients: restoration by IFN-gamma treatment. Nat Med. 1997;3(6):678-81. 126. Kox WJ, Bone RC, Krausch D, Döcke WD, Kox SN, Wauer H, et al. Interferon gamma-1b in the treatment of compensatory anti- inflammatory response syndrome. A new approach: proof of principle. Arch Intern Med. 1997;157(4):389-93. 127. Delsing CE, Gresnigt MS, Leentjens J, Preijers F, Frager FA, Kox M, et al. Interferon-gamma as adjunctive immunotherapy for invasive fungal infections: a case series. BMC Infect Dis. 2014;14:166. 128. Assendorp EL, Gresnigt MS, Sprenkeler EGG, Meis JF, Dors N, van der Linden JWM, et al. Adjunctive interferon-gamma immunotherapy in a pediatric case of Aspergillus terreus infection. Eur J Clin Microbiol Infect Dis. 2018;37(10):1915-22. 129. Drewry AM, Samra N, Skrupky LP, Fuller BM, Compton SM, Hotchkiss RS. Persistent lymphopenia after diagnosis of sepsis predicts mortality. Shock. 2014;42(5):383-91. 130. Mackall CL, Fry TJ, Gress RE. Harnessing the biology of IL-7 for therapeutic application. Nat Rev Immunol. 2011;11(5):330-42. 131. Francois B, Jeannet R, Daix T, Walton AH, Shotwell MS, Unsinger J, et al. Interleukin-7 restores lymphocytes in septic shock: the IRIS- 7 randomized clinical trial. JCI Insight. 2018;3(5). 132. Wolchok JD, Hodi FS, Weber JS, Allison JP, Urba WJ, Robert C, et al. Development of ipilimumab: a novel immunotherapeutic approach for the treatment of advanced melanoma. Ann N Y Acad Sci. 2013;1291(1):1-13. 133. Gong J, Chehrazi-Raffle A, Reddi S, Salgia R. Development of PD- 1 and PD-L1 inhibitors as a form of cancer immunotherapy: a

74 ANNA LANGE SLPI and soluble BTLA as immunological markers

comprehensive review of registration trials and future considerations. Journal for immunotherapy of cancer. 2018;6(1):8- . 134. Patil NK, Guo Y, Luan L, Sherwood ER. Targeting Immune Cell Checkpoints during Sepsis. Int J Mol Sci. 2017;18(11). 135. Chang K, Svabek C, Vazquez-Guillamet C, Sato B, Rasche D, Wilson S, et al. Targeting the programmed cell death 1: programmed cell death ligand 1 pathway reverses T cell exhaustion in patients with sepsis. Crit Care. 2014;18(1):R3. 136. Brahmamdam P, Inoue S, Unsinger J, Chang KC, McDunn JE, Hotchkiss RS. Delayed administration of anti-PD-1 antibody reverses immune dysfunction and improves survival during sepsis. J Leukoc Biol. 2010;88(2):233-40. 137. Zhang Y, Zhou Y, Lou J, Li J, Bo L, Zhu K, et al. PD-L1 blockade improves survival in experimental sepsis by inhibiting lymphocyte apoptosis and reversing monocyte dysfunction. Crit Care. 2010;14(6):R220. 138. Hotchkiss RS, Colston E, Yende S, Angus DC, Moldawer LL, Crouser ED, et al. Immune Checkpoint Inhibition in Sepsis: A Phase 1b Randomized, Placebo-Controlled, Single Ascending Dose Study of Antiprogrammed Cell Death-Ligand 1 Antibody (BMS-936559). Crit Care Med. 2019;47(5):632-42. 139. Hotchkiss RS, Colston E, Yende S, Crouser ED, Martin GS, Albertson T, et al. Immune checkpoint inhibition in sepsis: a Phase 1b randomized study to evaluate the safety, tolerability, pharmacokinetics, and pharmacodynamics of nivolumab. Intensive Care Med. 2019;45(10):1360-71. 140. Watanabe E, Nishida O, Kakihana Y, Odani M, Okamura T, Harada T, et al. Pharmacokinetics, Pharmacodynamics and Safety of Nivolumab in Patients With Sepsis-induced immunosuppression: A multicenter, open-label phase 1/2 study. Shock. 2019. 141. Inoue S, Bo L, Bian J, Unsinger J, Chang K, Hotchkiss RS. Dose- dependent effect of anti-CTLA-4 on survival in sepsis. Shock. 2011;36(1):38-44. 142. Ohlsson K, Tegner H. Inhibition of elastase from granulocytes by the low molecular weight bronchial protease inhibitor. Scand J Clin Lab Invest. 1976;36(5):437-45. 143. Thompson RC, Ohlsson K. Isolation, properties, and complete amino acid sequence of human secretory leukocyte protease inhibitor, a potent inhibitor of leukocyte elastase. Proc Natl Acad Sci U S A. 1986;83(18):6692-6.

ANNA LANGE SLPI and soluble BTLA as immunological markers 75

144. Abe T, Kobayashi N, Yoshimura K, Trapnell BC, Kim H, Hubbard RC, et al. Expression of the secretory leukoprotease inhibitor gene in epithelial cells. J Clin Invest. 1991;87(6):2207-15. 145. Jin FY, Nathan C, Radzioch D, Ding A. Secretory leukocyte protease inhibitor: a macrophage product induced by and antagonistic to bacterial lipopolysaccharide. Cell. 1997;88(3):417- 26. 146. Samsom JN, van der Marel AP, van Berkel LA, van Helvoort JM, Simons-Oosterhuis Y, Jansen W, et al. Secretory leukoprotease inhibitor in mucosal lymph node dendritic cells regulates the threshold for mucosal tolerance. J Immunol. 2007;179(10):6588- 95. 147. Sallenave JM, Si Tahar M, Cox G, Chignard M, Gauldie J. Secretory leukocyte proteinase inhibitor is a major leukocyte elastase inhibitor in human neutrophils. J Leukoc Biol. 1997;61(6):695-702. 148. Majchrzak-Gorecka M, Majewski P, Grygier B, Murzyn K, Cichy J. Secretory leukocyte protease inhibitor (SLPI), a multifunctional protein in the host defense response. Cytokine Growth Factor Rev. 2016;28:79-93. 149. Abbinante-Nissen JM, Simpson LG, Leikauf GD. Neutrophil elastase increases secretory leukocyte protease inhibitor transcript levels in airway epithelial cells. Am J Physiol. 1993;265(3 Pt 1):L286-92. 150. Grobmyer SR, Barie PS, Nathan CF, Fuortes M, Lin E, Lowry SF, et al. Secretory leukocyte protease inhibitor, an inhibitor of neutrophil activation, is elevated in serum in human sepsis and experimental endotoxemia. Crit Care Med. 2000;28(5):1276-82. 151. Sallenave JM, Shulmann J, Crossley J, Jordana M, Gauldie J. Regulation of secretory leukocyte proteinase inhibitor (SLPI) and elastase-specific inhibitor (ESI/elafin) in human airway epithelial cells by cytokines and neutrophilic enzymes. Am J Respir Cell Mol Biol. 1994;11(6):733-41. 152. Hiemstra PS, Maassen RJ, Stolk J, Heinzel-Wieland R, Steffens GJ, Dijkman JH. Antibacterial activity of antileukoprotease. Infect Immun. 1996;64(11):4520-4. 153. Pillay K, Coutsoudis A, Agadzi-Naqvi AK, Kuhn L, Coovadia HM, Janoff EN. Secretory leukocyte protease inhibitor in vaginal fluids and perinatal human immunodeficiency virus type 1 transmission. J Infect Dis. 2001;183(4):653-6. 154. McNeely TB, Shugars DC, Rosendahl M, Tucker C, Eisenberg SP, Wahl SM. Inhibition of human immunodeficiency virus type 1

76 ANNA LANGE SLPI and soluble BTLA as immunological markers

infectivity by secretory leukocyte protease inhibitor occurs prior to viral reverse transcription. Blood. 1997;90(3):1141-9. 155. Greene CM, McElvaney NG, O'Neill SJ, Taggart CC. Secretory leucoprotease inhibitor impairs Toll-like receptor 2- and 4- mediated responses in monocytic cells. Infect Immun. 2004;72(6):3684-7. 156. Taggart CC, Greene CM, McElvaney NG, O'Neill S. Secretory leucoprotease inhibitor prevents lipopolysaccharide-induced IkappaBalpha degradation without affecting phosphorylation or ubiquitination. J Biol Chem. 2002;277(37):33648-53. 157. Taggart CC, Cryan SA, Weldon S, Gibbons A, Greene CM, Kelly E, et al. Secretory leucoprotease inhibitor binds to NF-kappaB binding sites in monocytes and inhibits p65 binding. J Exp Med. 2005;202(12):1659-68. 158. Nakamura A, Mori Y, Hagiwara K, Suzuki T, Sakakibara T, Kikuchi T, et al. Increased susceptibility to LPS-induced endotoxin shock in secretory leukoprotease inhibitor (SLPI)-deficient mice. J Exp Med. 2003;197(5):669-74. 159. Menckeberg CL, Hol J, Simons-Oosterhuis Y, Raatgeep HR, de Ruiter LF, Lindenbergh-Kortleve DJ, et al. Human buccal epithelium acquires microbial hyporesponsiveness at birth, a role for secretory leukocyte protease inhibitor. Gut. 2015;64(6):884-93. 160. Zabieglo K, Majewski P, Majchrzak-Gorecka M, Wlodarczyk A, Grygier B, Zegar A, et al. The inhibitory effect of secretory leukocyte protease inhibitor (SLPI) on formation of neutrophil extracellular traps. J Leukoc Biol. 2015;98(1):99-106. 161. Klimenkova O, Ellerbeck W, Klimiankou M, Unalan M, Kandabarau S, Gigina A, et al. A lack of secretory leukocyte protease inhibitor (SLPI) causes defects in granulocytic differentiation. Blood. 2014;123(8):1239-49. 162. Parameswaran GI, Sethi S, Murphy TF. Effects of bacterial infection on airway antimicrobial peptides and proteins in COPD. Chest. 2011;140(3):611-7. 163. Twigg MS, Brockbank S, Lowry P, FitzGerald SP, Taggart C, Weldon S. The Role of Serine Proteases and Antiproteases in the Cystic Fibrosis Lung. Mediators Inflamm. 2015;2015:293053. 164. Hannila SS. Secretory Leukocyte Protease Inhibitor (SLPI): Emerging Roles in CNS Trauma and Repair. Neuroscientist. 2015;21(6):630-6. 165. Schneeberger S, Hautz T, Wahl SM, Brandacher G, Sucher R, Steinmassl O, et al. The effect of secretory leukocyte protease inhibitor (SLPI) on ischemia/reperfusion injury in cardiac transplantation. Am J Transplant. 2008;8(4):773-82.

ANNA LANGE SLPI and soluble BTLA as immunological markers 77

166. Prompunt E, Nernpermpisooth N, Sanit J, Kumphune S. Overexpression and pre-treatment of recombinant human Secretory Leukocyte Protease Inhibitor (rhSLPI) reduces an in vitro ischemia/reperfusion injury in rat cardiac myoblast (H9c2) cell. Biomol Concepts. 2018;9(1):17-32. 167. Duits LA, Tjabringa GS, Aarts NJ, van't Wout JW, Hiemstra PS, Nibbering PH, et al. Plasma secretory leukocyte protease inhibitor in febrile patients. Clin Microbiol Infect. 2003;9(7):605-13. 168. Averdunk L, Fitzner C, Levkovich T, Leaf DE, Sobotta M, Vieten J, et al. Secretory Leukocyte Protease Inhibitor (SLPI)-A Novel Predictive Biomarker of Acute Kidney Injury after Cardiac Surgery: A Prospective Observational Study. J Clin Med. 2019;8(11). 169. Antoniades CG, Khamri W, Abeles RD, Taams LS, Triantafyllou E, Possamai LA, et al. Secretory leukocyte protease inhibitor: a pivotal mediator of anti-inflammatory responses in acetaminophen-induced acute liver failure. Hepatology. 2014;59(4):1564-76. 170. Watanabe N, Gavrieli M, Sedy JR, Yang J, Fallarino F, Loftin SK, et al. BTLA is a lymphocyte inhibitory receptor with similarities to CTLA-4 and PD-1. Nat Immunol. 2003;4(7):670-9. 171. Han P, Goularte OD, Rufner K, Wilkinson B, Kaye J. An inhibitory Ig superfamily protein expressed by lymphocytes and APCs is also an early marker of thymocyte positive selection. J Immunol. 2004;172(10):5931-9. 172. Axelsson S, Magnuson A, Lange A, Alshamari A, Hörnquist EH, Hultgren O. A combination of the activation marker CD86 and the immune checkpoint marker B and T lymphocyte attenuator (BTLA) indicates a putative permissive activation state of B cell subtypes in healthy blood donors independent of age and sex. BMC Immunol. 2020;21(1):14. 173. Gonzalez LC, Loyet KM, Calemine-Fenaux J, Chauhan V, Wranik B, Ouyang W, et al. A coreceptor interaction between the CD28 and TNF receptor family members B and T lymphocyte attenuator and herpesvirus entry mediator. Proc Natl Acad Sci U S A. 2005;102(4):1116-21. 174. Cai G, Anumanthan A, Brown JA, Greenfield EA, Zhu B, Freeman GJ. CD160 inhibits activation of human CD4+ T cells through interaction with herpesvirus entry mediator. Nat Immunol. 2008;9(2):176-85. 175. Cai G, Freeman GJ. The CD160, BTLA, LIGHT/HVEM pathway: a bidirectional switch regulating T-cell activation. Immunol Rev. 2009;229(1):244-58.

78 ANNA LANGE SLPI and soluble BTLA as immunological markers

176. Wang Y, Subudhi SK, Anders RA, Lo J, Sun Y, Blink S, et al. The role of herpesvirus entry mediator as a negative regulator of T cell- mediated responses. J Clin Invest. 2005;115(3):711-7. 177. Cheung TC, Steinberg MW, Oborne LM, Macauley MG, Fukuyama S, Sanjo H, et al. Unconventional ligand activation of herpesvirus entry mediator signals cell survival. Proc Natl Acad Sci U S A. 2009;106(15):6244-9. 178. Ware CF, Sedý JR. TNF Superfamily Networks: bidirectional and interference pathways of the herpesvirus entry mediator (TNFSF14). Curr Opin Immunol. 2011;23(5):627-31. 179. Oster C, Wilde B, Specker C, Sun M, Kribben A, Witzke O, et al. BTLA Expression on Th1, Th2 and Th17 Effector T-Cells of Patients with Systemic Lupus Erythematosus Is Associated with Active Disease. International journal of molecular sciences. 2019;20(18):4505. 180. Chen YL, Lin HW, Chien CL, Lai YL, Sun WZ, Chen CA, et al. BTLA blockade enhances Cancer therapy by inhibiting IL-6/IL-10- induced CD19(high) B lymphocytes. J Immunother Cancer. 2019;7(1):313. 181. Shubin NJ, Monaghan SF, Heffernan DS, Chung CS, Ayala A. B and T lymphocyte attenuator expression on CD4+ T-cells associates with sepsis and subsequent infections in ICU patients. Crit Care. 2013;17(6):R276. 182. Shao R, Li CS, Fang Y, Zhao L, Hang C. Low B and T lymphocyte attenuator expression on CD4(+) T cells in the early stage of sepsis is associated with the severity and mortality of septic patients: a prospective cohort study. Crit Care. 2015;19(1):308. 183. Shubin NJ, Chung CS, Heffernan DS, Irwin LR, Monaghan SF, Ayala A. BTLA expression contributes to septic morbidity and mortality by inducing innate inflammatory cell dysfunction. J Leukoc Biol. 2012;92(3):593-603. 184. Cheng T, Bai J, Chung CS, Chen Y, Biron BM, Ayala A. Enhanced Innate Inflammation Induced by Anti-BTLA Antibody in Dual Insult Model of Hemorrhagic Shock/Sepsis. Shock. 2016;45(1):40- 9. 185. Monaghan SF, Banerjee D, Chung CS, Lomas-Neira J, Cygan KJ, Rhine CL, et al. Changes in the process of alternative RNA splicing results in soluble B and T lymphocyte attenuator with biological and clinical implications in critical illness. Mol Med. 2018;24(1):32. 186. Strålin K, Kaltoft MS, Konradsen HB, Olcén P, Holmberg H. Comparison of two urinary antigen tests for establishment of

ANNA LANGE SLPI and soluble BTLA as immunological markers 79

pneumococcal etiology of adult community-acquired pneumonia. J Clin Microbiol. 2004;42(8):3620-5. 187. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med. 1997;336(4):243- 50. 188. Loke YK, Kwok CS, Niruban A, Myint PK. Value of severity scales in predicting mortality from community-acquired pneumonia: systematic review and meta-analysis. Thorax. 2010;65(10):884-90. 189. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med. 1985;13(10):818-29. 190. Vincent JL, de Mendonça A, Cantraine F, Moreno R, Takala J, Suter PM, et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Working group on "sepsis-related problems" of the European Society of Intensive Care Medicine. Crit Care Med. 1998;26(11):1793-800. 191. Rubin DB. Inference and missing data. Biometrika. 1976;63(3):581-92. 192. Dekker FW, de Mutsert R, van Dijk PC, Zoccali C, Jager KJ. Survival analysis: time-dependent effects and time-varying risk factors. Kidney Int. 2008;74(8):994-7. 193. Kida K, Mizuuchi T, Takeyama K, Hiratsuka T, Jinno S, Hosoda K, et al. Serum secretory leukoprotease inhibitor levels to diagnose pneumonia in the elderly. Am Rev Respir Dis. 1992;146(6):1426- 9. 194. Leijte GP, Rimmelé T, Kox M, Bruse N, Monard C, Gossez M, et al. Monocytic HLA-DR expression kinetics in septic shock patients with different pathogens, sites of infection and adverse outcomes. Crit Care. 2020;24(1):110. 195. Berkestedt I, Herwald H, Ljunggren L, Nelson A, Bodelsson M. Elevated plasma levels of antimicrobial polypeptides in patients with severe sepsis. J Innate Immun. 2010;2(5):478-82. 196. Rintala E, Peuravuori H, Pulkki K, Voipio-Pulkki LM, Nevalainen T. Bactericidal/permeability-increasing protein (BPI) in sepsis correlates with the severity of sepsis and the outcome. Intensive Care Med. 2000;26(9):1248-51. 197. Zuo Y, Dang R, Peng H, Wu Z, Yang Y. LL-37 Exacerbates Local Inflammation in Sepsis-Induced Acute Lung Injury by Preventing Mitochondrial DNA (mtDNA) Degradation-Induced Autophagy. Med Sci Monit. 2019;25:6193-9203.

80 ANNA LANGE SLPI and soluble BTLA as immunological markers

198. Majewski K, elechowska P, Brzezi ska-Błaszczyk E. Circulating cathelicidin LL-37 in adult patients with pulmonary infectious diseases. Clin ŻInvest Med. 2017;40(1):E34ń -e9. 199. Shindo Y, Ito R, Kobayashi D, Ando M, Ichikawa M, Goto Y, et al. Risk factors for 30-day mortality in patients with pneumonia who receive appropriate initial antibiotics: an observational cohort study. Lancet Infect Dis. 2015;15(9):1055-65. 200. Ramachandran G. Gram-positive and gram-negative bacterial toxins in sepsis: a brief review. Virulence. 2014;5(1):213-8. 201. Toscano MG, Ganea D, Gamero AM. Cecal ligation puncture procedure. J Vis Exp. 2011(51). 202. Zhao Y, Jia Y, Li C, Shao R, Fang Y. Predictive Value of Soluble Programmed Death-1 for Severe Sepsis and Septic Shock During the First Week in an Intensive Care Unit. Shock (Augusta, Ga.). 2019;51(3):289-97. 203. Simone R, Pesce G, Antola P, Rumbullaku M, Bagnasco M, Bizzaro N, et al. The soluble form of CTLA-4 from serum of patients with autoimmune diseases regulates T-cell responses. Biomed Res Int. 2014;2014:215763. 204. Pistillo MP, Fontana V, Morabito A, Dozin B, Laurent S, Carosio R, et al. Soluble CTLA-4 as a favorable predictive biomarker in metastatic melanoma patients treated with ipilimumab: an Italian melanoma intergroup study. Cancer Immunol Immunother. 2019;68(1):97-107. 205. Westerdijk K, Simons KS, Zegers M, Wever PC, Pickkers P, de Jager CPC. The value of the neutrophil-lymphocyte count ratio in the diagnosis of sepsis in patients admitted to the Intensive Care Unit: A retrospective cohort study. PLoS One. 2019;14(2):e0212861. 206. Wang Q, Zhang J, Tu H, Liang D, Chang DW, Ye Y, et al. Soluble immune checkpoint-related proteins as predictors of tumor recurrence, survival, and T cell phenotypes in clear cell renal cell carcinoma patients. Journal for immunotherapy of cancer. 2019;7(1):334-. 207. Bian B, Fanale D, Dusetti N, Roque J, Pastor S, Chretien A-S, et al. Prognostic significance of circulating PD-1, PD-L1, pan-BTN3As, BTN3A1 and BTLA in patients with pancreatic adenocarcinoma. Oncoimmunology. 2019;8(4):e1561120-e. 208. Dong MP, Enomoto M, Thuy LTT, Hai H, Hieu VN, Hoang DV, et al. Clinical significance of circulating soluble immune checkpoint proteins in sorafenib-treated patients with advanced hepatocellular carcinoma. Sci Rep. 2020;10(1):3392-.

ANNA LANGE SLPI and soluble BTLA as immunological markers 81

209. Prescott HC, Angus DC. Enhancing Recovery From Sepsis: A Review. JAMA. 2018;319(1):62-75. 210. Yende S, Kellum JA, Talisa VB, Peck Palmer OM, Chang CH, Filbin MR, et al. Long-term Host Immune Response Trajectories Among Hospitalized Patients With Sepsis. JAMA Netw Open. 2019;2(8):e198686. 211. Riche F, Chousterman BG, Valleur P, Mebazaa A, Launay JM, Gayat E. Protracted immune disorders at one year after ICU discharge in patients with septic shock. Crit Care. 2018;22(1):42. 212. Hoffmann M, Kleine-Weber H, Schroeder S, Krüger N, Herrler T, Erichsen S, et al. SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor. Cell. 2020. 213. Kesic MJ, Meyer M, Bauer R, Jaspers I. Exposure to ozone modulates human airway protease/antiprotease balance contributing to increased influenza A infection. PLoS One. 2012;7(4):e35108.

82 ANNA LANGE SLPI and soluble BTLA as immunological markers